case study on online learning problems

• Published: 28 May 2021

Students’ online learning challenges during the pandemic and how they cope with them: The case of the Philippines

• Jessie S. Barrot   ORCID: orcid.org/0000-0001-8517-4058 1 ,
• Ian I. Llenares 1 &
• Leo S. del Rosario 1  

Education and Information Technologies volume  26 ,  pages 7321–7338 ( 2021 ) Cite this article

887k Accesses

174 Citations

4 Altmetric

Metrics details

Recently, the education system has faced an unprecedented health crisis that has shaken up its foundation. Given today’s uncertainties, it is vital to gain a nuanced understanding of students’ online learning experience in times of the COVID-19 pandemic. Although many studies have investigated this area, limited information is available regarding the challenges and the specific strategies that students employ to overcome them. Thus, this study attempts to fill in the void. Using a mixed-methods approach, the findings revealed that the online learning challenges of college students varied in terms of type and extent. Their greatest challenge was linked to their learning environment at home, while their least challenge was technological literacy and competency. The findings further revealed that the COVID-19 pandemic had the greatest impact on the quality of the learning experience and students’ mental health. In terms of strategies employed by students, the most frequently used were resource management and utilization, help-seeking, technical aptitude enhancement, time management, and learning environment control. Implications for classroom practice, policy-making, and future research are discussed.

Working on a manuscript?

1 introduction.

Since the 1990s, the world has seen significant changes in the landscape of education as a result of the ever-expanding influence of technology. One such development is the adoption of online learning across different learning contexts, whether formal or informal, academic and non-academic, and residential or remotely. We began to witness schools, teachers, and students increasingly adopt e-learning technologies that allow teachers to deliver instruction interactively, share resources seamlessly, and facilitate student collaboration and interaction (Elaish et al., 2019 ; Garcia et al., 2018 ). Although the efficacy of online learning has long been acknowledged by the education community (Barrot, 2020 , 2021 ; Cavanaugh et al., 2009 ; Kebritchi et al., 2017 ; Tallent-Runnels et al., 2006 ; Wallace, 2003 ), evidence on the challenges in its implementation continues to build up (e.g., Boelens et al., 2017 ; Rasheed et al., 2020 ).

Recently, the education system has faced an unprecedented health crisis (i.e., COVID-19 pandemic) that has shaken up its foundation. Thus, various governments across the globe have launched a crisis response to mitigate the adverse impact of the pandemic on education. This response includes, but is not limited to, curriculum revisions, provision for technological resources and infrastructure, shifts in the academic calendar, and policies on instructional delivery and assessment. Inevitably, these developments compelled educational institutions to migrate to full online learning until face-to-face instruction is allowed. The current circumstance is unique as it could aggravate the challenges experienced during online learning due to restrictions in movement and health protocols (Gonzales et al., 2020 ; Kapasia et al., 2020 ). Given today’s uncertainties, it is vital to gain a nuanced understanding of students’ online learning experience in times of the COVID-19 pandemic. To date, many studies have investigated this area with a focus on students’ mental health (Copeland et al., 2021 ; Fawaz et al., 2021 ), home learning (Suryaman et al., 2020 ), self-regulation (Carter et al., 2020 ), virtual learning environment (Almaiah et al., 2020 ; Hew et al., 2020 ; Tang et al., 2020 ), and students’ overall learning experience (e.g., Adarkwah, 2021 ; Day et al., 2021 ; Khalil et al., 2020 ; Singh et al., 2020 ). There are two key differences that set the current study apart from the previous studies. First, it sheds light on the direct impact of the pandemic on the challenges that students experience in an online learning space. Second, the current study explores students’ coping strategies in this new learning setup. Addressing these areas would shed light on the extent of challenges that students experience in a full online learning space, particularly within the context of the pandemic. Meanwhile, our nuanced understanding of the strategies that students use to overcome their challenges would provide relevant information to school administrators and teachers to better support the online learning needs of students. This information would also be critical in revisiting the typology of strategies in an online learning environment.

2 Literature review

2.1 education and the covid-19 pandemic.

In December 2019, an outbreak of a novel coronavirus, known as COVID-19, occurred in China and has spread rapidly across the globe within a few months. COVID-19 is an infectious disease caused by a new strain of coronavirus that attacks the respiratory system (World Health Organization, 2020 ). As of January 2021, COVID-19 has infected 94 million people and has caused 2 million deaths in 191 countries and territories (John Hopkins University, 2021 ). This pandemic has created a massive disruption of the educational systems, affecting over 1.5 billion students. It has forced the government to cancel national examinations and the schools to temporarily close, cease face-to-face instruction, and strictly observe physical distancing. These events have sparked the digital transformation of higher education and challenged its ability to respond promptly and effectively. Schools adopted relevant technologies, prepared learning and staff resources, set systems and infrastructure, established new teaching protocols, and adjusted their curricula. However, the transition was smooth for some schools but rough for others, particularly those from developing countries with limited infrastructure (Pham & Nguyen, 2020 ; Simbulan, 2020 ).

Inevitably, schools and other learning spaces were forced to migrate to full online learning as the world continues the battle to control the vicious spread of the virus. Online learning refers to a learning environment that uses the Internet and other technological devices and tools for synchronous and asynchronous instructional delivery and management of academic programs (Usher & Barak, 2020 ; Huang, 2019 ). Synchronous online learning involves real-time interactions between the teacher and the students, while asynchronous online learning occurs without a strict schedule for different students (Singh & Thurman, 2019 ). Within the context of the COVID-19 pandemic, online learning has taken the status of interim remote teaching that serves as a response to an exigency. However, the migration to a new learning space has faced several major concerns relating to policy, pedagogy, logistics, socioeconomic factors, technology, and psychosocial factors (Donitsa-Schmidt & Ramot, 2020 ; Khalil et al., 2020 ; Varea & González-Calvo, 2020 ). With reference to policies, government education agencies and schools scrambled to create fool-proof policies on governance structure, teacher management, and student management. Teachers, who were used to conventional teaching delivery, were also obliged to embrace technology despite their lack of technological literacy. To address this problem, online learning webinars and peer support systems were launched. On the part of the students, dropout rates increased due to economic, psychological, and academic reasons. Academically, although it is virtually possible for students to learn anything online, learning may perhaps be less than optimal, especially in courses that require face-to-face contact and direct interactions (Franchi, 2020 ).

2.2 Related studies

Recently, there has been an explosion of studies relating to the new normal in education. While many focused on national policies, professional development, and curriculum, others zeroed in on the specific learning experience of students during the pandemic. Among these are Copeland et al. ( 2021 ) and Fawaz et al. ( 2021 ) who examined the impact of COVID-19 on college students’ mental health and their coping mechanisms. Copeland et al. ( 2021 ) reported that the pandemic adversely affected students’ behavioral and emotional functioning, particularly attention and externalizing problems (i.e., mood and wellness behavior), which were caused by isolation, economic/health effects, and uncertainties. In Fawaz et al.’s ( 2021 ) study, students raised their concerns on learning and evaluation methods, overwhelming task load, technical difficulties, and confinement. To cope with these problems, students actively dealt with the situation by seeking help from their teachers and relatives and engaging in recreational activities. These active-oriented coping mechanisms of students were aligned with Carter et al.’s ( 2020 ), who explored students’ self-regulation strategies.

In another study, Tang et al. ( 2020 ) examined the efficacy of different online teaching modes among engineering students. Using a questionnaire, the results revealed that students were dissatisfied with online learning in general, particularly in the aspect of communication and question-and-answer modes. Nonetheless, the combined model of online teaching with flipped classrooms improved students’ attention, academic performance, and course evaluation. A parallel study was undertaken by Hew et al. ( 2020 ), who transformed conventional flipped classrooms into fully online flipped classes through a cloud-based video conferencing app. Their findings suggested that these two types of learning environments were equally effective. They also offered ways on how to effectively adopt videoconferencing-assisted online flipped classrooms. Unlike the two studies, Suryaman et al. ( 2020 ) looked into how learning occurred at home during the pandemic. Their findings showed that students faced many obstacles in a home learning environment, such as lack of mastery of technology, high Internet cost, and limited interaction/socialization between and among students. In a related study, Kapasia et al. ( 2020 ) investigated how lockdown impacts students’ learning performance. Their findings revealed that the lockdown made significant disruptions in students’ learning experience. The students also reported some challenges that they faced during their online classes. These include anxiety, depression, poor Internet service, and unfavorable home learning environment, which were aggravated when students are marginalized and from remote areas. Contrary to Kapasia et al.’s ( 2020 ) findings, Gonzales et al. ( 2020 ) found that confinement of students during the pandemic had significant positive effects on their performance. They attributed these results to students’ continuous use of learning strategies which, in turn, improved their learning efficiency.

Finally, there are those that focused on students’ overall online learning experience during the COVID-19 pandemic. One such study was that of Singh et al. ( 2020 ), who examined students’ experience during the COVID-19 pandemic using a quantitative descriptive approach. Their findings indicated that students appreciated the use of online learning during the pandemic. However, half of them believed that the traditional classroom setting was more effective than the online learning platform. Methodologically, the researchers acknowledge that the quantitative nature of their study restricts a deeper interpretation of the findings. Unlike the above study, Khalil et al. ( 2020 ) qualitatively explored the efficacy of synchronized online learning in a medical school in Saudi Arabia. The results indicated that students generally perceive synchronous online learning positively, particularly in terms of time management and efficacy. However, they also reported technical (internet connectivity and poor utility of tools), methodological (content delivery), and behavioral (individual personality) challenges. Their findings also highlighted the failure of the online learning environment to address the needs of courses that require hands-on practice despite efforts to adopt virtual laboratories. In a parallel study, Adarkwah ( 2021 ) examined students’ online learning experience during the pandemic using a narrative inquiry approach. The findings indicated that Ghanaian students considered online learning as ineffective due to several challenges that they encountered. Among these were lack of social interaction among students, poor communication, lack of ICT resources, and poor learning outcomes. More recently, Day et al. ( 2021 ) examined the immediate impact of COVID-19 on students’ learning experience. Evidence from six institutions across three countries revealed some positive experiences and pre-existing inequities. Among the reported challenges are lack of appropriate devices, poor learning space at home, stress among students, and lack of fieldwork and access to laboratories.

Although there are few studies that report the online learning challenges that higher education students experience during the pandemic, limited information is available regarding the specific strategies that they use to overcome them. It is in this context that the current study was undertaken. This mixed-methods study investigates students’ online learning experience in higher education. Specifically, the following research questions are addressed: (1) What is the extent of challenges that students experience in an online learning environment? (2) How did the COVID-19 pandemic impact the online learning challenges that students experience? (3) What strategies did students use to overcome the challenges?

2.3 Conceptual framework

The typology of challenges examined in this study is largely based on Rasheed et al.’s ( 2020 ) review of students’ experience in an online learning environment. These challenges are grouped into five general clusters, namely self-regulation (SRC), technological literacy and competency (TLCC), student isolation (SIC), technological sufficiency (TSC), and technological complexity (TCC) challenges (Rasheed et al., 2020 , p. 5). SRC refers to a set of behavior by which students exercise control over their emotions, actions, and thoughts to achieve learning objectives. TLCC relates to a set of challenges about students’ ability to effectively use technology for learning purposes. SIC relates to the emotional discomfort that students experience as a result of being lonely and secluded from their peers. TSC refers to a set of challenges that students experience when accessing available online technologies for learning. Finally, there is TCC which involves challenges that students experience when exposed to complex and over-sufficient technologies for online learning.

To extend Rasheed et al. ( 2020 ) categories and to cover other potential challenges during online classes, two more clusters were added, namely learning resource challenges (LRC) and learning environment challenges (LEC) (Buehler, 2004 ; Recker et al., 2004 ; Seplaki et al., 2014 ; Xue et al., 2020 ). LRC refers to a set of challenges that students face relating to their use of library resources and instructional materials, whereas LEC is a set of challenges that students experience related to the condition of their learning space that shapes their learning experiences, beliefs, and attitudes. Since learning environment at home and learning resources available to students has been reported to significantly impact the quality of learning and their achievement of learning outcomes (Drane et al., 2020 ; Suryaman et al., 2020 ), the inclusion of LRC and LEC would allow us to capture other important challenges that students experience during the pandemic, particularly those from developing regions. This comprehensive list would provide us a clearer and detailed picture of students’ experiences when engaged in online learning in an emergency. Given the restrictions in mobility at macro and micro levels during the pandemic, it is also expected that such conditions would aggravate these challenges. Therefore, this paper intends to understand these challenges from students’ perspectives since they are the ones that are ultimately impacted when the issue is about the learning experience. We also seek to explore areas that provide inconclusive findings, thereby setting the path for future research.

3 Material and methods

The present study adopted a descriptive, mixed-methods approach to address the research questions. This approach allowed the researchers to collect complex data about students’ experience in an online learning environment and to clearly understand the phenomena from their perspective.

3.1 Participants

This study involved 200 (66 male and 134 female) students from a private higher education institution in the Philippines. These participants were Psychology, Physical Education, and Sports Management majors whose ages ranged from 17 to 25 ( x̅  = 19.81; SD  = 1.80). The students have been engaged in online learning for at least two terms in both synchronous and asynchronous modes. The students belonged to low- and middle-income groups but were equipped with the basic online learning equipment (e.g., computer, headset, speakers) and computer skills necessary for their participation in online classes. Table 1 shows the primary and secondary platforms that students used during their online classes. The primary platforms are those that are formally adopted by teachers and students in a structured academic context, whereas the secondary platforms are those that are informally and spontaneously used by students and teachers for informal learning and to supplement instructional delivery. Note that almost all students identified MS Teams as their primary platform because it is the official learning management system of the university.

Informed consent was sought from the participants prior to their involvement. Before students signed the informed consent form, they were oriented about the objectives of the study and the extent of their involvement. They were also briefed about the confidentiality of information, their anonymity, and their right to refuse to participate in the investigation. Finally, the participants were informed that they would incur no additional cost from their participation.

3.2 Instrument and data collection

The data were collected using a retrospective self-report questionnaire and a focused group discussion (FGD). A self-report questionnaire was considered appropriate because the indicators relate to affective responses and attitude (Araujo et al., 2017 ; Barrot, 2016 ; Spector, 1994 ). Although the participants may tell more than what they know or do in a self-report survey (Matsumoto, 1994 ), this challenge was addressed by explaining to them in detail each of the indicators and using methodological triangulation through FGD. The questionnaire was divided into four sections: (1) participant’s personal information section, (2) the background information on the online learning environment, (3) the rating scale section for the online learning challenges, (4) the open-ended section. The personal information section asked about the students’ personal information (name, school, course, age, and sex), while the background information section explored the online learning mode and platforms (primary and secondary) used in class, and students’ length of engagement in online classes. The rating scale section contained 37 items that relate to SRC (6 items), TLCC (10 items), SIC (4 items), TSC (6 items), TCC (3 items), LRC (4 items), and LEC (4 items). The Likert scale uses six scores (i.e., 5– to a very great extent , 4– to a great extent , 3– to a moderate extent , 2– to some extent , 1– to a small extent , and 0 –not at all/negligible ) assigned to each of the 37 items. Finally, the open-ended questions asked about other challenges that students experienced, the impact of the pandemic on the intensity or extent of the challenges they experienced, and the strategies that the participants employed to overcome the eight different types of challenges during online learning. Two experienced educators and researchers reviewed the questionnaire for clarity, accuracy, and content and face validity. The piloting of the instrument revealed that the tool had good internal consistency (Cronbach’s α = 0.96).

The FGD protocol contains two major sections: the participants’ background information and the main questions. The background information section asked about the students’ names, age, courses being taken, online learning mode used in class. The items in the main questions section covered questions relating to the students’ overall attitude toward online learning during the pandemic, the reasons for the scores they assigned to each of the challenges they experienced, the impact of the pandemic on students’ challenges, and the strategies they employed to address the challenges. The same experts identified above validated the FGD protocol.

Both the questionnaire and the FGD were conducted online via Google survey and MS Teams, respectively. It took approximately 20 min to complete the questionnaire, while the FGD lasted for about 90 min. Students were allowed to ask for clarification and additional explanations relating to the questionnaire content, FGD, and procedure. Online surveys and interview were used because of the ongoing lockdown in the city. For the purpose of triangulation, 20 (10 from Psychology and 10 from Physical Education and Sports Management) randomly selected students were invited to participate in the FGD. Two separate FGDs were scheduled for each group and were facilitated by researcher 2 and researcher 3, respectively. The interviewers ensured that the participants were comfortable and open to talk freely during the FGD to avoid social desirability biases (Bergen & Labonté, 2020 ). These were done by informing the participants that there are no wrong responses and that their identity and responses would be handled with the utmost confidentiality. With the permission of the participants, the FGD was recorded to ensure that all relevant information was accurately captured for transcription and analysis.

3.3 Data analysis

To address the research questions, we used both quantitative and qualitative analyses. For the quantitative analysis, we entered all the data into an excel spreadsheet. Then, we computed the mean scores ( M ) and standard deviations ( SD ) to determine the level of challenges experienced by students during online learning. The mean score for each descriptor was interpreted using the following scheme: 4.18 to 5.00 ( to a very great extent ), 3.34 to 4.17 ( to a great extent ), 2.51 to 3.33 ( to a moderate extent ), 1.68 to 2.50 ( to some extent ), 0.84 to 1.67 ( to a small extent ), and 0 to 0.83 ( not at all/negligible ). The equal interval was adopted because it produces more reliable and valid information than other types of scales (Cicchetti et al., 2006 ).

For the qualitative data, we analyzed the students’ responses in the open-ended questions and the transcribed FGD using the predetermined categories in the conceptual framework. Specifically, we used multilevel coding in classifying the codes from the transcripts (Birks & Mills, 2011 ). To do this, we identified the relevant codes from the responses of the participants and categorized these codes based on the similarities or relatedness of their properties and dimensions. Then, we performed a constant comparative and progressive analysis of cases to allow the initially identified subcategories to emerge and take shape. To ensure the reliability of the analysis, two coders independently analyzed the qualitative data. Both coders familiarize themselves with the purpose, research questions, research method, and codes and coding scheme of the study. They also had a calibration session and discussed ways on how they could consistently analyze the qualitative data. Percent of agreement between the two coders was 86 percent. Any disagreements in the analysis were discussed by the coders until an agreement was achieved.

This study investigated students’ online learning experience in higher education within the context of the pandemic. Specifically, we identified the extent of challenges that students experienced, how the COVID-19 pandemic impacted their online learning experience, and the strategies that they used to confront these challenges.

4.1 The extent of students’ online learning challenges

Table 2 presents the mean scores and SD for the extent of challenges that students’ experienced during online learning. Overall, the students experienced the identified challenges to a moderate extent ( x̅  = 2.62, SD  = 1.03) with scores ranging from x̅  = 1.72 ( to some extent ) to x̅  = 3.58 ( to a great extent ). More specifically, the greatest challenge that students experienced was related to the learning environment ( x̅  = 3.49, SD  = 1.27), particularly on distractions at home, limitations in completing the requirements for certain subjects, and difficulties in selecting the learning areas and study schedule. It is, however, found that the least challenge was on technological literacy and competency ( x̅  = 2.10, SD  = 1.13), particularly on knowledge and training in the use of technology, technological intimidation, and resistance to learning technologies. Other areas that students experienced the least challenge are Internet access under TSC and procrastination under SRC. Nonetheless, nearly half of the students’ responses per indicator rated the challenges they experienced as moderate (14 of the 37 indicators), particularly in TCC ( x̅  = 2.51, SD  = 1.31), SIC ( x̅  = 2.77, SD  = 1.34), and LRC ( x̅  = 2.93, SD  = 1.31).

Out of 200 students, 181 responded to the question about other challenges that they experienced. Most of their responses were already covered by the seven predetermined categories, except for 18 responses related to physical discomfort ( N  = 5) and financial challenges ( N  = 13). For instance, S108 commented that “when it comes to eyes and head, my eyes and head get ache if the session of class was 3 h straight in front of my gadget.” In the same vein, S194 reported that “the long exposure to gadgets especially laptop, resulting in body pain & headaches.” With reference to physical financial challenges, S66 noted that “not all the time I have money to load”, while S121 claimed that “I don't know until when are we going to afford budgeting our money instead of buying essentials.”

4.2 Impact of the pandemic on students’ online learning challenges

Another objective of this study was to identify how COVID-19 influenced the online learning challenges that students experienced. As shown in Table 3 , most of the students’ responses were related to teaching and learning quality ( N  = 86) and anxiety and other mental health issues ( N  = 52). Regarding the adverse impact on teaching and learning quality, most of the comments relate to the lack of preparation for the transition to online platforms (e.g., S23, S64), limited infrastructure (e.g., S13, S65, S99, S117), and poor Internet service (e.g., S3, S9, S17, S41, S65, S99). For the anxiety and mental health issues, most students reported that the anxiety, boredom, sadness, and isolation they experienced had adversely impacted the way they learn (e.g., S11, S130), completing their tasks/activities (e.g., S56, S156), and their motivation to continue studying (e.g., S122, S192). The data also reveal that COVID-19 aggravated the financial difficulties experienced by some students ( N  = 16), consequently affecting their online learning experience. This financial impact mainly revolved around the lack of funding for their online classes as a result of their parents’ unemployment and the high cost of Internet data (e.g., S18, S113, S167). Meanwhile, few concerns were raised in relation to COVID-19’s impact on mobility ( N  = 7) and face-to-face interactions ( N  = 7). For instance, some commented that the lack of face-to-face interaction with her classmates had a detrimental effect on her learning (S46) and socialization skills (S36), while others reported that restrictions in mobility limited their learning experience (S78, S110). Very few comments were related to no effect ( N  = 4) and positive effect ( N  = 2). The above findings suggest the pandemic had additive adverse effects on students’ online learning experience.

4.3 Students’ strategies to overcome challenges in an online learning environment

The third objective of this study is to identify the strategies that students employed to overcome the different online learning challenges they experienced. Table 4 presents that the most commonly used strategies used by students were resource management and utilization ( N  = 181), help-seeking ( N  = 155), technical aptitude enhancement ( N  = 122), time management ( N  = 98), and learning environment control ( N  = 73). Not surprisingly, the top two strategies were also the most consistently used across different challenges. However, looking closely at each of the seven challenges, the frequency of using a particular strategy varies. For TSC and LRC, the most frequently used strategy was resource management and utilization ( N  = 52, N  = 89, respectively), whereas technical aptitude enhancement was the students’ most preferred strategy to address TLCC ( N  = 77) and TCC ( N  = 38). In the case of SRC, SIC, and LEC, the most frequently employed strategies were time management ( N  = 71), psychological support ( N  = 53), and learning environment control ( N  = 60). In terms of consistency, help-seeking appears to be the most consistent across the different challenges in an online learning environment. Table 4 further reveals that strategies used by students within a specific type of challenge vary.

5 Discussion and conclusions

The current study explores the challenges that students experienced in an online learning environment and how the pandemic impacted their online learning experience. The findings revealed that the online learning challenges of students varied in terms of type and extent. Their greatest challenge was linked to their learning environment at home, while their least challenge was technological literacy and competency. Based on the students’ responses, their challenges were also found to be aggravated by the pandemic, especially in terms of quality of learning experience, mental health, finances, interaction, and mobility. With reference to previous studies (i.e., Adarkwah, 2021 ; Copeland et al., 2021 ; Day et al., 2021 ; Fawaz et al., 2021 ; Kapasia et al., 2020 ; Khalil et al., 2020 ; Singh et al., 2020 ), the current study has complemented their findings on the pedagogical, logistical, socioeconomic, technological, and psychosocial online learning challenges that students experience within the context of the COVID-19 pandemic. Further, this study extended previous studies and our understanding of students’ online learning experience by identifying both the presence and extent of online learning challenges and by shedding light on the specific strategies they employed to overcome them.

Overall findings indicate that the extent of challenges and strategies varied from one student to another. Hence, they should be viewed as a consequence of interaction several many factors. Students’ responses suggest that their online learning challenges and strategies were mediated by the resources available to them, their interaction with their teachers and peers, and the school’s existing policies and guidelines for online learning. In the context of the pandemic, the imposed lockdowns and students’ socioeconomic condition aggravated the challenges that students experience.

While most studies revealed that technology use and competency were the most common challenges that students face during the online classes (see Rasheed et al., 2020 ), the case is a bit different in developing countries in times of pandemic. As the findings have shown, the learning environment is the greatest challenge that students needed to hurdle, particularly distractions at home (e.g., noise) and limitations in learning space and facilities. This data suggests that online learning challenges during the pandemic somehow vary from the typical challenges that students experience in a pre-pandemic online learning environment. One possible explanation for this result is that restriction in mobility may have aggravated this challenge since they could not go to the school or other learning spaces beyond the vicinity of their respective houses. As shown in the data, the imposition of lockdown restricted students’ learning experience (e.g., internship and laboratory experiments), limited their interaction with peers and teachers, caused depression, stress, and anxiety among students, and depleted the financial resources of those who belong to lower-income group. All of these adversely impacted students’ learning experience. This finding complemented earlier reports on the adverse impact of lockdown on students’ learning experience and the challenges posed by the home learning environment (e.g., Day et al., 2021 ; Kapasia et al., 2020 ). Nonetheless, further studies are required to validate the impact of restrictions on mobility on students’ online learning experience. The second reason that may explain the findings relates to students’ socioeconomic profile. Consistent with the findings of Adarkwah ( 2021 ) and Day et al. ( 2021 ), the current study reveals that the pandemic somehow exposed the many inequities in the educational systems within and across countries. In the case of a developing country, families from lower socioeconomic strata (as in the case of the students in this study) have limited learning space at home, access to quality Internet service, and online learning resources. This is the reason the learning environment and learning resources recorded the highest level of challenges. The socioeconomic profile of the students (i.e., low and middle-income group) is the same reason financial problems frequently surfaced from their responses. These students frequently linked the lack of financial resources to their access to the Internet, educational materials, and equipment necessary for online learning. Therefore, caution should be made when interpreting and extending the findings of this study to other contexts, particularly those from higher socioeconomic strata.

Among all the different online learning challenges, the students experienced the least challenge on technological literacy and competency. This is not surprising considering a plethora of research confirming Gen Z students’ (born since 1996) high technological and digital literacy (Barrot, 2018 ; Ng, 2012 ; Roblek et al., 2019 ). Regarding the impact of COVID-19 on students’ online learning experience, the findings reveal that teaching and learning quality and students’ mental health were the most affected. The anxiety that students experienced does not only come from the threats of COVID-19 itself but also from social and physical restrictions, unfamiliarity with new learning platforms, technical issues, and concerns about financial resources. These findings are consistent with that of Copeland et al. ( 2021 ) and Fawaz et al. ( 2021 ), who reported the adverse effects of the pandemic on students’ mental and emotional well-being. This data highlights the need to provide serious attention to the mediating effects of mental health, restrictions in mobility, and preparedness in delivering online learning.

Nonetheless, students employed a variety of strategies to overcome the challenges they faced during online learning. For instance, to address the home learning environment problems, students talked to their family (e.g., S12, S24), transferred to a quieter place (e.g., S7, S 26), studied at late night where all family members are sleeping already (e.g., S51), and consulted with their classmates and teachers (e.g., S3, S9, S156, S193). To overcome the challenges in learning resources, students used the Internet (e.g., S20, S27, S54, S91), joined Facebook groups that share free resources (e.g., S5), asked help from family members (e.g., S16), used resources available at home (e.g., S32), and consulted with the teachers (e.g., S124). The varying strategies of students confirmed earlier reports on the active orientation that students take when faced with academic- and non-academic-related issues in an online learning space (see Fawaz et al., 2021 ). The specific strategies that each student adopted may have been shaped by different factors surrounding him/her, such as available resources, student personality, family structure, relationship with peers and teacher, and aptitude. To expand this study, researchers may further investigate this area and explore how and why different factors shape their use of certain strategies.

Several implications can be drawn from the findings of this study. First, this study highlighted the importance of emergency response capability and readiness of higher education institutions in case another crisis strikes again. Critical areas that need utmost attention include (but not limited to) national and institutional policies, protocol and guidelines, technological infrastructure and resources, instructional delivery, staff development, potential inequalities, and collaboration among key stakeholders (i.e., parents, students, teachers, school leaders, industry, government education agencies, and community). Second, the findings have expanded our understanding of the different challenges that students might confront when we abruptly shift to full online learning, particularly those from countries with limited resources, poor Internet infrastructure, and poor home learning environment. Schools with a similar learning context could use the findings of this study in developing and enhancing their respective learning continuity plans to mitigate the adverse impact of the pandemic. This study would also provide students relevant information needed to reflect on the possible strategies that they may employ to overcome the challenges. These are critical information necessary for effective policymaking, decision-making, and future implementation of online learning. Third, teachers may find the results useful in providing proper interventions to address the reported challenges, particularly in the most critical areas. Finally, the findings provided us a nuanced understanding of the interdependence of learning tools, learners, and learning outcomes within an online learning environment; thus, giving us a multiperspective of hows and whys of a successful migration to full online learning.

Some limitations in this study need to be acknowledged and addressed in future studies. One limitation of this study is that it exclusively focused on students’ perspectives. Future studies may widen the sample by including all other actors taking part in the teaching–learning process. Researchers may go deeper by investigating teachers’ views and experience to have a complete view of the situation and how different elements interact between them or affect the others. Future studies may also identify some teacher-related factors that could influence students’ online learning experience. In the case of students, their age, sex, and degree programs may be examined in relation to the specific challenges and strategies they experience. Although the study involved a relatively large sample size, the participants were limited to college students from a Philippine university. To increase the robustness of the findings, future studies may expand the learning context to K-12 and several higher education institutions from different geographical regions. As a final note, this pandemic has undoubtedly reshaped and pushed the education system to its limits. However, this unprecedented event is the same thing that will make the education system stronger and survive future threats.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Adarkwah, M. A. (2021). “I’m not against online teaching, but what about us?”: ICT in Ghana post Covid-19. Education and Information Technologies, 26 (2), 1665–1685.

Article   Google Scholar  

Almaiah, M. A., Al-Khasawneh, A., & Althunibat, A. (2020). Exploring the critical challenges and factors influencing the E-learning system usage during COVID-19 pandemic. Education and Information Technologies, 25 , 5261–5280.

Araujo, T., Wonneberger, A., Neijens, P., & de Vreese, C. (2017). How much time do you spend online? Understanding and improving the accuracy of self-reported measures of Internet use. Communication Methods and Measures, 11 (3), 173–190.

Barrot, J. S. (2016). Using Facebook-based e-portfolio in ESL writing classrooms: Impact and challenges. Language, Culture and Curriculum, 29 (3), 286–301.

Barrot, J. S. (2018). Facebook as a learning environment for language teaching and learning: A critical analysis of the literature from 2010 to 2017. Journal of Computer Assisted Learning, 34 (6), 863–875.

Barrot, J. S. (2020). Scientific mapping of social media in education: A decade of exponential growth. Journal of Educational Computing Research .  https://doi.org/10.1177/0735633120972010 .

Barrot, J. S. (2021). Social media as a language learning environment: A systematic review of the literature (2008–2019). Computer Assisted Language Learning . https://doi.org/10.1080/09588221.2021.1883673 .

Bergen, N., & Labonté, R. (2020). “Everything is perfect, and we have no problems”: Detecting and limiting social desirability bias in qualitative research. Qualitative Health Research, 30 (5), 783–792.

Birks, M., & Mills, J. (2011). Grounded theory: A practical guide . Sage.

Boelens, R., De Wever, B., & Voet, M. (2017). Four key challenges to the design of blended learning: A systematic literature review. Educational Research Review, 22 , 1–18.

Buehler, M. A. (2004). Where is the library in course management software? Journal of Library Administration, 41 (1–2), 75–84.

Carter, R. A., Jr., Rice, M., Yang, S., & Jackson, H. A. (2020). Self-regulated learning in online learning environments: Strategies for remote learning. Information and Learning Sciences, 121 (5/6), 321–329.

Cavanaugh, C. S., Barbour, M. K., & Clark, T. (2009). Research and practice in K-12 online learning: A review of open access literature. The International Review of Research in Open and Distributed Learning, 10 (1), 1–22.

Cicchetti, D., Bronen, R., Spencer, S., Haut, S., Berg, A., Oliver, P., & Tyrer, P. (2006). Rating scales, scales of measurement, issues of reliability: Resolving some critical issues for clinicians and researchers. The Journal of Nervous and Mental Disease, 194 (8), 557–564.

Copeland, W. E., McGinnis, E., Bai, Y., Adams, Z., Nardone, H., Devadanam, V., & Hudziak, J. J. (2021). Impact of COVID-19 pandemic on college student mental health and wellness. Journal of the American Academy of Child & Adolescent Psychiatry, 60 (1), 134–141.

Day, T., Chang, I. C. C., Chung, C. K. L., Doolittle, W. E., Housel, J., & McDaniel, P. N. (2021). The immediate impact of COVID-19 on postsecondary teaching and learning. The Professional Geographer, 73 (1), 1–13.

Donitsa-Schmidt, S., & Ramot, R. (2020). Opportunities and challenges: Teacher education in Israel in the Covid-19 pandemic. Journal of Education for Teaching, 46 (4), 586–595.

Drane, C., Vernon, L., & O’Shea, S. (2020). The impact of ‘learning at home’on the educational outcomes of vulnerable children in Australia during the COVID-19 pandemic. Literature Review Prepared by the National Centre for Student Equity in Higher Education. Curtin University, Australia.

Elaish, M., Shuib, L., Ghani, N., & Yadegaridehkordi, E. (2019). Mobile English language learning (MELL): A literature review. Educational Review, 71 (2), 257–276.

Fawaz, M., Al Nakhal, M., & Itani, M. (2021). COVID-19 quarantine stressors and management among Lebanese students: A qualitative study.  Current Psychology , 1–8.

Franchi, T. (2020). The impact of the Covid-19 pandemic on current anatomy education and future careers: A student’s perspective. Anatomical Sciences Education, 13 (3), 312–315.

Garcia, R., Falkner, K., & Vivian, R. (2018). Systematic literature review: Self-regulated learning strategies using e-learning tools for computer science. Computers & Education, 123 , 150–163.

Gonzalez, T., De La Rubia, M. A., Hincz, K. P., Comas-Lopez, M., Subirats, L., Fort, S., & Sacha, G. M. (2020). Influence of COVID-19 confinement on students’ performance in higher education. PLoS One, 15 (10), e0239490.

Hew, K. F., Jia, C., Gonda, D. E., & Bai, S. (2020). Transitioning to the “new normal” of learning in unpredictable times: Pedagogical practices and learning performance in fully online flipped classrooms. International Journal of Educational Technology in Higher Education, 17 (1), 1–22.

Huang, Q. (2019). Comparing teacher’s roles of F2F learning and online learning in a blended English course. Computer Assisted Language Learning, 32 (3), 190–209.

John Hopkins University. (2021). Global map . https://coronavirus.jhu.edu/

Kapasia, N., Paul, P., Roy, A., Saha, J., Zaveri, A., Mallick, R., & Chouhan, P. (2020). Impact of lockdown on learning status of undergraduate and postgraduate students during COVID-19 pandemic in West Bengal. India . Children and Youth Services Review, 116 , 105194.

Kebritchi, M., Lipschuetz, A., & Santiague, L. (2017). Issues and challenges for teaching successful online courses in higher education: A literature review. Journal of Educational Technology Systems, 46 (1), 4–29.

Khalil, R., Mansour, A. E., Fadda, W. A., Almisnid, K., Aldamegh, M., Al-Nafeesah, A., & Al-Wutayd, O. (2020). The sudden transition to synchronized online learning during the COVID-19 pandemic in Saudi Arabia: A qualitative study exploring medical students’ perspectives. BMC Medical Education, 20 (1), 1–10.

Matsumoto, K. (1994). Introspection, verbal reports and second language learning strategy research. Canadian Modern Language Review, 50 (2), 363–386.

Ng, W. (2012). Can we teach digital natives digital literacy? Computers & Education, 59 (3), 1065–1078.

Pham, T., & Nguyen, H. (2020). COVID-19: Challenges and opportunities for Vietnamese higher education. Higher Education in Southeast Asia and beyond, 8 , 22–24.

Google Scholar  

Rasheed, R. A., Kamsin, A., & Abdullah, N. A. (2020). Challenges in the online component of blended learning: A systematic review. Computers & Education, 144 , 103701.

Recker, M. M., Dorward, J., & Nelson, L. M. (2004). Discovery and use of online learning resources: Case study findings. Educational Technology & Society, 7 (2), 93–104.

Roblek, V., Mesko, M., Dimovski, V., & Peterlin, J. (2019). Smart technologies as social innovation and complex social issues of the Z generation. Kybernetes, 48 (1), 91–107.

Seplaki, C. L., Agree, E. M., Weiss, C. O., Szanton, S. L., Bandeen-Roche, K., & Fried, L. P. (2014). Assistive devices in context: Cross-sectional association between challenges in the home environment and use of assistive devices for mobility. The Gerontologist, 54 (4), 651–660.

Simbulan, N. (2020). COVID-19 and its impact on higher education in the Philippines. Higher Education in Southeast Asia and beyond, 8 , 15–18.

Singh, K., Srivastav, S., Bhardwaj, A., Dixit, A., & Misra, S. (2020). Medical education during the COVID-19 pandemic: a single institution experience. Indian Pediatrics, 57 (7), 678–679.

Singh, V., & Thurman, A. (2019). How many ways can we define online learning? A systematic literature review of definitions of online learning (1988–2018). American Journal of Distance Education, 33 (4), 289–306.

Spector, P. (1994). Using self-report questionnaires in OB research: A comment on the use of a controversial method. Journal of Organizational Behavior, 15 (5), 385–392.

Suryaman, M., Cahyono, Y., Muliansyah, D., Bustani, O., Suryani, P., Fahlevi, M., & Munthe, A. P. (2020). COVID-19 pandemic and home online learning system: Does it affect the quality of pharmacy school learning? Systematic Reviews in Pharmacy, 11 , 524–530.

Tallent-Runnels, M. K., Thomas, J. A., Lan, W. Y., Cooper, S., Ahern, T. C., Shaw, S. M., & Liu, X. (2006). Teaching courses online: A review of the research. Review of Educational Research, 76 (1), 93–135.

Tang, T., Abuhmaid, A. M., Olaimat, M., Oudat, D. M., Aldhaeebi, M., & Bamanger, E. (2020). Efficiency of flipped classroom with online-based teaching under COVID-19.  Interactive Learning Environments , 1–12.

Usher, M., & Barak, M. (2020). Team diversity as a predictor of innovation in team projects of face-to-face and online learners. Computers & Education, 144 , 103702.

Varea, V., & González-Calvo, G. (2020). Touchless classes and absent bodies: Teaching physical education in times of Covid-19.  Sport, Education and Society , 1–15.

Wallace, R. M. (2003). Online learning in higher education: A review of research on interactions among teachers and students. Education, Communication & Information, 3 (2), 241–280.

World Health Organization (2020). Coronavirus . https://www.who.int/health-topics/coronavirus#tab=tab_1

Xue, E., Li, J., Li, T., & Shang, W. (2020). China’s education response to COVID-19: A perspective of policy analysis.  Educational Philosophy and Theory , 1–13.

Download references

No funding was received in the conduct of this study.

Author information

Authors and affiliations.

College of Education, Arts and Sciences, National University, Manila, Philippines

Jessie S. Barrot, Ian I. Llenares & Leo S. del Rosario

You can also search for this author in PubMed   Google Scholar

Contributions

Jessie Barrot led the planning, prepared the instrument, wrote the report, and processed and analyzed data. Ian Llenares participated in the planning, fielded the instrument, processed and analyzed data, reviewed the instrument, and contributed to report writing. Leo del Rosario participated in the planning, fielded the instrument, processed and analyzed data, reviewed the instrument, and contributed to report writing.

Corresponding author

Correspondence to Jessie S. Barrot .

Ethics declarations

Ethics approval.

The study has undergone appropriate ethics protocol.

Consent to participate

Informed consent was sought from the participants.

Consent for publication

Authors consented the publication. Participants consented to publication as long as confidentiality is observed.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Cite this article.

Barrot, J.S., Llenares, I.I. & del Rosario, L.S. Students’ online learning challenges during the pandemic and how they cope with them: The case of the Philippines. Educ Inf Technol 26 , 7321–7338 (2021). https://doi.org/10.1007/s10639-021-10589-x

Download citation

Received : 22 January 2021

Accepted : 17 May 2021

Published : 28 May 2021

Issue Date : November 2021

DOI : https://doi.org/10.1007/s10639-021-10589-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Remote learning
• Online learning
• Online learning strategies
• Higher education
• Find a journal
• Publish with us

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• Advanced Search
• Journal List
• Wiley - PMC COVID-19 Collection

Students’ experience of online learning during the COVID‐19 pandemic: A province‐wide survey study

Lixiang yan.

1 Centre for Learning Analytics at Monash, Faculty of Information Technology, Monash University, Clayton VIC, Australia

Alexander Whitelock‐Wainwright

2 Portfolio of the Deputy Vice‐Chancellor (Education), Monash University, Melbourne VIC, Australia

Quanlong Guan

3 Department of Computer Science, Jinan University, Guangzhou China

Gangxin Wen

4 College of Cyber Security, Jinan University, Guangzhou China

Dragan Gašević

Guanliang chen, associated data.

The data is not openly available as it is restricted by the Chinese government.

Online learning is currently adopted by educational institutions worldwide to provide students with ongoing education during the COVID‐19 pandemic. Even though online learning research has been advancing in uncovering student experiences in various settings (i.e., tertiary, adult, and professional education), very little progress has been achieved in understanding the experience of the K‐12 student population, especially when narrowed down to different school‐year segments (i.e., primary and secondary school students). This study explores how students at different stages of their K‐12 education reacted to the mandatory full‐time online learning during the COVID‐19 pandemic. For this purpose, we conducted a province‐wide survey study in which the online learning experience of 1,170,769 Chinese students was collected from the Guangdong Province of China. We performed cross‐tabulation and Chi‐square analysis to compare students’ online learning conditions, experiences, and expectations. Results from this survey study provide evidence that students’ online learning experiences are significantly different across school years. Foremost, policy implications were made to advise government authorises and schools on improving the delivery of online learning, and potential directions were identified for future research into K‐12 online learning.

Practitioner notes

What is already known about this topic

• Online learning has been widely adopted during the COVID‐19 pandemic to ensure the continuation of K‐12 education.
• Student success in K‐12 online education is substantially lower than in conventional schools.
• Students experienced various difficulties related to the delivery of online learning.

What this paper adds

• Provide empirical evidence for the online learning experience of students in different school years.
• Identify the different needs of students in primary, middle, and high school.
• Identify the challenges of delivering online learning to students of different age.

Implications for practice and/or policy

• Authority and schools need to provide sufficient technical support to students in online learning.
• The delivery of online learning needs to be customised for students in different school years.

INTRODUCTION

The ongoing COVID‐19 pandemic poses significant challenges to the global education system. By July 2020, the UN Educational, Scientific and Cultural Organization (2020) reported nationwide school closure in 111 countries, affecting over 1.07 billion students, which is around 61% of the global student population. Traditional brick‐and‐mortar schools are forced to transform into full‐time virtual schools to provide students with ongoing education (Van Lancker & Parolin,  2020 ). Consequently, students must adapt to the transition from face‐to‐face learning to fully remote online learning, where synchronous video conferences, social media, and asynchronous discussion forums become their primary venues for knowledge construction and peer communication.

For K‐12 students, this sudden transition is problematic as they often lack prior online learning experience (Barbour & Reeves,  2009 ). Barbour and LaBonte ( 2017 ) estimated that even in countries where online learning is growing rapidly, such as USA and Canada, less than 10% of the K‐12 student population had prior experience with this format. Maladaptation to online learning could expose inexperienced students to various vulnerabilities, including decrements in academic performance (Molnar et al.,  2019 ), feeling of isolation (Song et al.,  2004 ), and lack of learning motivation (Muilenburg & Berge,  2005 ). Unfortunately, with confirmed cases continuing to rise each day, and new outbreaks occur on a global scale, full‐time online learning for most students could last longer than anticipated (World Health Organization,  2020 ). Even after the pandemic, the current mass adoption of online learning could have lasting impacts on the global education system, and potentially accelerate and expand the rapid growth of virtual schools on a global scale (Molnar et al.,  2019 ). Thus, understanding students' learning conditions and their experiences of online learning during the COVID pandemic becomes imperative.

Emerging evidence on students’ online learning experience during the COVID‐19 pandemic has identified several major concerns, including issues with internet connection (Agung et al.,  2020 ; Basuony et al.,  2020 ), problems with IT equipment (Bączek et al.,  2021 ; Niemi & Kousa,  2020 ), limited collaborative learning opportunities (Bączek et al.,  2021 ; Yates et al.,  2020 ), reduced learning motivation (Basuony et al.,  2020 ; Niemi & Kousa,  2020 ; Yates et al.,  2020 ), and increased learning burdens (Niemi & Kousa,  2020 ). Although these findings provided valuable insights about the issues students experienced during online learning, information about their learning conditions and future expectations were less mentioned. Such information could assist educational authorises and institutions to better comprehend students’ difficulties and potentially improve their online learning experience. Additionally, most of these recent studies were limited to higher education, except for Yates et al. ( 2020 ) and Niemi and Kousa’s ( 2020 ) studies on senior high school students. Empirical research targeting the full spectrum of K‐12students remain scarce. Therefore, to address these gaps, the current paper reports the findings of a large‐scale study that sought to explore K‐12 students’ online learning experience during the COVID‐19 pandemic in a provincial sample of over one million Chinese students. The findings of this study provide policy recommendations to educational institutions and authorities regarding the delivery of K‐12 online education.

LITERATURE REVIEW

Learning conditions and technologies.

Having stable access to the internet is critical to students’ learning experience during online learning. Berge ( 2005 ) expressed the concern of the divide in digital‐readiness, and the pedagogical approach between different countries could influence students’ online learning experience. Digital‐readiness is the availability and adoption of information technologies and infrastructures in a country. Western countries like America (3rd) scored significantly higher in digital‐readiness compared to Asian countries like China (54th; Cisco,  2019 ). Students from low digital‐readiness countries could experience additional technology‐related problems. Supporting evidence is emerging in recent studies conducted during the COVID‐19 pandemic. In Egypt's capital city, Basuony et al. ( 2020 ) found that only around 13.9%of the students experienced issues with their internet connection. Whereas more than two‐thirds of the students in rural Indonesia reported issues of unstable internet, insufficient internet data, and incompatible learning device (Agung et al.,  2020 ).

Another influential factor for K‐12 students to adequately adapt to online learning is the accessibility of appropriate technological devices, especially having access to a desktop or a laptop (Barbour et al., 2018 ). However, it is unlikely for most of the students to satisfy this requirement. Even in higher education, around 76% of students reported having incompatible devices for online learning and only 15% of students used laptop for online learning, whereas around 85% of them used smartphone (Agung et al.,  2020 ). It is very likely that K‐12 students also suffer from this availability issue as they depend on their parents to provide access to relevant learning devices.

Technical issues surrounding technological devices could also influence students’ experience in online learning. (Barbour & Reeves,  2009 ) argues that students need to have a high level of digital literacy to find and use relevant information and communicate with others through technological devices. Students lacking this ability could experience difficulties in online learning. Bączek et al. ( 2021 ) found that around 54% of the medical students experienced technical problems with IT equipment and this issue was more prevalent in students with lower years of tertiary education. Likewise, Niemi and Kousa ( 2020 ) also find that students in a Finish high school experienced increased amounts of technical problems during the examination period, which involved additional technical applications. These findings are concerning as young children and adolescent in primary and lower secondary school could be more vulnerable to these technical problems as they are less experienced with the technologies in online learning (Barbour & LaBonte,  2017 ). Therefore, it is essential to investigate the learning conditions and the related difficulties experienced by students in K‐12 education as the extend of effects on them remain underexplored.

Learning experience and interactions

Apart from the aforementioned issues, the extent of interaction and collaborative learning opportunities available in online learning could also influence students’ experience. The literature on online learning has long emphasised the role of effective interaction for the success of student learning. According to Muirhead and Juwah ( 2004 ), interaction is an event that can take the shape of any type of communication between two or subjects and objects. Specifically, the literature acknowledges the three typical forms of interactions (Moore,  1989 ): (i) student‐content, (ii) student‐student, and (iii) student‐teacher. Anderson ( 2003 ) posits, in the well‐known interaction equivalency theorem, learning experiences will not deteriorate if only one of the three interaction is of high quality, and the other two can be reduced or even eliminated. Quality interaction can be accomplished by across two dimensions: (i) structure—pedagogical means that guide student interaction with contents or other students and (ii) dialogue—communication that happens between students and teachers and among students. To be able to scale online learning and prevent the growth of teaching costs, the emphasise is typically on structure (i.e., pedagogy) that can promote effective student‐content and student‐student interaction. The role of technology and media is typically recognised as a way to amplify the effect of pedagogy (Lou et al.,  2006 ). Novel technological innovations—for example learning analytics‐based personalised feedback at scale (Pardo et al.,  2019 ) —can also empower teachers to promote their interaction with students.

Online education can lead to a sense of isolation, which can be detrimental to student success (McInnerney & Roberts,  2004 ). Therefore, integration of social interaction into pedagogy for online learning is essential, especially at the times when students do not actually know each other or have communication and collaboration skills underdeveloped (Garrison et al.,  2010 ; Gašević et al.,  2015 ). Unfortunately, existing evidence suggested that online learning delivery during the COVID‐19 pandemic often lacks interactivity and collaborative experiences (Bączek et al.,  2021 ; Yates et al.,  2020 ). Bączek et al., ( 2021 ) found that around half of the medical students reported reduced interaction with teachers, and only 4% of students think online learning classes are interactive. Likewise, Yates et al. ( 2020 )’s study in high school students also revealed that over half of the students preferred in‐class collaboration over online collaboration as they value the immediate support and the proximity to teachers and peers from in‐class interaction.

Learning expectations and age differentiation

Although these studies have provided valuable insights and stressed the need for more interactivity in online learning, K‐12 students in different school years could exhibit different expectations for the desired activities in online learning. Piaget's Cognitive Developmental Theory illustrated children's difficulties in understanding abstract and hypothetical concepts (Thomas,  2000 ). Primary school students will encounter many abstract concepts in their STEM education (Uttal & Cohen,  2012 ). In face‐to‐face learning, teachers provide constant guidance on students’ learning progress and can help them to understand difficult concepts. Unfortunately, the level of guidance significantly drops in online learning, and, in most cases, children have to face learning obstacles by themselves (Barbour,  2013 ). Additionally, lower primary school students may lack the metacognitive skills to use various online learning functions, maintain engagement in synchronous online learning, develop and execute self‐regulated learning plans, and engage in meaningful peer interactions during online learning (Barbour,  2013 ; Broadbent & Poon,  2015 ; Huffaker & Calvert, 2003; Wang et al.,  2013 ). Thus, understanding these younger students’ expectations is imperative as delivering online learning to them in the same way as a virtual high school could hinder their learning experiences. For students with more matured metacognition, their expectations of online learning could be substantially different from younger students. Niemi et al.’s study ( 2020 ) with students in a Finish high school have found that students often reported heavy workload and fatigue during online learning. These issues could cause anxiety and reduce students’ learning motivation, which would have negative consequences on their emotional well‐being and academic performance (Niemi & Kousa,  2020 ; Yates et al.,  2020 ), especially for senior students who are under the pressure of examinations. Consequently, their expectations of online learning could be orientated toward having additional learning support functions and materials. Likewise, they could also prefer having more opportunities for peer interactions as these interactions are beneficial to their emotional well‐being and learning performance (Gašević et al., 2013 ; Montague & Rinaldi, 2001 ). Therefore, it is imperative to investigate the differences between online learning expectations in students of different school years to suit their needs better.

Research questions

By building upon the aforementioned relevant works, this study aimed to contribute to the online learning literature with a comprehensive understanding of the online learning experience that K‐12 students had during the COVID‐19 pandemic period in China. Additionally, this study also aimed to provide a thorough discussion of what potential actions can be undertaken to improve online learning delivery. Formally, this study was guided by three research questions (RQs):

RQ1 . What learning conditions were experienced by students across 12 years of education during their online learning process in the pandemic period? RQ2 . What benefits and obstacles were perceived by students across 12 years of education when performing online learning? RQ3 . What expectations do students, across 12 years of education, have for future online learning practices ?

Participants

The total number of K‐12 students in the Guangdong Province of China is around 15 million. In China, students of Year 1–6, Year 7–9, and Year 10–12 are referred to as students of primary school, middle school, and high school, respectively. Typically, students in China start their study in primary school at the age of around six. At the end of their high‐school study, students have to take the National College Entrance Examination (NCEE; also known as Gaokao) to apply for tertiary education. The survey was administrated across the whole Guangdong Province, that is the survey was exposed to all of the 15 million K‐12 students, though it was not mandatory for those students to accomplish the survey. A total of 1,170,769 students completed the survey, which accounts for a response rate of 7.80%. After removing responses with missing values and responses submitted from the same IP address (duplicates), we had 1,048,575 valid responses, which accounts to about 7% of the total K‐12 students in the Guangdong Province. The number of students in different school years is shown in Figure  1 . Overall, students were evenly distributed across different school years, except for a smaller sample in students of Year 10–12.

The number of students in each school year

Survey design

The survey was designed collaboratively by multiple relevant parties. Firstly, three educational researchers working in colleges and universities and three educational practitioners working in the Department of Education in Guangdong Province were recruited to co‐design the survey. Then, the initial draft of the survey was sent to 30 teachers from different primary and secondary schools, whose feedback and suggestions were considered to improve the survey. The final survey consisted of a total of 20 questions, which, broadly, can be classified into four categories: demographic, behaviours, experiences, and expectations. Details are available in Appendix.

All K‐12 students in the Guangdong Province were made to have full‐time online learning from March 1, 2020 after the outbreak of COVID‐19 in January in China. A province‐level online learning platform was provided to all schools by the government. In addition to the learning platform, these schools can also use additional third‐party platforms to facilitate the teaching activities, for example WeChat and Dingding, which provide services similar to WhatsApp and Zoom. The main change for most teachers was that they had to shift the classroom‐based lectures to online lectures with the aid of web‐conferencing tools. Similarly, these teachers also needed to perform homework marking and have consultation sessions in an online manner.

The Department of Education in the Guangdong Province of China distributed the survey to all K‐12 schools in the province on March 21, 2020 and collected responses on March 26, 2020. Students could access and answer the survey anonymously by either scan the Quick Response code along with the survey or click the survey address link on their mobile device. The survey was administrated in a completely voluntary manner and no incentives were given to the participants. Ethical approval was granted by the Department of Education in the Guangdong Province. Parental approval was not required since the survey was entirely anonymous and facilitated by the regulating authority, which satisfies China's ethical process.

The original survey was in Chinese, which was later translated by two bilingual researchers and verified by an external translator who is certified by the Australian National Accreditation Authority of Translators and Interpreters. The original and translated survey questionnaires are available in Supporting Information. Given the limited space we have here and the fact that not every survey item is relevant to the RQs, the following items were chosen to answer the RQs: item Q3 (learning media) and Q11 (learning approaches) for RQ1, item Q13 (perceived obstacle) and Q19 (perceived benefits) for RQ2, and item Q19 (expected learning activities) for RQ3. Cross‐tabulation based approaches were used to analyse the collected data. To scrutinise whether the differences displayed by students of different school years were statistically significant, we performed Chi‐square tests and calculated the Cramer's V to assess the strengths of the association after chi‐square had determined significance.

For the analyses, students were segmented into four categories based on their school years, that is Year 1–3, Year 4–6, Year 7–9, and Year 10–12, to provide a clear understanding of the different experiences and needs that different students had for online learning. This segmentation was based on the educational structure of Chinese schools: elementary school (Year 1–6), middle school (Year 7–9), and high school (Year 10–12). Children in elementary school can further be segmented into junior (Year 1–3) or senior (Year 4–6) students because senior elementary students in China are facing more workloads compared to junior students due to the provincial Middle School Entry Examination at the end of Year 6.

Learning conditions—RQ1

Learning media.

The Chi‐square test showed significant association between school years and students’ reported usage of learning media, χ 2 (55, N  = 1,853,952) = 46,675.38, p  < 0.001. The Cramer's V is 0.07 ( df ∗ = 5), which indicates a small‐to‐medium effect according to Cohen’s ( 1988 ) guidelines. Based on Figure  2 , we observed that an average of up to 87.39% students used smartphones to perform online learning, while only 25.43% students used computer, which suggests that smartphones, with widespread availability in China (2020), have been adopted by students for online learning. As for the prevalence of the two media, we noticed that both smartphones ( χ 2 (3, N  = 1,048,575) = 9,395.05, p < 0.001, Cramer's V  = 0.10 ( df ∗ = 1)) and computers ( χ 2 (3, N  = 1,048,575) = 11,025.58, p <.001, Cramer's V  = 0.10 ( df ∗ = 1)) were more adopted by high‐school‐year (Year 7–12) than early‐school‐year students (Year 1–6), both with a small effect size. Besides, apparent discrepancies can be observed between the usages of TV and paper‐based materials across different school years, that is early‐school‐year students reported more TV usage ( χ 2 (3, N  = 1,048,575) = 19,505.08, p <.001), with a small‐to‐medium effect size, Cramer's V  = 0.14( df ∗ = 1). High‐school‐year students (especially Year 10–12) reported more usage of paper‐based materials ( χ 2 (3, N  = 1,048,575) = 23,401.64, p < 0.001), with a small‐to‐medium effect size, Cramer's V  = 0.15( df ∗ = 1).

Learning media used by students in online learning

Learning approaches

School years is also significantly associated with the different learning approaches students used to tackle difficult concepts during online learning, χ 2 (55, N  = 2,383,751) = 58,030.74, p < 0.001. The strength of this association is weak to moderate as shown by the Cramer's V (0.07, df ∗ = 5; Cohen,  1988 ). When encountering problems related to difficult concepts, students typically chose to “solve independently by searching online” or “rewatch recorded lectures” instead of consulting to their teachers or peers (Figure  3 ). This is probably because, compared to classroom‐based education, it is relatively less convenient and more challenging for students to seek help from others when performing online learning. Besides, compared to high‐school‐year students, early‐school‐year students (Year 1–6), reported much less use of “solve independently by searching online” ( χ 2 (3, N  = 1,048,575) = 48,100.15, p <.001), with a small‐to‐medium effect size, Cramer's V  = 0.21 ( df ∗ = 1). Also, among those approaches of seeking help from others, significantly more high‐school‐year students preferred “communicating with other students” than early‐school‐year students ( χ 2 (3, N  = 1,048,575) = 81,723.37, p < 0.001), with a medium effect size, Cramer's V  = 0.28 ( df ∗ = 1).

Learning approaches used by students in online learning

Perceived benefits and obstacles—RQ2

Perceived benefits.

The association between school years and perceived benefits in online learning is statistically significant, χ 2 (66, N  = 2,716,127) = 29,534.23, p  < 0.001, and the Cramer's V (0.04, df ∗ = 6) indicates a small effect (Cohen,  1988 ). Unsurprisingly, benefits brought by the convenience of online learning are widely recognised by students across all school years (Figure  4 ), that is up to 75% of students reported that it is “more convenient to review course content” and 54% said that they “can learn anytime and anywhere” . Besides, we noticed that about 50% of early‐school‐year students appreciated the “access to courses delivered by famous teachers” and 40%–47% of high‐school‐year students indicated that online learning is “helpful to develop self‐regulation and autonomy” .

Perceived benefits of online learning reported by students

Perceived obstacles

The Chi‐square test shows a significant association between school years and students’ perceived obstacles in online learning, χ 2 (77, N  = 2,699,003) = 31,987.56, p < 0.001. This association is relatively weak as shown by the Cramer's V (0.04, df ∗ = 7; Cohen,  1988 ). As shown in Figure  5 , the biggest obstacles encountered by up to 73% of students were the “eyestrain caused by long staring at screens” . Disengagement caused by nearby disturbance was reported by around 40% of students, especially those of Year 1–3 and 10–12. Technological‐wise, about 50% of students experienced poor Internet connection during their learning process, and around 20% of students reported the “confusion in setting up the platforms” across of school years.

Perceived obstacles of online learning reported by students

Expectations for future practices of online learning – RQ3

Online learning activities.

The association between school years and students’ expected online learning activities is significant, χ 2 (66, N  = 2,416,093) = 38,784.81, p < 0.001. The Cramer's V is 0.05 ( df ∗ = 6) which suggests a small effect (Cohen,  1988 ). As shown in Figure  6 , the most expected activity for future online learning is “real‐time interaction with teachers” (55%), followed by “online group discussion and collaboration” (38%). We also observed that more early‐school‐year students expect reflective activities, such as “regular online practice examinations” ( χ 2 (3, N  = 1,048,575) = 11,644.98, p < 0.001), with a small effect size, Cramer's V  = 0.11 ( df ∗ = 1). In contrast, more high‐school‐year students expect “intelligent recommendation system …” ( χ 2 (3, N  = 1,048,575) = 15,327.00, p < 0.001), with a small effect size, Cramer's V  = 0.12 ( df ∗ = 1).

Students’ expected online learning activities

Regarding students’ learning conditions, substantial differences were observed in learning media, family dependency, and learning approaches adopted in online learning between students in different school years. The finding of more computer and smartphone usage in high‐school‐year than early‐school‐year students can probably be explained by that, with the growing abilities in utilising these media as well as the educational systems and tools which run on these media, high‐school‐year students tend to make better use of these media for online learning practices. Whereas, the differences in paper‐based materials may imply that high‐school‐year students in China have to accomplish a substantial amount of exercise, assignments, and exam papers to prepare for the National College Entrance Examination (NCEE), whose delivery was not entirely digitised due to the sudden transition to online learning. Meanwhile, high‐school‐year students may also have preferred using paper‐based materials for exam practice, as eventually, they would take their NCEE in the paper format. Therefore, these substantial differences in students’ usage of learning media should be addressed by customising the delivery method of online learning for different school years.

Other than these between‐age differences in learning media, the prevalence of smartphone in online learning resonates with Agung et al.’s ( 2020 ) finding on the issues surrounding the availability of compatible learning device. The prevalence of smartphone in K‐12 students is potentially problematic as the majority of the online learning platform and content is designed for computer‐based learning (Berge,  2005 ; Molnar et al.,  2019 ). Whereas learning with smartphones has its own unique challenges. For example, Gikas and Grant ( 2013 ) discovered that students who learn with smartphone experienced frustration with the small screen‐size, especially when trying to type with the tiny keypad. Another challenge relates to the distraction of various social media applications. Although similar distractions exist in computer and web‐based social media, the level of popularity, especially in the young generation, are much higher in mobile‐based social media (Montag et al.,  2018 ). In particular, the message notification function in smartphones could disengage students from learning activities and allure them to social media applications (Gikas & Grant,  2013 ). Given these challenges of learning with smartphones, more research efforts should be devoted to analysing students’ online learning behaviour in the setting of mobile learning to accommodate their needs better.

The differences in learning approaches, once again, illustrated that early‐school‐year students have different needs compared to high‐school‐year students. In particular, the low usage of the independent learning methods in early‐school‐year students may reflect their inability to engage in independent learning. Besides, the differences in help seeking behaviours demonstrated the distinctive needs for communication and interaction between different students, that is early‐school‐year students have a strong reliance on teachers and high‐school‐year students, who are equipped with stronger communication ability, are more inclined to interact with their peers. This finding implies that the design of online learning platforms should take students’ different needs into account. Thus, customisation is urgently needed for the delivery of online learning to different school years.

In terms of the perceived benefits and challenges of online learning, our results resonate with several previous findings. In particular, the benefits of convenience are in line with the flexibility advantages of online learning, which were mentioned in prior works (Appana,  2008 ; Bączek et al.,  2021 ; Barbour,  2013 ; Basuony et al.,  2020 ; Harvey et al.,  2014 ). Early‐school‐year students’ higher appreciation in having “access to courses delivered by famous teachers” and lower appreciation in the independent learning skills developed through online learning are also in line with previous literature (Barbour,  2013 ; Harvey et al.,  2014 ; Oliver et al.,  2009 ). Again, these similar findings may indicate the strong reliance that early‐school‐year students place on teachers, while high‐school‐year students are more capable of adapting to online learning by developing independent learning skills.

Technology‐wise, students’ experience of poor internet connection and confusion in setting up online learning platforms are particularly concerning. The problem of poor internet connection corroborated the findings reported in prior studies (Agung et al.,  2020 ; Barbour,  2013 ; Basuony et al.,  2020 ; Berge,  2005 ; Rice,  2006 ), that is the access issue surrounded the digital divide as one of the main challenges of online learning. In the era of 4G and 5G networks, educational authorities and institutions that deliver online education could fall into the misconception of most students have a stable internet connection at home. The internet issue we observed is particularly vital to students’ online learning experience as most students prefer real‐time communications (Figure  6 ), which rely heavily on stable internet connection. Likewise, the finding of students’ confusion in technology is also consistent with prior studies (Bączek et al.,  2021 ; Muilenburg & Berge,  2005 ; Niemi & Kousa,  2020 ; Song et al.,  2004 ). Students who were unsuccessfully in setting up the online learning platforms could potentially experience declines in confidence and enthusiasm for online learning, which would cause a subsequent unpleasant learning experience. Therefore, both the readiness of internet infrastructure and student technical skills remain as the significant challenges for the mass‐adoption of online learning.

On the other hand, students’ experience of eyestrain from extended screen time provided empirical evidence to support Spitzer’s ( 2001 ) speculation about the potential ergonomic impact of online learning. This negative effect is potentially related to the prevalence of smartphone device and the limited screen size of these devices. This finding not only demonstrates the potential ergonomic issues that would be caused by smartphone‐based online learning but also resonates with the aforementioned necessity of different platforms and content designs for different students.

A less‐mentioned problem in previous studies on online learning experiences is the disengagement caused by nearby disturbance, especially in Year 1–3 and 10–12. It is likely that early‐school‐year students suffered from this problem because of their underdeveloped metacognitive skills to concentrate on online learning without teachers’ guidance. As for high‐school‐year students, the reasons behind their disengagement require further investigation in the future. Especially it would be worthwhile to scrutinise whether this type of disengagement is caused by the substantial amount of coursework they have to undertake and the subsequent a higher level of pressure and a lower level of concentration while learning.

Across age‐level differences are also apparent in terms of students’ expectations of online learning. Although, our results demonstrated students’ needs of gaining social interaction with others during online learning, findings (Bączek et al.,  2021 ; Harvey et al.,  2014 ; Kuo et al.,  2014 ; Liu & Cavanaugh,  2012 ; Yates et al.,  2020 ). This need manifested differently across school years, with early‐school‐year students preferring more teacher interactions and learning regulation support. Once again, this finding may imply that early‐school‐year students are inadequate in engaging with online learning without proper guidance from their teachers. Whereas, high‐school‐year students prefer more peer interactions and recommendation to learning resources. This expectation can probably be explained by the large amount of coursework exposed to them. Thus, high‐school‐year students need further guidance to help them better direct their learning efforts. These differences in students’ expectations for future practices could guide the customisation of online learning delivery.

Implications

As shown in our results, improving the delivery of online learning not only requires the efforts of policymakers but also depend on the actions of teachers and parents. The following sub‐sections will provide recommendations for relevant stakeholders and discuss their essential roles in supporting online education.

Technical support

The majority of the students has experienced technical problems during online learning, including the internet lagging and confusion in setting up the learning platforms. These problems with technology could impair students’ learning experience (Kauffman,  2015 ; Muilenburg & Berge,  2005 ). Educational authorities and schools should always provide a thorough guide and assistance for students who are experiencing technical problems with online learning platforms or other related tools. Early screening and detection could also assist schools and teachers to direct their efforts more effectively in helping students with low technology skills (Wilkinson et al.,  2010 ). A potential identification method involves distributing age‐specific surveys that assess students’ Information and Communication Technology (ICT) skills at the beginning of online learning. For example, there are empirical validated ICT surveys available for both primary (Aesaert et al.,  2014 ) and high school (Claro et al.,  2012 ) students.

For students who had problems with internet lagging, the delivery of online learning should provide options that require fewer data and bandwidth. Lecture recording is the existing option but fails to address students’ need for real‐time interaction (Clark et al.,  2015 ; Malik & Fatima,  2017 ). A potential alternative involves providing students with the option to learn with digital or physical textbooks and audio‐conferencing, instead of screen sharing and video‐conferencing. This approach significantly reduces the amount of data usage and lowers the requirement of bandwidth for students to engage in smooth online interactions (Cisco,  2018 ). It also requires little additional efforts from teachers as official textbooks are often available for each school year, and thus, they only need to guide students through the materials during audio‐conferencing. Educational authority can further support this approach by making digital textbooks available for teachers and students, especially those in financial hardship. However, the lack of visual and instructor presence could potentially reduce students’ attention, recall of information, and satisfaction in online learning (Wang & Antonenko,  2017 ). Therefore, further research is required to understand whether the combination of digital or physical textbooks and audio‐conferencing is appropriate for students with internet problems. Alternatively, suppose the local technological infrastructure is well developed. In that case, governments and schools can also collaborate with internet providers to issue data and bandwidth vouchers for students who are experiencing internet problems due to financial hardship.

For future adoption of online learning, policymakers should consider the readiness of the local internet infrastructure. This recommendation is particularly important for developing countries, like Bangladesh, where the majority of the students reported the lack of internet infrastructure (Ramij & Sultana,  2020 ). In such environments, online education may become infeasible, and alternative delivery method could be more appropriate, for example, the Telesecundaria program provides TV education for rural areas of Mexico (Calderoni,  1998 ).

Other than technical problems, choosing a suitable online learning platform is also vital for providing students with a better learning experience. Governments and schools should choose an online learning platform that is customised for smartphone‐based learning, as the majority of students could be using smartphones for online learning. This recommendation is highly relevant for situations where students are forced or involuntarily engaged in online learning, like during the COVID‐19 pandemic, as they might not have access to a personal computer (Molnar et al.,  2019 ).

Customisation of delivery methods

Customising the delivery of online learning for students in different school years is the theme that appeared consistently across our findings. This customisation process is vital for making online learning an opportunity for students to develop independent learning skills, which could help prepare them for tertiary education and lifelong learning. However, the pedagogical design of K‐12 online learning programs should be differentiated from adult‐orientated programs as these programs are designed for independent learners, which is rarely the case for students in K‐12 education (Barbour & Reeves,  2009 ).

For early‐school‐year students, especially Year 1–3 students, providing them with sufficient guidance from both teachers and parents should be the priority as these students often lack the ability to monitor and reflect on learning progress. In particular, these students would prefer more real‐time interaction with teachers, tutoring from parents, and regular online practice examinations. These forms of guidance could help early‐school‐year students to cope with involuntary online learning, and potentially enhance their experience in future online learning. It should be noted that, early‐school‐year students demonstrated interest in intelligent monitoring and feedback systems for learning. Additional research is required to understand whether these young children are capable of understanding and using learning analytics that relay information on their learning progress. Similarly, future research should also investigate whether young children can communicate effectively through digital tools as potential inability could hinder student learning in online group activities. Therefore, the design of online learning for early‐school‐year students should focus less on independent learning but ensuring that students are learning effective under the guidance of teachers and parents.

In contrast, group learning and peer interaction are essential for older children and adolescents. The delivery of online learning for these students should focus on providing them with more opportunities to communicate with each other and engage in collaborative learning. Potential methods to achieve this goal involve assigning or encouraging students to form study groups (Lee et al.,  2011 ), directing students to use social media for peer communication (Dabbagh & Kitsantas,  2012 ), and providing students with online group assignments (Bickle & Rucker,  2018 ).

Special attention should be paid to students enrolled in high schools. For high‐school‐year students, in particular, students in Year 10–12, we also recommend to provide them with sufficient access to paper‐based learning materials, such as revision booklet and practice exam papers, so they remain familiar with paper‐based examinations. This recommendation applies to any students who engage in online learning but has to take their final examination in paper format. It is also imperative to assist high‐school‐year students who are facing examinations to direct their learning efforts better. Teachers can fulfil this need by sharing useful learning resources on the learning management system, if it is available, or through social media groups. Alternatively, students are interested in intelligent recommendation systems for learning resources, which are emerging in the literature (Corbi & Solans,  2014 ; Shishehchi et al.,  2010 ). These systems could provide personalised recommendations based on a series of evaluation on learners’ knowledge. Although it is infeasible for situations where the transformation to online learning happened rapidly (i.e., during the COVID‐19 pandemic), policymakers can consider embedding such systems in future online education.

Limitations

The current findings are limited to primary and secondary Chinese students who were involuntarily engaged in online learning during the COVID‐19 pandemic. Despite the large sample size, the population may not be representative as participants are all from a single province. Also, information about the quality of online learning platforms, teaching contents, and pedagogy approaches were missing because of the large scale of our study. It is likely that the infrastructures of online learning in China, such as learning platforms, instructional designs, and teachers’ knowledge about online pedagogy, were underprepared for the sudden transition. Thus, our findings may not represent the experience of students who voluntarily participated in well‐prepared online learning programs, in particular, the virtual school programs in America and Canada (Barbour & LaBonte,  2017 ; Molnar et al.,  2019 ). Lastly, the survey was only evaluated and validated by teachers but not students. Therefore, students with the lowest reading comprehension levels might have a different understanding of the items’ meaning, especially terminologies that involve abstract contracts like self‐regulation and autonomy in item Q17.

In conclusion, we identified across‐year differences between primary and secondary school students’ online learning experience during the COVID‐19 pandemic. Several recommendations were made for the future practice and research of online learning in the K‐12 student population. First, educational authorities and schools should provide sufficient technical support to help students to overcome potential internet and technical problems, as well as choosing online learning platforms that have been customised for smartphones. Second, customising the online pedagogy design for students in different school years, in particular, focusing on providing sufficient guidance for young children, more online collaborative opportunity for older children and adolescent, and additional learning resource for senior students who are facing final examinations.

CONFLICT OF INTEREST

There is no potential conflict of interest in this study.

ETHICS STATEMENT

The data are collected by the Department of Education of the Guangdong Province who also has the authority to approve research studies in K12 education in the province.

Supporting information

Supplementary Material

ACKNOWLEDGEMENTS

This work is supported by the National Natural Science Foundation of China (62077028, 61877029), the Science and Technology Planning Project of Guangdong (2020B0909030005, 2020B1212030003, 2020ZDZX3013, 2019B1515120010, 2018KTSCX016, 2019A050510024), the Science and Technology Planning Project of Guangzhou (201902010041), and the Fundamental Research Funds for the Central Universities (21617408, 21619404).

SURVEY ITEMS

Yan, L , Whitelock‐Wainwright, A , Guan, Q , Wen, G , Gašević, D , & Chen, G . Students’ experience of online learning during the COVID‐19 pandemic: A province‐wide survey study . Br J Educ Technol . 2021; 52 :2038–2057. 10.1111/bjet.13102 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]

DATA AVAILABILITY STATEMENT

• Aesaert, K. , Van Nijlen, D. , Vanderlinde, R. , & van Braak, J. (2014). Direct measures of digital information processing and communication skills in primary education: Using item response theory for the development and validation of an ICT competence scale . Computers & Education , 76 , 168–181. 10.1016/j.compedu.2014.03.013 [ CrossRef ] [ Google Scholar ]
• Agung, A. S. N. , Surtikanti, M. W. , & Quinones, C. A. (2020). Students’ perception of online learning during COVID‐19 pandemic: A case study on the English students of STKIP Pamane Talino . SOSHUM: Jurnal Sosial Dan Humaniora , 10 ( 2 ), 225–235. 10.31940/soshum.v10i2.1316 [ CrossRef ] [ Google Scholar ]
• Anderson, T. (2003). Getting the mix right again: An updated and theoretical rationale for interaction . The International Review of Research in Open and Distributed Learning , 4 ( 2 ). 10.19173/irrodl.v4i2.149 [ CrossRef ] [ Google Scholar ]
• Appana, S. (2008). A review of benefits and limitations of online learning in the context of the student, the instructor and the tenured faculty . International Journal on E‐learning , 7 ( 1 ), 5–22. [ Google Scholar ]
• Bączek, M. , Zagańczyk‐Bączek, M. , Szpringer, M. , Jaroszyński, A. , & Wożakowska‐Kapłon, B. (2021). Students’ perception of online learning during the COVID‐19 pandemic: A survey study of Polish medical students . Medicine , 100 ( 7 ), e24821. 10.1097/MD.0000000000024821 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Barbour, M. K. (2013). The landscape of k‐12 online learning: Examining what is known . Handbook of Distance Education , 3 , 574–593. [ Google Scholar ]
• Barbour, M. , Huerta, L. , & Miron, G. (2018). Virtual schools in the US: Case studies of policy, performance and research evidence. In Society for information technology & teacher education international conference (pp. 672–677). Association for the Advancement of Computing in Education (AACE). [ Google Scholar ]
• Barbour, M. K. , & LaBonte, R. (2017). State of the nation: K‐12 e‐learning in Canada, 2017 edition . http://k12sotn.ca/wp‐content/uploads/2018/02/StateNation17.pdf [ Google Scholar ]
• Barbour, M. K. , & Reeves, T. C. (2009). The reality of virtual schools: A review of the literature . Computers & Education , 52 ( 2 ), 402–416. [ Google Scholar ]
• Basuony, M. A. K. , EmadEldeen, R. , Farghaly, M. , El‐Bassiouny, N. , & Mohamed, E. K. A. (2020). The factors affecting student satisfaction with online education during the COVID‐19 pandemic: An empirical study of an emerging Muslim country . Journal of Islamic Marketing . 10.1108/JIMA-09-2020-0301 [ CrossRef ] [ Google Scholar ]
• Berge, Z. L. (2005). Virtual schools: Planning for success . Teachers College Press, Columbia University. [ Google Scholar ]
• Bickle, M. C. , & Rucker, R. (2018). Student‐to‐student interaction: Humanizing the online classroom using technology and group assignments . Quarterly Review of Distance Education , 19 ( 1 ), 1–56. [ Google Scholar ]
• Broadbent, J. , & Poon, W. L. (2015). Self‐regulated learning strategies & academic achievement in online higher education learning environments: A systematic review . The Internet and Higher Education , 27 , 1–13. [ Google Scholar ]
• Calderoni, J. (1998). Telesecundaria: Using TV to bring education to rural Mexico (Tech. Rep.). The World Bank. [ Google Scholar ]
• Cisco . (2018). Bandwidth requirements for meetings with cisco Webex and collaboration meeting rooms white paper . http://dwz.date/dpbc [ Google Scholar ]
• Cisco . (2019). Cisco digital readiness 2019 . https://www.cisco.com/c/m/en_us/about/corporate‐social‐responsibility/research‐resources/digital‐readiness‐index.html#/ (Library Catalog: www.cisco.com). [ Google Scholar ]
• Clark, C. , Strudler, N. , & Grove, K. (2015). Comparing asynchronous and synchronous video vs. text based discussions in an online teacher education course . Online Learning , 19 ( 3 ), 48–69. [ Google Scholar ]
• Claro, M. , Preiss, D. D. , San Martín, E. , Jara, I. , Hinostroza, J. E. , Valenzuela, S. , Cortes, F. , & Nussbaum, M. (2012). Assessment of 21st century ICT skills in Chile: Test design and results from high school level students . Computers & Education , 59 ( 3 ), 1042–1053. 10.1016/j.compedu.2012.04.004 [ CrossRef ] [ Google Scholar ]
• Cohen, J. (1988). Statistical power analysis for the behavioral sciences . Routledge Academic. [ Google Scholar ]
• Corbi, A. , & Solans, D. B. (2014). Review of current student‐monitoring techniques used in elearning‐focused recommender systems and learning analytics: The experience API & LIME model case study . IJIMAI , 2 ( 7 ), 44–52. [ Google Scholar ]
• Dabbagh, N. , & Kitsantas, A. (2012). Personal learning environments, social media, and self‐regulated learning: A natural formula for connecting formal and informal learning . The Internet and Higher Education , 15 ( 1 ), 3–8. 10.1016/j.iheduc.2011.06.002 [ CrossRef ] [ Google Scholar ]
• Garrison, D. R. , Cleveland‐Innes, M. , & Fung, T. S. (2010). Exploring causal relationships among teaching, cognitive and social presence: Student perceptions of the community of inquiry framework . The Internet and Higher Education , 13 ( 1–2 ), 31–36. 10.1016/j.iheduc.2009.10.002 [ CrossRef ] [ Google Scholar ]
• Gašević, D. , Adesope, O. , Joksimović, S. , & Kovanović, V. (2015). Externally‐facilitated regulation scaffolding and role assignment to develop cognitive presence in asynchronous online discussions . The Internet and Higher Education , 24 , 53–65. 10.1016/j.iheduc.2014.09.006 [ CrossRef ] [ Google Scholar ]
• Gašević, D. , Zouaq, A. , & Janzen, R. (2013). “Choose your classmates, your GPA is at stake!” The association of cross‐class social ties and academic performance . American Behavioral Scientist , 57 ( 10 ), 1460–1479. [ Google Scholar ]
• Gikas, J. , & Grant, M. M. (2013). Mobile computing devices in higher education: Student perspectives on learning with cellphones, smartphones & social media . The Internet and Higher Education , 19 , 18–26. [ Google Scholar ]
• Harvey, D. , Greer, D. , Basham, J. , & Hu, B. (2014). From the student perspective: Experiences of middle and high school students in online learning . American Journal of Distance Education , 28 ( 1 ), 14–26. 10.1080/08923647.2014.868739 [ CrossRef ] [ Google Scholar ]
• Kauffman, H. (2015). A review of predictive factors of student success in and satisfaction with online learning . Research in Learning Technology , 23 . 10.3402/rlt.v23.26507 [ CrossRef ] [ Google Scholar ]
• Kuo, Y.‐C. , Walker, A. E. , Belland, B. R. , Schroder, K. E. , & Kuo, Y.‐T. (2014). A case study of integrating interwise: Interaction, internet self‐efficacy, and satisfaction in synchronous online learning environments . International Review of Research in Open and Distributed Learning , 15 ( 1 ), 161–181. 10.19173/irrodl.v15i1.1664 [ CrossRef ] [ Google Scholar ]
• Lee, S. J. , Srinivasan, S. , Trail, T. , Lewis, D. , & Lopez, S. (2011). Examining the relationship among student perception of support, course satisfaction, and learning outcomes in online learning . The Internet and Higher Education , 14 ( 3 ), 158–163. 10.1016/j.iheduc.2011.04.001 [ CrossRef ] [ Google Scholar ]
• Liu, F. , & Cavanaugh, C. (2012). Factors influencing student academic performance in online high school algebra . Open Learning: The Journal of Open, Distance and e‐Learning , 27 ( 2 ), 149–167. 10.1080/02680513.2012.678613 [ CrossRef ] [ Google Scholar ]
• Lou, Y. , Bernard, R. M. , & Abrami, P. C. (2006). Media and pedagogy in undergraduate distance education: A theory‐based meta‐analysis of empirical literature . Educational Technology Research and Development , 54 ( 2 ), 141–176. 10.1007/s11423-006-8252-x [ CrossRef ] [ Google Scholar ]
• Malik, M. , & Fatima, G. (2017). E‐learning: Students’ perspectives about asynchronous and synchronous resources at higher education level . Bulletin of Education and Research , 39 ( 2 ), 183–195. [ Google Scholar ]
• McInnerney, J. M. , & Roberts, T. S. (2004). Online learning: Social interaction and the creation of a sense of community . Journal of Educational Technology & Society , 7 ( 3 ), 73–81. [ Google Scholar ]
• Molnar, A. , Miron, G. , Elgeberi, N. , Barbour, M. K. , Huerta, L. , Shafer, S. R. , & Rice, J. K. (2019). Virtual schools in the US 2019 . National Education Policy Center. [ Google Scholar ]
• Montague, M. , & Rinaldi, C. (2001). Classroom dynamics and children at risk: A followup . Learning Disability Quarterly , 24 ( 2 ), 75–83. [ Google Scholar ]
• Montag, C. , Becker, B. , & Gan, C. (2018). The multipurpose application Wechat: A review on recent research . Frontiers in Psychology , 9 , 2247. 10.3389/fpsyg.2018.02247 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Moore, M. G. (1989). Editorial: Three types of interaction . American Journal of Distance Education , 3 ( 2 ), 1–7. 10.1080/08923648909526659 [ CrossRef ] [ Google Scholar ]
• Muilenburg, L. Y. , & Berge, Z. L. (2005). Student barriers to online learning: A factor analytic study . Distance Education , 26 ( 1 ), 29–48. 10.1080/01587910500081269 [ CrossRef ] [ Google Scholar ]
• Muirhead, B. , & Juwah, C. (2004). Interactivity in computer‐mediated college and university education: A recent review of the literature . Journal of Educational Technology & Society , 7 ( 1 ), 12–20. [ Google Scholar ]
• Niemi, H. M. , & Kousa, P. (2020). A case study of students’ and teachers’ perceptions in a finnish high school during the COVID pandemic . International Journal of Technology in Education and Science , 4 ( 4 ), 352–369. 10.46328/ijtes.v4i4.167 [ CrossRef ] [ Google Scholar ]
• Oliver, K. , Osborne, J. , & Brady, K. (2009). What are secondary students’ expectations for teachers in virtual school environments? Distance Education , 30 ( 1 ), 23–45. 10.1080/01587910902845923 [ CrossRef ] [ Google Scholar ]
• Pardo, A. , Jovanovic, J. , Dawson, S. , Gašević, D. , & Mirriahi, N. (2019). Using learning analytics to scale the provision of personalised feedback . British Journal of Educational Technology , 50 ( 1 ), 128–138. 10.1111/bjet.12592 [ CrossRef ] [ Google Scholar ]
• Ramij, M. , & Sultana, A. (2020). Preparedness of online classes in developing countries amid covid‐19 outbreak: A perspective from Bangladesh. Afrin, Preparedness of Online Classes in Developing Countries amid COVID‐19 Outbreak: A Perspective from Bangladesh (June 29, 2020) .
• Rice, K. L. (2006). A comprehensive look at distance education in the k–12 context . Journal of Research on Technology in Education , 38 ( 4 ), 425–448. 10.1080/15391523.2006.10782468 [ CrossRef ] [ Google Scholar ]
• Shishehchi, S. , Banihashem, S. Y. , & Zin, N. A. M. (2010). A proposed semantic recommendation system for elearning: A rule and ontology based e‐learning recommendation system. In 2010 international symposium on information technology (Vol. 1, pp. 1–5).
• Song, L. , Singleton, E. S. , Hill, J. R. , & Koh, M. H. (2004). Improving online learning: Student perceptions of useful and challenging characteristics . The Internet and Higher Education , 7 ( 1 ), 59–70. 10.1016/j.iheduc.2003.11.003 [ CrossRef ] [ Google Scholar ]
• Spitzer, D. R. (2001). Don’t forget the high‐touch with the high‐tech in distance learning . Educational Technology , 41 ( 2 ), 51–55. [ Google Scholar ]
• Thomas, R. M. (2000). Comparing theories of child development. Wadsworth/Thomson Learning. United Nations Educational, Scientific and Cultural Organization. (2020, March). Education: From disruption to recovery . https://en.unesco.org/covid19/educationresponse (Library Catalog: en.unesco.org)
• Uttal, D. H. , & Cohen, C. A. (2012). Spatial thinking and stem education: When, why, and how? In Psychology of learning and motivation (Vol. 57 , pp. 147–181). Elsevier. [ Google Scholar ]
• Van Lancker, W. , & Parolin, Z. (2020). Covid‐19, school closures, and child poverty: A social crisis in the making . The Lancet Public Health , 5 ( 5 ), e243–e244. 10.1016/S2468-2667(20)30084-0 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Wang, C.‐H. , Shannon, D. M. , & Ross, M. E. (2013). Students’ characteristics, self‐regulated learning, technology self‐efficacy, and course outcomes in online learning . Distance Education , 34 ( 3 ), 302–323. 10.1080/01587919.2013.835779 [ CrossRef ] [ Google Scholar ]
• Wang, J. , & Antonenko, P. D. (2017). Instructor presence in instructional video: Effects on visual attention, recall, and perceived learning . Computers in Human Behavior , 71 , 79–89. 10.1016/j.chb.2017.01.049 [ CrossRef ] [ Google Scholar ]
• Wilkinson, A. , Roberts, J. , & While, A. E. (2010). Construction of an instrument to measure student information and communication technology skills, experience and attitudes to e‐learning . Computers in Human Behavior , 26 ( 6 ), 1369–1376. 10.1016/j.chb.2010.04.010 [ CrossRef ] [ Google Scholar ]
• World Health Organization . (2020, July). Coronavirus disease 2019 (COVID‐19): Situation Report‐164 (Situation Report No. 164). https://www.who.int/docs/default‐source/coronaviruse/situation‐reports/20200702‐covid‐19‐sitrep‐164.pdf?sfvrsn$=$ac074f58$_$2
• Yates, A. , Starkey, L. , Egerton, B. , & Flueggen, F. (2020). High school students’ experience of online learning during Covid‐19: The influence of technology and pedagogy . Technology, Pedagogy and Education , 9 , 1–15. 10.1080/1475939X.2020.1854337 [ CrossRef ] [ Google Scholar ]

• Open access
• Published: 10 March 2022

Collaborative learning in small groups in an online course – a case study

• Mildrid Jorunn Haugland 1 ,
• Ivar Rosenberg 2 &
• Katrine Aasekjær 3  

BMC Medical Education volume  22 , Article number:  165 ( 2022 ) Cite this article

5031 Accesses

6 Citations

Metrics details

The ability to learn collaboratively and work in teams is an essential competency in both educational and healthcare settings, and collaborative student activities are acknowledged as being an important part of the pedagogical approach in higher education and teaching. The course that was the focus of this research, a 15-ECTS-credit online course in philosophy of science, ethics, and research methods, was offered online as part of 11 master’s-level health programmes at a university in Norway. Collaborative learning in combination with digital teaching tools was the preferred pedagogical approach in the online course. The aim of the study was to describe, explore and discuss how the students collaborated in small groups in an online course to learn.

We performed six focus groups and 13 individual interviews from February 2018 to May 2019, conducting a qualitative case study with a content analysis of the data collected. The participants were master students in the same faculty at a university in Norway. All the included participants had fulfilled the 15 ECTS credit course.

Our study revealed that the collaboration in small groups resulted in three different working processes, depending on the students’ ability to be flexible and take responsibility for their own and common learning. The three different working processes that emerged from our data were 1. joint responsibility – flexible organization; 2. individual responsibility – flexible organization; and 3. individual responsibility – unorganized. None of the groups changed their working process during their course, even though some experienced their strategy as inadequate.

Conclusions

Our study showed that despite similar factors such as context, assignments and student autonomy, the students chose different collaboration strategies to accomplish the online course learning objectives. Each group chose their own working process, but only the strategy 1. joint responsibility – flexible organization seemed to promote collaboration, discussion, and team work to complete the complex assignments in the online course. The result from our study may be helpful in designing and planning future online courses; hence online learning requires a focus on how students collaborate and learn online, to gain knowledge and understanding through group discussion.

Peer Review reports

Education is increasingly being offered online, and there is growing demand in higher education for online studies and courses using online resources in teaching and learning [ 1 , 2 ]. E-learning worldwide is expected to account for 30% of all educational provision [ 3 , 4 ]. This has led to an increase in educational provision offered online (all or in part) and the need for improved articulation between technology and pedagogy in higher education [ 5 ]. However, even though online teaching is in demand in both educational institutions and among students themselves, studies show that the ability to complete online education is reduced, compared to face-to-face teaching [ 6 , 7 , 8 ]. One suggestion for reducing dropout rates is to have a mix of online and face-to-face courses in study programmes [ 9 ].

Collaborative learning (CL) and teamwork skills developed through working in groups are important competencies for healthcare workers [ 10 ]. Group work is therefore a pedagogical method that is widely used in health and social science education (e.g., problem-based learning (PBL), team-based learning (TBL) and simulation training), and much research has been carried out into these pedagogical methods [ 11 ]. Collaborative learning is defined as teaching or learning activities that promote an individual’s own learning and that of others in small groups (two to five students) or collaboration (cooperation) to achieve common goals [ 12 , 13 ]. In CL, learning is a dynamic process involving interaction between individual students’ drive to learn and a social activity in a specific context [ 14 , 15 , 16 ]. While students in CL are mutually dependent on one another, to be able to discuss and reflect and thus achieve a deeper understanding of the subject matter, they learn from one another through reflection in the situation and on the situation [ 17 , 18 , 19 , 20 ].

Research has shown which design and group working process factors that can positively influence collaboration within groups. Design factors include group size (three–five), group composition and the nature of the assignments [ 20 ]. Positive interdependence and individual accountability are important factors for group working processes [ 20 , 21 ]. In Norway, two reports [ 22 , 23 ] have concluded that students’ learning depends on how digital tools are implemented and how these tools are used within the pedagogical situation. It is therefore important to look at which factors are important for CL to achieve discussion and reflection, thereby facilitating in-depth learning and collaboration, when students complete assignments. An understanding of the factors that facilitate students’ collaboration is critical to understanding how this approach to learning can be used more effectively in online courses in higher education.

The aim of the study was to describe, explore, and discuss how students undertaking an online course collaborated in small groups. Our research question was therefore: How did the students collaborate in small groups to achieve learning online?

We adopted a social-constructivist approach to learning in this study. This approach emphasizes that understanding CL and the various roles students have in the learning process requires examining the interaction that is taking place and the context of this interaction [ 16 , 24 , 25 , 26 , 27 ]. In this approach learning is seen as a dynamic social process where increased knowledge is considered a consequence of social interaction.

When the study was conducted, all master’s students at the Western Norway University of Applied Sciences, Faculty of Health and Social Sciences, had completed an online course in the philosophy of science, ethics, and research methods (MaMet).

Each programme of study created in MaMet its own assignments, making them relevant to their specific professions/programmes and ensuring an appropriate level of difficulty for all students. The courses were run at different time periods for each programme and by teacher associated with that specific programme. In some programmes, this was the only online course, others programmes had several online courses. A key learning method in this course was CL in small groups, through written assignments and peer reviewing of fellow students’ assignments.

Description of MaMet

The online course in philosophy of science, ethics, and research methods was completed in 11 master’s programmes, and each study programme was responsible for administering the course by facilitating, guiding and following the students over the course of this module in their master’s programme. The number of students enrolled in each programme differed, ranging from 12 to 90. MaMet is grounded on a small-scale online course, where there is planned discussion and feedback among teachers and students throughout the whole course. Student activity and CL are a cornerstone in the pedagogical and didactical thinking in MaMet, and work in small groups, with assignments, is the most prominent methodology in implementation of the course. By completing every problem-based assignment in the course, students gain the knowledge and skills to be able to design their own study protocol. The digital resources in MaMet include design focusing on learning outcomes, enabling students to develop the ability to understand and perform research projects.

Each small group in the study consisted of the same students throughout the whole course and included three to five students. Group members were responsible for the work processes of the group, how task problems were solved, when and how the group members met and the collaborative structure within the group. Group composition and the content of assignments were defined by the master’s programme. The groups did not have supervisors for small-group collaboration.

This case study involved individual and focus group interviews with master’s students in the Faculty of Health and Social Sciences, to gain feedback about their experience of collaborative online learning in small groups in an online course. By conducting a qualitative case study, we were able to generate an in-depth understanding of a complex issue in a real-life context [ 28 , 29 ].

The study was approved by the Norwegian Centre for Research Data (NSD 60336) and the academic institution. All the participants gave written consent after receiving written and oral information about the study and were given the option to withdraw from the study if they wanted to. All data were anonymized and kept confidential, in compliance with the ethical guidelines of the Declaration of Helsinki.

We conducted both focus-group and individual interviews because some participants felt that taking part in focus groups with their fellow students would be difficult as their experiences were connected to collaboration with peers. Differentiation of focus-group or individual interviews was done in collaboration with the participants, and with a focus on including students from all the master’s programmes who had completed the online course in philosophy of science, ethics, and research methods. Given that the participants completed the 15-ECTS programme at different points during the year, it was not possible to combine students from different programmes in the focus group interview.

Participants and settings

Participants for the individual and focus group interviews were selected by means of purposeful sampling. This sampling method enabled us to include participants who could contribute information relevant to the aim of the study [ 28 , 29 ]. Participants were master’s students who had participated in the 15-ECTS-credit online course in philosophy of science, ethics, and research methods. A total of 260 students completed the course, split into 65 small groups, and these students received a written invitation (via the online course) to participate in the study and an oral invitation (when attending lectures on campus). Thirty students from all 11 master’s programmes expressed an interest in taking part, and all 30 were included. We conducted six focus group meetings and 13 individual interviews. The 30 participants represented 25 different groups. Two of 30 participants were males, which reflects the overall gender distribution in the programmes. Each focus group consisted of between two and six students.

Data collection

The same person (MJH) moderated all individual and focus group interviews, and IR co-moderated the focus group interviews conducted between February 2018 and May 2019. MJH had not been involved in development of the online course and had never met the students before. The focus group interviews were conducted face to face. Some individual interviews were conducted over the phone if that best suited the students. We conducted focus group interviews since our understanding is that students will be influenced by and have an influence on others present, providing a collection of rich and meaningful data [ 30 , 31 ]. We believe that interaction through focus groups can inspire students to reflect and talk about the challenges associated with the topic. To facilitate this interaction, we conducted the focus group interviews in settings free of disturbance, on campus. The interviewer made all the arrangements regarding time and place, in agreement with the students.

The data collection was conducted within 2 months of the course ending. In this way, the students would still be able to remember their experiences while, at the same time, having a certain amount of distance from them. The themes in the semi-structured interview guide were as follows: a typical day (what activities/events took place and when); use of resources; motivation for online learning; collaboration in students’ respective small groups; and collaboration with teachers and fellow students. If any theme was not mentioned by the students during the conversation, the interviewer asked about it. The interviews lasted from 30 to 75 min, until data saturation occurred [ 31 , 32 ]. The interviews were recorded and transcribed verbatim by MJH and KAA and approved by all the authors, ensuring that no essential information was lost during the transcription process.

Data analysis

We used content analysis to analyse the data [ 30 , 33 ], starting with all three authors reading and re-reading the interviews to get an overall impression of the data. Two of the researchers (MJH and KAA) worked separately and divided the text into units of meaning. They then grouped and coded these. This was done for each interview. We compared and discussed the codes across the interviews, before identifying categories. All interviews were then analysed again, with a focus on codes extracted to form categories. At the end of the analysis process, we created a condensed narrative with quotes, to illustrate what appeared in the categories. Relevant subthemes were identified to highlight key similarities and differences in the three main themes, based on what we found in the data. All three researchers discussed the sub-categories and further abstracted and reorganized these into themes and subthemes. Table  1 gives examples of the abstraction process from meaning units, code, and themes to subthemes. The condensed narrative formed the basis for the results presented.

Our analytical process revealed themes and subthemes underpinning experiences essential to understanding how students collaborated in small groups in an online course. We found that when students were collaborating online in a group, the groups developed different strategies to solve the course assignments. All groups had the same goal for their work but used different working processes to reach that goal. Figure  1 summarizes the three different working processes that emerged: 1. joint responsibility – flexible organization; 2. individual responsibility – flexible organization; and 3. individual responsibility – unorganized.

An overview of how the various groups organized their work

The different working processes reflected the main characteristics of the group. We found seven subthemes that characterized the work process: understanding of the tasks, expectation of the group members, responsibility for the group work, preparedness for the group meetings, organization of the group work, group loyalty, and responsibility for fellow students’ learning. Each group seemed to maintain its working process throughout the online course, even if students told us that they experienced, as the work progressed, that there could be other and more appropriate ways of collaborating to complete the various assignments. The students’ explanation for not attempting to change their working process, was that they wanted to avoid conflicts and damage the atmosphere in the group. There is no point in complaining, it will not solve the problem, only cause unpleasant feelings.” Interview 13.

Regardless of which working processes the students engaged in, the students reported that group assignments were important for learning philosophy and methods relevant to science, and that problem-based assignments enabled them to use all the learning resources in the online course as the assignments were so closely linked to the learning resources. Furthermore, the students stated that it was important to continue the collaboration with the same students throughout the whole course. The topics were complex and difficult to understand, and by having the same group members, the working process was more predictable. The students also thought that the group sizes were appropriate.

Joint responsibility – flexible organization

These students reported that they were prepared and informed about the assignments ahead of them. The work was characterized by joint responsibility and had a clear structure and framework to promote collaboration. The structure was such that there was still some flexibility, and it could be adjusted during the working process to suit the needs of the groups. There was a loyalty in relation to the work process to be carried out, and group members’ level of participation was high. This working process was characterized by discussion and reflection on the assignments and an understanding that learning was promoted through the group working process. The groups worked independently, with less need for input from the teachers.

“By collaborating online, we had to have strict rules within the group, so we didn’t spend time interrupting each other. We met on a regular basis since none of the group members lived in the same city, and it was great starting at times that matched our schedule. It made the work flexible. We solved all the assignments in collaboration, and all the members had a common responsibility for the assignments.” Interview 8.

We found that with this model (joint responsibility – flexible organization), the groups collaborated in different ways to solve the assigned problem. In some of the groups, the students started the work together, distributing the work among group members, and then came together to discuss progress, distributing further work within the group as required. The work was characterized by short meetings to clarify a common understanding and to distribute responsibilities. Other groups spent a long time on the working processes and did all the work together online, focusing on common understanding.

“We met every day at 10. Sometimes we discussed in a videoconference, sometimes just chatting, or emailing. We divided the task and had individual responsibility. But at the same tame we gave both written and orally feedback on fellow students work. So, we had an individual part, but were involved in the whole task” Interview 13

Despite some differences in working processes, the main characteristic of collaboration was that the groups had a mutual aim (i.e., that all students in the group had a common understanding of the task in hand and respective contributions), and they consequently took responsibility for their own and fellow students’ learning.

Individual responsibility – flexible organization

In this working process, the rationale for the online course and how the assignments would contribute to the students becoming qualified professionals was less clear to the students. Consequently, the group members had varying degrees of preparedness, and use of the digital learning resources was more fragmented and limited to the assignments that the group had to complete.

“I don’t understand why I need philosophy and method in my profession. I am not going to be a scientist, I am supposed to be [an anaesthetic nurse, operating room nurse, intensive care nurse]. I really don’t see the point. I believe only those who actually want to do a master thesis should do this.” Interview 7.

When these students met for the first time, they were not prepared. They organized the work by dividing the assignments into smaller sections, and they completed different sections of the overall assignment separately, bringing their respective contributions together to assemble an answer at the end. In this collaboration process, work was also distributed within groups differently. Assignments were tackled either by distributing an entire task to each of the members or by dividing the relevant task into smaller units so that each student contributed to each task. In both approaches, individuals had responsibility for completing part of the overall task.

“We organized the group by delegating one assignment to two group members at a time. Meaning that two students had the main responsibility for one assignment, and the other members gave some comments on the work. This gave us a greater flexibility and not so much work. But I don’t have so much knowledge and control over the themes that I didn’t have responsibility over.” Interview 10.

In this working model (individual responsibility – flexible organization), there was less focus on meeting one another to discuss the work in progress, but rather, a greater focus on delivering a product – the assignment. Discussion about the product was characterized by whether the assignment contained what it needed for it to be approved, and there was less discussion of the group members’ understanding of the task. Group members’ input tended to be presented to the group individually, rather than during group discussion, which could result in a situation with two conflicting inputs. It was thus up to the person responsible for delivering the assignment to assess which input should be considered or whether the input should be considered at all.

“I wanted the answer to be the best possible. Everyone did their part and pasted it into the document. And then almost nothing happened. We were left with a fragmented answer with many yellow boxes and comments on the page. I had to take responsibility for the last bit to put [it all] together and make it a whole. Students stopped contributing when they felt they had finished with their part Often I had the impression that they did not care and that the focus was elsewhere.” Interview 12

Individual responsibility – unorganized

This working process had no group structure because how the work was organized depended on individuals taking responsibility on behalf of the group. The main characteristic of this working process was that the student(s) who took responsibility were the same throughout the online course, and these students were highly motivated. They recognized the importance of learning and therefore knew how to go about solving the assignment. This working process lacked structure and organization, and there was an absence of cooperation and discussion. Most of the group members gave their input only when the product was available.

“It became an extra workload on my account, because I felt that I had to do the assignments so that we could deliver a product. You depend on those who are supposed to participate to take responsibility for their own learning. There is no point in complaining, it will not solve the problem, only cause unpleasant feelings.” Interview 13.

In this working process, one or two students in each group were responsible for the work. Group members who did not take any responsibility only attended group meetings when these did not conflict with their own needs or priorities. Group dynamics existed to a limited extent because the groups were characterized by an absence of participation, discussion and flexibility among group members. Those individuals who took responsibility had little opportunity to assume responsibility for anything other than their own learning and understanding, but they expressed the desire for a community where they could share knowledge through discussion and reflection and thus enable everyone to gain greater knowledge.

“It gave an unpleasant feeling – the responsibility of doing the assignments on your own. I didn’t feel we were a group who could share and help each other. We didn’t share common responsibility, and therefore we lost the possibility for discussions and reflection. I believe we could have had more; we could have achieved better learning if we had worked differently.” Interview 14.

The aim of the study was to describe, explore, and discuss how students undertaking an online course collaborated in small groups. Overall, students reported a positive experience of studying the philosophy of science and methods in an online context, and most of the students reported that working in small groups was essential for learning complex aspects of this subject. The participants in our study reported that group size (three–five) was a significant characteristic for the work process, along with a reciprocal blend of digital resources and assignments. These characteristics and continuity of group members throughout the online course made it possible to complete and deliver complex assignments. Our findings are in line with factors identified by Scager [ 20 ] as being important when working together. By working in small groups, the students experienced a level of support and understanding among their fellow students, and the fact that all assignments in the online course were group-based forced them to collaborate to achieve the learning outcomes.

Even though the students thought that CL was essential for their learning, not all reflected on the relationship between the working processes within the group and their learning. We found that the different working processes adopted during the online course could be differentiated into three main group working processes. These processes were not all focused on collaborative learning. Rather, some focused more on the students’ own learning and competencies. These findings are in line with those of Johnson et al. [ 13 ], who also found variations in small-group working. The three main working processes were consistent throughout the whole online course. Although some students felt that the work process could have been better, and that they therefore had to do additional work, they resisted changing the process in fear of ruining the atmosphere in the group.

Among the groups using working process number 1, two approaches to the organization of work could be observed, meaning that the students often had meetings, but how they met and collaborated was different. Some groups met and distributed the work, clarifying assignments, discussing different opinions and interpretations, and coming to an agreement and common understanding. They worked on their specific set of tasks and then met again to finalize the work and complete the assignment. The other approach was to have a discussion and work together towards completion of the assignment, changing the campus group to an online group, focusing on collaborating in real time. Students in groups using both forms worked together and discussed all parts of the tasks as a team to complete different aspects of the complex assignments As some students expressed; “I experienced that I learned more by participating in discussions with my peers and solving tasks together than spending time alone with my books” Interview 13. They involved themselves and their fellow students and were committed to working as a team and to the subject matter [ 20 , 34 ].

In working process number 2 (individual responsibility, with flexible organization), students acted more as individual contributors than as team members. The students’ started their collaboration by dividing the assignment into different sub-assignments, with students taking individual responsibility for their task. Groups used a “stapler” (as described by Scager [ 20 ]), i.e., a group member who was responsible for integrating each student’s work into a group paper. The groups did not seek to establish common knowledge or a shared understanding of the topic, and each student had individual responsibility for seeking out the necessary knowledge to complete his or her contribution to the assignment. Due to a lack of continuity in their interaction and collaboration, these groups might have lost the potential learning effect of collaboration. Johnson and Johnson [ 35 ] have called this behaviour “pseudo learning”. Although a sense of team cohesiveness is maintained through equal contributions from each member and by agreeing on distribution of the workload, this method of organizing work does not ensure that students perceive their work as an activity which facilitates learning; rather, they see it more as a way to “get the job done”.

Students adopting working process number 3 (individual responsibility, with unorganized groups) were organized as a group but did not act as a group, and the groups did not organize themselves. Only one or a few students took responsibility and got involved in the work, while the other group members did not participate in small-group collaboration. The students who took responsibility worked and collaborated in many of the ways that the students in working process 1 did, taking responsibility, using available resources and completing the assignments. As stated in one of the interviews – “we were two who took responsibility, and we sent documents and discussions back and forth between the two of us. That worked very well, but the rest of the group were absent, and that felt wrong” Interview 11.

Bliss and Lawrence [ 36 ] claim that one of the biggest obstacles to group learning is students who do not participate. Our findings demonstrate that this can indeed be a problem. The students who became involved were deprived of the benefits that could be achieved through discussion. This has also been reported by Bliss and Lawrence [ 36 ] and Liu and Tsai [ 37 ]. On the other hand, these students became well acquainted with the subject matter and were able to complete the assignments.

Our findings suggest that how students perceive the subject they are studying is related to the importance of the subject. Another key factor affecting adoption of collaborative work practices is having a common understanding of the subject and students’ expectations regarding their own level of participation in the subject. The students who organized their work using working process number 1 had a common understanding of the collaboration within the group. This made the members aware of what to do and their expectations of one another. It is uncertain whether all members fully recognized the meaning of the objectives of the online course, but how the group organized the work could have led to a common understanding. There was a random composition of the groups, and how all members within one group perceived the objectives of the course is uncertain. It may have been the case that members in groups focusing on individual responsibility could initially have had the same opinions as those working with joint responsibility. It seems that how the group organized (or did not organize) the work affected the students’ understanding of the course in philosophy and science, and this understanding could have changed because of the different working processes. Lave and Wenger [ 26 ] stated that active participation in the social context creates what they do as a group. It is more likely that the working processes arose due to a lack of understanding of the consequences of different group working processes for both individual members’ and fellow students’ learning, and the strategies used by the group at the outset persisted throughout the whole online course. The students who used working process number 3 had also not clarified the working process in advance, and this way of organizing the group seemed to leave the responsibility to a few members and not the entire group.

All three working processes identified seem to be in line with the definition of collaborative learning that emphasizes collaboration (cooperation) to achieve common goals [ 12 , 13 ]. It is uncertain whether all working processes facilitate an individual’s own learning and that of others in small groups equally well. When learning is understood as a dynamic social process where knowledge is considered a response to social interaction, and a prerequisite is discussion and reflection on everyone’s contributions, it seems that not all of the three working processes identified in our study can be understood as CL.

The working process joint responsibility and flexible organization encouraged discussion and reflection, with all group members developing an understanding of the assignments. Many studies have found that in-depth learning is achieved through discussion and reflection with peers [ 17 , 18 , 19 , 20 , 21 , 22 ], and we found that students working with joint responsibility acted in line with this definition of in-depth learning. In this working process, the students had shared responsibility for assignments and a common understanding of the work, with all participants acquiring greater knowledge and understanding of all parts of the tasks. Through sharing responsibility, group discussions and feedback, the students not only acted as a group but more like a team. Bang and Midelfart [ 38 ] define a team as a group that has a task for which the group members are collectively responsible and where they are interdependent.

The three different collaboration processes adopted during the online course that we studied do not differ significantly from how students collaborate in in-person courses as shown in the literature [ 13 , 19 , 20 , 21 , 22 ]. It seems that students adopt the same patterns and structures independently from the context they operate in. This could indicate that students collaborate and are influenced by the same factors regardless of where the collaboration takes place.

Strengths and limitations

This case study gave us the opportunity to explore, in depth and over time, students’ experience of collaborative learning in small groups in online courses. Since there is little research (to the best of our knowledge) on how students learn in online subjects that are part of an otherwise campus-based education, it was important for us to gain an insight into different approaches to collaboration. This study was based on semi-structured interviews with students in relation to one case. We have described the case thoroughly so that readers can understand and recognize the parameters and relate them to their own situation [ 20 , 39 ]. In this way, the findings may be useful in designing and implementing similar online courses. The small number of students in some of the focus groups and the fact that the students came from the same programme could be a limitation of the study. Students from the same programme may have been too self-conscious to reveal certain aspects of group work in front of their peers. Some students mentioned this, so we conducted individual interviews. However, there may have been students who also felt the same without saying so. Having only two to five students in each group did not enable the full benefits of the group processes to be revealed in such groups [ 32 ]. Author IR was present during the focus groups interviews as a co-moderator. If the students knew that he was one of the course designers and the person who solved their technical problems this may have influenced the participants’ responses.

Another strength, yet also a possible weakness, of the study is that KAA and IR were the ones who developed the online course. This gave them an understanding of the challenges and strengths of the course, but at the same time, they had to work to maintain an analytical distance from the data. The third researcher was not directly involved in development of the course and could therefore view the data impartially.

This study contributes to knowledge of how students working in groups approach learning and identifies important factors about collaborative learning during online courses. This knowledge may be useful for educators designing and facilitating online courses and for instructors supervising groups. This study shows that even if design factors are the same (e.g., group size, challenging and relevant assignments, and student autonomy in terms of being able to organize group working processes), the working process that each group chooses can differ.

Although the identified working processes were found to promote collaboration only one working process promoted group discussion of all parts of the tasks, working as a team completing different aspects of complex assignments. Future online teaching might require an even stronger focus on students’ internal motivation for learning and the importance of teacher presence and teachers’ ability to facilitate online education.

Availability of data and materials

The datasets generated and/or analysed during the current study are not publicly available due to the confidentiality of the participants but are available from the corresponding author on reasonable request and with permission of the participants.

Abbreviations

• Collaborative learning

Online course in the philosophy of science and methods

European Credit Transfer and Accumulation System

Jara M, Mohamad F. Pedagogical templates for e-learning. Leading education and social research. London: Institute of Education, University of London; 2007.

Google Scholar  

Saqr M, Fors U, Tedre M. How the study of online collaborative learning can guide teachers and predict students’ performance in a medical course. BMC Med Educ. 2018;18(1):24.

Article   Google Scholar  

Lillejord S, Børte K, Nesje K, Ruud E. Learning and teaching with technology in higher education – a systematic review. Oslo: Knowledge Centre for Education; 2018. www.kunnskapssenter.no . 

European Commission. Report to the EU Commission on new modes of learning and teaching in higher education. 2014. http://ec.europa.eu/dgs/education_culture/repository/education/library/reports/modernisation-universities_en.pdf .

Jones A, Bennett R. Reaching beyond an online/offline divide: invoking the rhizome in higher education course design. Technol Pedagog Educ. 2017;26(2):193–210.

Allen IE, Seaman J. Changing course: ten years of tracking online education in the United States. Boston: Sloan Consortium (NJ1); 2013.

Ali R, Leeds E. The impact of face-to-face orientation on online retention: a pilot study. Online J Distance Learn Admin. 2009;12(4):1–10.

Bawa P. Retention in online courses: exploring issues and solutions—a literature review. SAGE Open. 2016;6:1–11.

Nilson L, Goodson L. Online teaching at its best. Merging Instructional Design with Teaching and Learning Research. San Fransico: Jossey-Bass; 2018.

Iqbal MP, Velan GM, O’Sullivan AJ, Balasooriya C. The collaborative learning development exercise (CLeD-EX): an educational instrument to promote key collaborative learning behaviours in medical students. BMC Med Educ. 2020;20:62.

Wilson KJ, Brickmann P, Bramie CJ, Group work. Evidence-based teaching guides. CBE Life Sci Educ. 2018;17(1):1–5.

Le H, Janssen J, Wubbels T. Collaborative learning practices: teacher and student perceived obstacles to effective student collaboration. Camb J Educ. 2018;48(1):103–22.

Johnson DW, Johnson RT, Smith KA. Cooperative learning: Improving university instruction by basing practice on validated theory. J Excell Coll Teach. 2014;25(3&4):85–118.

Cacciattolo K. Defining workplace learning. Eur Sci J. 2015;1:243–50.

Illeris K. A comprehensive understanding of human learning. In: Illeris K, editor. Contemporary Theories of Learning. Learning Theorists. In Their Own Words. 2nd ed. New York: Routledge; 2018. p. 8–22.

Chapter   Google Scholar  

Patton N, Higgs J, Smith M. Using theories of learning in workplaces to enhance physiotherapy clinical education. Physiother Theory Pract. 2013;29:493–503.

Billett S, Choy S. Learning through work: emerging perspectives and new challenges. J Work Learn. 2013;25:264–76.

Harasim L. Learning theory and online technologies. Taylor & Francis eBooks. 2nd ed. New York: Routledge; 2017.

Book   Google Scholar  

Laal M, Laal M. Collaborative learning: what is it? Proc Soc Behav Sci. 2012;31:491–5.

Scager K, Boonstra J, Peeters T, Vulperhorst J, Wiegant F. Collaborative learning in higher education: evoking positive interdependence. CBE Life Sci Educ. 2016;15(4):ar69.

Johnson DW, Johnson RT. An Educational Psychology success story: social interdependence theory and cooperative learning. Educ Res. 2009;38(5):365–79.

Damşa C, de Lange T, Elken M, Esterhazy R, Fossland T, Frølich N, et al. Quality in Norwegian higher education: a review of research on aspects affecting student learning. NIFU report. Oslo: NIFU/UiO; 2015. p. 24.

Nerland M, Prøitz TS. Pathways to quality in higher education: case studies of educational practices in eight courses. NIFU Rep. 2018;3(35):1–38.

Coster S, Norman I, Murrells T, Kitchen S, Meerabeau E, Sooboodoo E, et al. Interprofessional attitudes amongst undergraduate students in the health professions: a longitudinal questionnaire survey. Int J Nurs Stud. 2008;45:1667–81.

Hutchings M, Scammel J, Quinney A. Praxis and reflexivity for interprofessional education: towards an inclusive theoretical framework for learning. J Interprof Care. 2013;27:358–66.

Lave J, Wenger E. Situated learning. Legitimate peripheral participation. Cambridge: Cambridge University Press; 1991.

Haugland M, Brenna SJ, Aanes MM. Interprofessional education as a contributor to professional and interprofessional identities. J Interprof Care. 2019;12:1–7.

Crowe S, Cresswell K, Robertson A, Huby G, Avery A, Sheikh A. The case study approach. BMC Med Res Methodol. 2011;11:100.

Creswell JW, Creswell JD. Research design: qualitative, quantitative, and mixed methods approaches. Los Angeles: Sage; 2018.

Lindgren BM, Lundman B, Graneheim UH. Abstraction and interpretation during the qualitative content analysis process. Int J Nurs Stud. 2020;108:103632.

Patton MQ. Qualitative research & evaluation methods. 4th ed. Thousand Oaks: Sage Publications; 2015.

Krueger RA, Casey MA. A practical guide for applied research. California: Sage Publications Inc.; 2000.

Graneheim UH, Lundman B. Qualitative content analysis in nursing research: concepts, procedures and measures to achieve trustworthiness. Nurse Educ Today. 2004;24(2):105–12.

Pelz B. Three principles of effective online pedagogy. J Asynchron Learn Netw. 2004;14(1):103–16.

Johnson DW, Johnson RT. Learning together and alone: cooperative, competitive, and individualistic learning. 5th ed. Boston: Allyn & Bacon; 1999.

Bliss C, Lawrence B. From posts to patterns: a metric to characterize discussion board activity in online courses. Online Learn. 2019;13(2):1–18.

Liu CC, Tsai CC. An analysis of peer interaction patterns as discoursed by on-line small group problem-solving activity. Comput Educ. 2008;50:627–38.

Bang H, Midelfart TN. Effektive ledergrupper 2.utg. Oslo: Gyldendal; 2019.

Karlgren K, Larsson F, Dahlstrøm A. Eye-opening facilitator behaviours: an interaction analysis of facilitator behaviours that advance debriefings. BMJ Simul Technol Enhanc Learn. 2020;6:220–8.

Download references

Acknowledgements

We are very grateful to all the students who shared their experiences with us.

The Faculty of Health and Social Sciences, Western Norway University of Applied Sciences, Norway, funded working hours for MJH, IR and KAA.

Author information

Authors and affiliations.

Faculty of Health and Social Sciences, Department of Health and Functioning, Western Norway University of Applied Sciences/Høgskulen på Vestlandet, Inndalsveien 28, 5063, Bergen, Norway

Mildrid Jorunn Haugland

Faculty of Health and Social Sciences, Academic Affairs, Western Norway University of Applied Sciences/Høgskulen på Vestlandet, Inndalsveien 28, 5063, Bergen, Norway

Ivar Rosenberg

Faculty of Health and Social Sciences, Department of Health and Caring Sciences, Western Norway University of Applied Sciences/Høgskulen på Vestlandet, Inndalsveien 28, 5063, Bergen, Norway

Katrine Aasekjær

You can also search for this author in PubMed   Google Scholar

Contributions

The manuscript was read and approved by all the named authors. All the authors contributed to the design, data collection, analysis, writing, revision and approval of the manuscript. MJH moderated all individual and focus group interviews. IR co-moderated all the focus group interviews. MJH and KAA divided the text into units of meaning and grouped and coded these. KAA created a condensed narrative with quotes, to illustrate what appeared in the categories. All three researchers discussed the sub-categories and further abstracted and reorganized these into themes and subthemes. All authors reviewed the manuscript. We further confirm that the order of authors listed in the manuscript has been approved by all of us.

Corresponding author

Correspondence to Mildrid Jorunn Haugland .

Ethics declarations

Ethics approval and consent to participate.

The study was approved by the Norwegian Centre for Research Data (NSD-60336) and the academic institution. All the participants gave written consent after receiving written and oral information about the study and were given the option to withdraw from the study if they wanted to. All data were anonymized and kept confidential, in compliance with the ethical guidelines of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Cite this article.

Haugland, M.J., Rosenberg, I. & Aasekjær, K. Collaborative learning in small groups in an online course – a case study. BMC Med Educ 22 , 165 (2022). https://doi.org/10.1186/s12909-022-03232-x

Download citation

Received : 27 September 2021

Accepted : 03 March 2022

Published : 10 March 2022

DOI : https://doi.org/10.1186/s12909-022-03232-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Small group
• Online course
• Qualitative study
• Working process

BMC Medical Education

ISSN: 1472-6920

• Submission enquiries: [email protected]
• General enquiries: [email protected]

• Our Mission

Making Learning Relevant With Case Studies

The open-ended problems presented in case studies give students work that feels connected to their lives.

To prepare students for jobs that haven’t been created yet, we need to teach them how to be great problem solvers so that they’ll be ready for anything. One way to do this is by teaching content and skills using real-world case studies, a learning model that’s focused on reflection during the problem-solving process. It’s similar to project-based learning, but PBL is more focused on students creating a product.

Case studies have been used for years by businesses, law and medical schools, physicians on rounds, and artists critiquing work. Like other forms of problem-based learning, case studies can be accessible for every age group, both in one subject and in interdisciplinary work.

You can get started with case studies by tackling relatable questions like these with your students:

• How can we limit food waste in the cafeteria?
• How can we get our school to recycle and compost waste? (Or, if you want to be more complex, how can our school reduce its carbon footprint?)
• How can we improve school attendance?
• How can we reduce the number of people who get sick at school during cold and flu season?

Addressing questions like these leads students to identify topics they need to learn more about. In researching the first question, for example, students may see that they need to research food chains and nutrition. Students often ask, reasonably, why they need to learn something, or when they’ll use their knowledge in the future. Learning is most successful for students when the content and skills they’re studying are relevant, and case studies offer one way to create that sense of relevance.

Teaching With Case Studies

Ultimately, a case study is simply an interesting problem with many correct answers. What does case study work look like in classrooms? Teachers generally start by having students read the case or watch a video that summarizes the case. Students then work in small groups or individually to solve the case study. Teachers set milestones defining what students should accomplish to help them manage their time.

During the case study learning process, student assessment of learning should be focused on reflection. Arthur L. Costa and Bena Kallick’s Learning and Leading With Habits of Mind gives several examples of what this reflection can look like in a classroom: 

Journaling: At the end of each work period, have students write an entry summarizing what they worked on, what worked well, what didn’t, and why. Sentence starters and clear rubrics or guidelines will help students be successful. At the end of a case study project, as Costa and Kallick write, it’s helpful to have students “select significant learnings, envision how they could apply these learnings to future situations, and commit to an action plan to consciously modify their behaviors.”

Interviews: While working on a case study, students can interview each other about their progress and learning. Teachers can interview students individually or in small groups to assess their learning process and their progress.

Student discussion: Discussions can be unstructured—students can talk about what they worked on that day in a think-pair-share or as a full class—or structured, using Socratic seminars or fishbowl discussions. If your class is tackling a case study in small groups, create a second set of small groups with a representative from each of the case study groups so that the groups can share their learning.

4 Tips for Setting Up a Case Study

1. Identify a problem to investigate: This should be something accessible and relevant to students’ lives. The problem should also be challenging and complex enough to yield multiple solutions with many layers.

2. Give context: Think of this step as a movie preview or book summary. Hook the learners to help them understand just enough about the problem to want to learn more.

3. Have a clear rubric: Giving structure to your definition of quality group work and products will lead to stronger end products. You may be able to have your learners help build these definitions.

4. Provide structures for presenting solutions: The amount of scaffolding you build in depends on your students’ skill level and development. A case study product can be something like several pieces of evidence of students collaborating to solve the case study, and ultimately presenting their solution with a detailed slide deck or an essay—you can scaffold this by providing specified headings for the sections of the essay.

Problem-Based Teaching Resources

There are many high-quality, peer-reviewed resources that are open source and easily accessible online.

• The National Center for Case Study Teaching in Science at the University at Buffalo built an online collection of more than 800 cases that cover topics ranging from biochemistry to economics. There are resources for middle and high school students.
• Models of Excellence , a project maintained by EL Education and the Harvard Graduate School of Education, has examples of great problem- and project-based tasks—and corresponding exemplary student work—for grades pre-K to 12.
• The Interdisciplinary Journal of Problem-Based Learning at Purdue University is an open-source journal that publishes examples of problem-based learning in K–12 and post-secondary classrooms.
• The Tech Edvocate has a list of websites and tools related to problem-based learning.

In their book Problems as Possibilities , Linda Torp and Sara Sage write that at the elementary school level, students particularly appreciate how they feel that they are taken seriously when solving case studies. At the middle school level, “researchers stress the importance of relating middle school curriculum to issues of student concern and interest.” And high schoolers, they write, find the case study method “beneficial in preparing them for their future.”

Create Your Course

7 top challenges with online learning for students (and solutions), share this article.

We'll discuss the biggest challenges of online learning and possible solutions to these problems to create a more impactful experience for students.

The COVID-19 pandemic put the world at a standstill. But not many things were transformed as much as the education system as the virus spread. Students’ education was abruptly interrupted and schools scrambled to find alternative ways to help students continue their education. 

And there was one glaring solution: Online learning.

Schools started investing in EdTech , and students started taking classes and assignments via Zoom, Google Meet, or some other video conferencing platform.

While distance learning has lowered costs, increased flexibility, and reduced the need for physical infrastructure for both students and teachers, it does not come without its downsides . 

Related: Advantages and Disadvantages of Online Learning

According to Statista, 38% of parents said that one of the major challenges of remote learning is that their children lack the motivation to pay attention and engage during classes. In another study done by Pew Research, 65% of students preferred in-person classes to remote or hybrid learning options. 

These statistics show that online learning has some disadvantages that teachers and students alike need to know about, and try to solve. 

In this piece, we’ll discuss the biggest challenges of online learning and possible solutions to these problems. This way, if you’re a teacher, you’ll know how to support students who are struggling. And if you’re a student, you’ll know what to do when you get into a difficult e-learning-related problem.

Skip ahead here:

• Tech Issues
• Distractions
• Time Management
• Barriers to learning (Disabilities / Special Needs)

Related: 10 Steps To Creating A Wildly Successful Online Course

7 Biggest Problems with Online Learning for Students

Below are the seven biggest challenges of online learning for students (and teachers), and how to solve them. 

Feelings of Isolation

Humans, by nature, are social animals. Most people like to interact with and get to know others, especially in social settings. And although students get to interact with their classmates over Zoom or Google Meet, it is not the same as physical interaction.  

Online learning affords students the ability to study, work, and pursue other interests all at the same time. But the absence of their coursemates and teachers in the immediate environment can cause students to feel isolated. They start to feel disconnected from the class and might not engage the way they normally would in a physical setting.

So it’s no surprise when students turn off their webcams and doze off during online classes. Not only does this foster indiscipline, it also causes students’ academic performance to suffer as teachers cannot personally attend to each student’s needs. 

It’s easy to get frustrated when you can’t talk to your teachers and classmates face to face and voice concerns immediately. However, there are things you can do to power through. Here are some of them: 

• Find out if your school has a student support system in place . Some online schools have advisors and academic staff that guide and support students throughout the duration of their online classes.
• Check if your school offers a networking opportunity for students. Some schools allow students to interact with their peers via chats and forums. It’s similar to interacting with classmates in a physical class, except students have to use online communication etiquette.
• Interact with your teachers and classmates during lectures as much as possible . Ask questions at the right time, organize team projects, and have group discussions with your peers.

For professors, try to make yourself available at certain hours for students who want to reach you. Be dedicated to your students and try to help them achieve their academic and life goals .  

Lack of Motivation

Most students start online classes pumped and ready to go, but as the courses progress, they find that they’re no longer motivated to even attend classes. 

Due to the lack of face-to-face interaction, some students find it hard to focus during online classes. The physical absence of teachers or classmates takes away the sense of urgency and motivation that students need to attend classes on time, meet deadlines, and make progress. This could lead to procrastination and declining grades. 

Contrary to popular opinion, long texts, learning assignments, and quizzes don’t help matters, and can even contribute to students losing motivation to attend classes. 

Lack of motivation is a common issue amongst students. Here are some ways students can maintain a work-life balance and succeed academically: 

• Set realistic long-term and short-term goals and plans to  help them stay on track with classes, assignments, and projects. To-do lists are also important for meeting deadlines. Crossing activities off a to-do list can be highly motivating.
• Students who need direction or help can check out websites and self-help books. 
• Practice positive affirmations. Giving yourself short pep talk to affirm that you can do whatever you set your mind to can motivate you, especially during tough times. 
• Try to interact with teachers and classmates as much as possible. Log in to class daily to see class discussions and course updates. Ask questions, share your opinions, and engage in healthy debate. Being active in class can provide a sense of belonging that keeps you motivated to continue.

Teachers can also incorporate gamification in their online courses  to motivate their students to attend and participate during classes.

Lack of Technical Equipment 

To attend online classes and succeed at remote learning, students need a device with a strong internet connection that they can type assignments on, e.g. laptop, desktop computer, and tablet with a keyboard. 

These devices don’t come cheap, especially for low-income students.

• Some schools give out devices to students that can’t afford them. So if you can’t afford necessary devices for your online classes, ask your school if they provide laptops or tablets to remote students. 
• Use a library. In some regions, public libraries have computers they allow students to use. If you have a library like this in your area, try to use it.
• Borrow from family or friends. If these don’t work out, you could ask relatives or friends who have a laptop to lend you theirs for some time till you’re able to get your own. 

Technical Issues 

Millennials and Gen Zs, as they’re called, are generally proficient in using computers and technology. But this doesn’t mean that they don’t face technical issues from time to time. Learning with computers requires students to understand how to use multiple software–some of which have steep learning curves. 

If a student facing technical issues were on a physical campus, they could easily ask for help from the IT department. With online classes, the student has to try to figure things out alone. If they’re lucky, they’ll have someone close by to help them, but chances are, the person won’t be available all the time. 

Technical issues are not limited to students, though. Teachers face them too–low internet bandwidth, spotty reception, and video glitches, among other things. These issues disrupt learning flow and make learning tedious.

To reduce the technical issues that students and teachers experience during online classes, here’s some measures they could take: 

• Before enrolling in an online class, students should check if they have access to the necessary technology they need to succeed at home. If they don’t, they should check if the school offers technical help (via phone, email, and live chat) to online students.
• When attending online classes, students and teachers should use a high-quality internet service provider (ISP) for fast connection . If they don’t have access to a good ISP at home, they can use free Wi-Fi at a public library or coffee shop nearby. 

There, they can attend classes, participate in group discussions, talk with teachers and coursemates, and turn in assignments. 

• Teachers should provide a comprehensive guide that contains technical times, digital literacy guidelines, and online attendance regulations.
• Teachers should record class sessions on their computer for students who couldn’t make it to class. 

Launch your online learning product for free

Use Thinkific to create, market, and sell online courses, communities, and memberships — all from a single platform.

Online Distractions 

As wonderful as the internet is for learning purposes, it also comes with a ton of distractions. 

Constant notifications from blogs, videos, and social media platforms can distract students from their classes and assignments. And once they’re distracted by these notifications, it’s very easy for them to start scrolling through these platforms mindlessly.

This can cause them to forget that they have classes, assignments, quizzes, or exams.

To be productive in online classes, you need to identify things that can distract you and stop you from achieving your objectives. If you are getting distracted by the internet and social media, here are some things you can do to stop it:

• Turn on social media blockers during classes and exams. Or you can turn off your notifications completely. When you’re done studying, you can go back online or do some other fun activity. 
• Tell people around you about your daily schedule. This way, if they see you getting distracted by technology, they can remind you to do what you planned to do that day. Think of them like human alarm clocks. 
• Find a quiet place to complete your coursework. Even if it’s night, doing your work in a quiet place without your phone (or other unnecessary gadgets) present will help you pay maximum attention to your studies.

To help students pay more attention during classes, here are some things teachers can do:

• Use dynamic learning design to make classes engaging for students . Encouraging your students to build things, take surveys, and have debates can help them concentrate more on their studies, as opposed to the ‘teacher speak, student listen’ learning model.
• Organize tests and quizzes that require students to respond verbally. When students interact physically and mentally during a class, they become less inclined to look through social media and/or blogs.

Related: Easy Topics for Group Discussion With Your Students

Bad Time Management 

It’s hard enough to juggle your normal day-to-day activities without being a student. Online learning adds a few more items to a student’s to-do list, and it can be hard to navigate all these responsibilities. 

While online learning provides students with unparalleled flexibility to do other activities, they have to be good time managers to do their duties effectively and successfully. 

Time management is an important skill that helps students stay focused and disciplined. Students need to learn this skill to effectively manage their academic work and still have time for family, friends, and leisure. Here are few ways students can manage their time better: 

• Try to multitask. Doing two (or more) things at once will save time, and ensure that students will attend classes on time and meet deadlines.
• Make daily to-do lists. Not everyone can multitask. For some, doing multiple things at once can be stressful. If you cannot multitask, create a to-do list and assign times to do each task. When you complete a task, cross it off the list. With time, this habit will increase your overall productivity.

Good teachers can also help their students manage their time better by conducting periodic surveys. If you’re a teacher, these surveys will give you actionable insights into how your students spend their time in a day. 

If you find anything troubling about a student’s schedule, you can offer them personalized guidance and advice.

Disabilities and Special Needs 

Some students may have problems with online classes due to learning difficulties or disabilities. Students with dyslexia, autism, poor vision, hearing impairment, and other disabilities need extra attention to succeed academically. And they can only get that in a physical class.

If you, a teacher, have students with disabilities, check if your online course is universally accessible to all learners . If it isn’t, try to improve usability for everyone. Here are some ways to do that:

• Include captions to your audio and video resources for students with hearing impairments.
• Have voice-over descriptions of text and images.
• Provide alternative learning options like keyboard shortcuts for certain exercises.
• Use AI-powered personal assistants for students with special needs.
• What Is Learning Experience Design? (Tips And Examples)
• How To Design Your Online Course (Visually And Structurally)

Overcome The Challenges of Online Classes

Like most, if not all, things in life, online learning has its upsides and downsides. While it can be overwhelming to juggle attending lectures, working, and interacting with friends, always remember that there are things you can do to make online learning easier for you.

As you ask for help from friends and family, remember to ask your online school for support, too. If you feel stuck, talk to your classmates to see if they’re going through the same thing and how they’re coping.

Also, seek help from teachers and student advisors. And if you’re a teacher or advisor, be willing to support your students and create a conducive setting for learning.

If you, a student, ask for help and follow the tips outlined above, you’ll be able to navigate the challenges with online learning, stay on track, and achieve your goals. 

Colin is a Content Marketer at Thinkific, writing about everything from online entrepreneurship & course creation to digital marketing strategy.

• How To Design An Effective Learning Path
• 4 Cooperative Learning Strategies for Your Class
• 4 Strategies To Boost Participation In Online Courses
• 5 Techniques to Create an Effective Online Course

Related Articles

Steve dotto: how to launch mini-courses that grow your business.

Online course creation expert Steve Dotto's shraes his 5-step process for creating mini-courses to build your list and grow your business.

The 16 Best Personal Finance Courses

Online personal finance courses give you the knowledge and tools you need to meet your money management and investing goals.

How to Increase Completion Rates in Online Courses (Dr. Carrie Rose Interview)

Thinkific Teach Online TV interview with education expert Dr. Carrie Rose on increasing student engagement and completion rates in online courses.

Try Thinkific for yourself!

Accomplish your course creation and student success goals faster with thinkific..

Download this guide and start building your online program!

It is on its way to your inbox

Case Studies for eLearning: How and when they work best

Case studies in education are an age old teaching strategy. They provide meaningful, content-related experiences through which learners can discover and imagine abstract principles in real world settings. In this article, we talk about effective approaches to using case studies in eLearning environments.

Case studies make excellent reading and comprehension activities, while simultaneously serving as information providing tools. These are discovery activities for learners that focus your learners’ analytical and problem-solving skills on the scenario presented in the case study. They are also a great way to demonstrate a real incident or an event that conveys a crucial lesson for best practices. Through this, learners connect intimately and directly with the industry they are training for. In a nutshell, case studies are their first line of contact with the future work environment .

Moreover, case studies make great content for an eLearning interface: think about an eLearning screen with tabs like “About”, “Synopsis”, “Events”, “References”, “Assignment”. Each page includes text and multimedia for learners to tinker and play with. The idea is to display the case in a fun and explorative fashion. This is as opposed to a simple pdf file with lines and lines of text that becomes harder to read with each page!

The goal of a case study is to relay relevant learning materials and then ask questions based on the reading. In eLearning, case studies are richer. The material has more variety, as it is linked from the current resources available on the Internet.

The eLearning medium also provides an interactive environment for case studies with plenty of room for collaboration and online discussion. A complex case study can be simplified using eLearning tools and also become more engaging than the traditional case study delivery method.

The content of your case study could even be retrieved, with the appropriate citations, from a local newspaper. Has there been an article or a news story on some interesting business or an incident? Is it related to your learning objective? Use it as a case study.

Case studies are great for teaching complex knowledge that cannot be taught by using simple formulas. They are especially good for teaching judgement skills and decision making skills required when dealing with real-world dilemmas. They can also be presented in multiple formats. For example, in an instructor-led case study, the case is explained by the instructor and so are the assignments related to it.

Virtual field trips are also great examples of eLearning case studies . Mini-case studies, or vignettes, presented in the beginning of the chapters also serve as great classroom-to-outside-world connection tools. Some case studies can be utilized to develop “Reaction Papers”, in which learners create a summary and reaction to the events they read in the case study.

Simple Strategies to Provide Case Studies

So what are the best practices for case studies? In an eLearning environment, case studies are a rich mixture of multimedia. In the following cases, using a scenario-based approach will work well:

1. When you can create a simulation of an actual system or have extensive video content on it. Find documentaries related to the content, in the form of YouTube videos. Additionally, create an interactive screen with buttons and dialogues between actors to simulate a scene in the case study.

2. Explain charts, diagrams and other technical and business graphics using case studies.

3. Give some life to your numerical data on spreadsheets by narrating a story in the form of a case study.

4. When explaining blueprints, drawings and specifications of products and systems.

5. When demonstrating conventional business documents such as reports, contracts, instruction manuals, email messages, memos and letters.

When using case studies in your eLearning , provide prompts to learners to prepare them that they are about to read a real case. Provide specific guidance and facts to understand the case better. Explain how the case relates to the learning objective or the topic in the eLearning course . What are the important features of the case? What should the learners focus on? Provide sufficient clues on where to start the examination of the case. Also, provide questions in the end of the case to help with brainstorming and critical thinking.

Using case studies is an excellent teaching strategy. Case studies are good for teaching complex knowledge. They promote observing and reflecting on new knowledge and experiences.

Try to use case studies that are based on the local work context of your learners. Learners should be able to relate to the experiences narrated in the case. For your global learners, try to include case studies based on international businesses.

Case studies are also recommended for teaching judgement skills necessary to deal with complex and contradictory situations common in the real world setting.

If you are looking for a way to connect your eLearning course to the outside world, simply integrate a case study.

Improve your employee, partner and customer training with our enterprise-ready learning management system. Book a demo now and see why our diverse portfolio of customers consistently give us 5 stars (out of 5!)

Book a demo

You may also like

Questions to ask before buying an LMS: 60+ tips for selecting the right training platform

Maximizing the benefits of a customizable LMS: How to make a training portal truly yours

Transitioning to a training role? 8 employee training tools to get started

Training designed for everyone: 10 must-have LMS features

5 strategies to spark maintenance staff motivation for online training

How it works

For Business

Join Mind Tools

Article • 10 min read

Case Study-Based Learning

Enhancing learning through immediate application.

By the Mind Tools Content Team

If you've ever tried to learn a new concept, you probably appreciate that "knowing" is different from "doing." When you have an opportunity to apply your knowledge, the lesson typically becomes much more real.

Adults often learn differently from children, and we have different motivations for learning. Typically, we learn new skills because we want to. We recognize the need to learn and grow, and we usually need – or want – to apply our newfound knowledge soon after we've learned it.

A popular theory of adult learning is andragogy (the art and science of leading man, or adults), as opposed to the better-known pedagogy (the art and science of leading children). Malcolm Knowles , a professor of adult education, was considered the father of andragogy, which is based on four key observations of adult learners:

• Adults learn best if they know why they're learning something.
• Adults often learn best through experience.
• Adults tend to view learning as an opportunity to solve problems.
• Adults learn best when the topic is relevant to them and immediately applicable.

This means that you'll get the best results with adults when they're fully involved in the learning experience. Give an adult an opportunity to practice and work with a new skill, and you have a solid foundation for high-quality learning that the person will likely retain over time.

So, how can you best use these adult learning principles in your training and development efforts? Case studies provide an excellent way of practicing and applying new concepts. As such, they're very useful tools in adult learning, and it's important to understand how to get the maximum value from them.

What Is a Case Study?

Case studies are a form of problem-based learning, where you present a situation that needs a resolution. A typical business case study is a detailed account, or story, of what happened in a particular company, industry, or project over a set period of time.

The learner is given details about the situation, often in a historical context. The key players are introduced. Objectives and challenges are outlined. This is followed by specific examples and data, which the learner then uses to analyze the situation, determine what happened, and make recommendations.

The depth of a case depends on the lesson being taught. A case study can be two pages, 20 pages, or more. A good case study makes the reader think critically about the information presented, and then develop a thorough assessment of the situation, leading to a well-thought-out solution or recommendation.

Why Use a Case Study?

Case studies are a great way to improve a learning experience, because they get the learner involved, and encourage immediate use of newly acquired skills.

They differ from lectures or assigned readings because they require participation and deliberate application of a broad range of skills. For example, if you study financial analysis through straightforward learning methods, you may have to calculate and understand a long list of financial ratios (don't worry if you don't know what these are). Likewise, you may be given a set of financial statements to complete a ratio analysis. But until you put the exercise into context, you may not really know why you're doing the analysis.

With a case study, however, you might explore whether a bank should provide financing to a borrower, or whether a company is about to make a good acquisition. Suddenly, the act of calculating ratios becomes secondary – it's more important to understand what the ratios tell you. This is how case studies can make the difference between knowing what to do, and knowing how, when, and why to do it.

Then, what really separates case studies from other practical forms of learning – like scenarios and simulations – is the ability to compare the learner's recommendations with what actually happened. When you know what really happened, it's much easier to evaluate the "correctness" of the answers given.

When to Use a Case Study

As you can see, case studies are powerful and effective training tools. They also work best with practical, applied training, so make sure you use them appropriately.

Remember these tips:

• Case studies tend to focus on why and how to apply a skill or concept, not on remembering facts and details. Use case studies when understanding the concept is more important than memorizing correct responses.
• Case studies are great team-building opportunities. When a team gets together to solve a case, they'll have to work through different opinions, methods, and perspectives.
• Use case studies to build problem-solving skills, particularly those that are valuable when applied, but are likely to be used infrequently. This helps people get practice with these skills that they might not otherwise get.
• Case studies can be used to evaluate past problem solving. People can be asked what they'd do in that situation, and think about what could have been done differently.

Ensuring Maximum Value From Case Studies

The first thing to remember is that you already need to have enough theoretical knowledge to handle the questions and challenges in the case study. Otherwise, it can be like trying to solve a puzzle with some of the pieces missing.

Here are some additional tips for how to approach a case study. Depending on the exact nature of the case, some tips will be more relevant than others.

• Read the case at least three times before you start any analysis. Case studies usually have lots of details, and it's easy to miss something in your first, or even second, reading.
• Once you're thoroughly familiar with the case, note the facts. Identify which are relevant to the tasks you've been assigned. In a good case study, there are often many more facts than you need for your analysis.
• If the case contains large amounts of data, analyze this data for relevant trends. For example, have sales dropped steadily, or was there an unexpected high or low point?
• If the case involves a description of a company's history, find the key events, and consider how they may have impacted the current situation.
• Consider using techniques like SWOT analysis and Porter's Five Forces Analysis to understand the organization's strategic position.
• Stay with the facts when you draw conclusions. These include facts given in the case as well as established facts about the environmental context. Don't rely on personal opinions when you put together your answers.

Writing a Case Study

You may have to write a case study yourself. These are complex documents that take a while to research and compile. The quality of the case study influences the quality of the analysis. Here are some tips if you want to write your own:

• Write your case study as a structured story. The goal is to capture an interesting situation or challenge and then bring it to life with words and information. You want the reader to feel a part of what's happening.
• Present information so that a "right" answer isn't obvious. The goal is to develop the learner's ability to analyze and assess, not necessarily to make the same decision as the people in the actual case.
• Do background research to fully understand what happened and why. You may need to talk to key stakeholders to get their perspectives as well.
• Determine the key challenge. What needs to be resolved? The case study should focus on one main question or issue.
• Define the context. Talk about significant events leading up to the situation. What organizational factors are important for understanding the problem and assessing what should be done? Include cultural factors where possible.
• Identify key decision makers and stakeholders. Describe their roles and perspectives, as well as their motivations and interests.
• Make sure that you provide the right data to allow people to reach appropriate conclusions.
• Make sure that you have permission to use any information you include.

A typical case study structure includes these elements:

• Executive summary. Define the objective, and state the key challenge.
• Opening paragraph. Capture the reader's interest.
• Scope. Describe the background, context, approach, and issues involved.
• Presentation of facts. Develop an objective picture of what's happening.
• Description of key issues. Present viewpoints, decisions, and interests of key parties.

Because case studies have proved to be such effective teaching tools, many are already written. Some excellent sources of free cases are The Times 100 , CasePlace.org , and Schroeder & Schroeder Inc . You can often search for cases by topic or industry. These cases are expertly prepared, based mostly on real situations, and used extensively in business schools to teach management concepts.

Case studies are a great way to improve learning and training. They provide learners with an opportunity to solve a problem by applying what they know.

There are no unpleasant consequences for getting it "wrong," and cases give learners a much better understanding of what they really know and what they need to practice.

Case studies can be used in many ways, as team-building tools, and for skill development. You can write your own case study, but a large number are already prepared. Given the enormous benefits of practical learning applications like this, case studies are definitely something to consider adding to your next training session.

Knowles, M. (1973). 'The Adult Learner: A Neglected Species [online].' Available here .

You've accessed 1 of your 2 free resources.

Get unlimited access

Discover more content

How structure affects strategy.

Building a Framework for Success

Book Insights

Good Charts and Good Charts Workbook

Scott Berinato

Add comment

Comments (0)

Be the first to comment!

Try Mind Tools for free

Get unlimited access to all our career-boosting content and member benefits with our 7-day free trial. 

Sign-up to our newsletter

Subscribing to the Mind Tools newsletter will keep you up-to-date with our latest updates and newest resources.

Subscribe now

Business Skills

Personal Development

Leadership and Management

Most Popular

Newest Releases

Seven Surprises for New Managers

How to Work Effectively With Consultants

Mind Tools Store

About Mind Tools Content

Discover something new today

Risk management and risk analysis.

Assessing and Managing Risks

How to Answer Interview Questions

Responding Confidently While Under Pressure

How Emotionally Intelligent Are You?

Boosting Your People Skills

Self-Assessment

What's Your Leadership Style?

Learn About the Strengths and Weaknesses of the Way You Like to Lead

Recommended for you

Work-life balance overview.

This Article Provides an Overview of Work-Life Balance

168 Hours: You Have More Time Than You Think

Laura Vanderkam

Business Operations and Process Management

Strategy Tools

Customer Service

Business Ethics and Values

Handling Information and Data

Project Management

Knowledge Management

Self-Development and Goal Setting

Time Management

Presentation Skills

Learning Skills

Career Skills

Communication Skills

Negotiation, Persuasion and Influence

Working With Others

Difficult Conversations

Creativity Tools

Self-Management

Work-Life Balance

Stress Management and Wellbeing

Coaching and Mentoring

Change Management

Team Management

Managing Conflict

Delegation and Empowerment

Performance Management

Leadership Skills

Developing Your Team

Talent Management

Problem Solving

Decision Making

• SUGGESTED TOPICS
• The Magazine
• Newsletters
• Managing Yourself
• Managing Teams
• Work-life Balance
• The Big Idea
• Data & Visuals
• Reading Lists
• Case Selections
• HBR Learning
• Topic Feeds
• Account Settings
• Email Preferences

What the Case Study Method Really Teaches

• Nitin Nohria

Seven meta-skills that stick even if the cases fade from memory.

It’s been 100 years since Harvard Business School began using the case study method. Beyond teaching specific subject matter, the case study method excels in instilling meta-skills in students. This article explains the importance of seven such skills: preparation, discernment, bias recognition, judgement, collaboration, curiosity, and self-confidence.

During my decade as dean of Harvard Business School, I spent hundreds of hours talking with our alumni. To enliven these conversations, I relied on a favorite question: “What was the most important thing you learned from your time in our MBA program?”

• Nitin Nohria is a professor and former dean at Harvard Business School and the chairman of Thrive Capital, a venture capital firm based in New York.

Partner Center

This site uses cookies to improve your experience. By viewing our content, you are accepting the use of cookies. To help us insure we adhere to various privacy regulations, please select your country/region of residence. If you do not select a country we will assume you are from the United States. View our privacy policy and terms of use.

• Instructional Design
• Authoring Tools
• Blended Learning
• Virtual Classroom
• Learning Strategy
• Remote Learning
• Gamification

Case Study: Using templates to make eLearning development easier

B Online Learning

JULY 22, 2015

B Online Learning recently went through this process when we were selected via a competitive tendering process by NSW Health to develop a comprehensive online training […].

B Online Learning recently went through this process when we were selected by NSW Health to be the sole developer for […]. The email has arrived, the “BIG” project has been signed off, all systems are go, you just need to produce over a thousand screens of content before the grand unveiling and you have two months.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Trending Sources

Infopro Learning

• E-Learning Provocateur
• Experiencing eLearning
• InSync Training
• Learning Rebels
• Upside Learning
• Clark Quinn

Case Studies: Successful Implementation of Generative AI in E-Learning

SEPTEMBER 22, 2023

Executive Summary: The production, and personalization of content have all been improved by generative AI, completely changing the face of e- learning . In this blog, we examine case studies demonstrating the Successful Implementation of Generative AI in E- Learning and its impact on educational outcomes.

How a NPO is Leveraging Online Learning | GoSkills

GoSkills eLearning

APRIL 17, 2023

Discover how nonprofits can benefit from online learning . Download Building Blocks Mentoring Program's case study and its inspiring journey with GoSkills.

Online Learning Courses – The New Reality for Institutions and Students 

Hurix Digital

APRIL 1, 2020

Students and institutions across the world are not depending on the archaic, uni-dimensional methods of traditional learning anymore. So much so, that almost 65% of educational institutions in the US alone have begun to support state-of-the-art online learning courses regularly! Resource Efficiency. Better Productivity.

Custom eLearning Development: 3 Modern Learning Trends That Will Drive Future Success

MARCH 15, 2021

Investing in online learning programs enables organizations to provide more flexible employee development programs that better suit modern learners. The learning solution that best fits your organization is highly dependent on your business’s needs and goals. Custom eLearning solutions have long been on the rise.

Case Study: FinTalent

MARCH 19, 2019

The month-long competition comprised of three key elements: Gamified mobile learning on SmartUp’s platform. FinTech learning content from industry thought leaders in snappy bite sized online learning . Click here to download a copy of this case study . The results and student reactions. 85% students.

APRIL 27, 2023

Case study: How EY brought its employees together with an online learning event

CLO Magazine

JUNE 1, 2021

“We briefed them about the theme and offered suggestions on how to use it into their case studies . For many years, we concluded our learning event with drinks and a party. We wanted to retain this fun element, but what could possibly be appropriate for our organization and this online learning event?

Case Study: Convert PPT to eLearning Interactive Course

Brilliant Teams

JUNE 12, 2023

Requirement discussion: Convert PPT to eLearning The health and safety consulting firm approached us with a request to convert their existing classroom training materials, consisting of approximately 275 Convert PPT to e- learning course. Transform static presentations into interactive e- learning modules by converting PPT to e- learning .

Falling Head Over Heels for Online Learning: The 5 Love Languages of the Instructional Designer

FEBRUARY 12, 2021

Online learning can conjure some strong love-hate emotions in learners everywhere. Those who are leery of online courses have likely been burned by a bad course experience or burned out trying to create a great course experience for learners of their own. Learning to Love . Want to learn more about CourseArc?

Q&A: Low Country Virtual

SEPTEMBER 28, 2023

After year one of the pandemic, the district superintendents wanted to discontinue emergency remote learning and shift into high-quality online learning based on the National Standards of Quality. We like to distinguish what occurred at the beginning of the pandemic from what online learning should really be.

Accessible eLearning Using Inclusive Design – A Case Study

Harbinger Interactive Learning

OCTOBER 4, 2022

In this blog post, we will look at an interesting case study on eLearning accessibility. This case involved the use of a high level of interactivities including simulation-based assessments for learners to successfully complete the courses. Design decisions to consider when creating accessible courses. What was the Objective?

Training Strategies for the Insurance Workforce: Unlocking Success through Practical Approaches

JUNE 27, 2023

Various strategies can be implemented to ensure that training is results-oriented; these are: Simulations offer hands-on learning experiences but may only partially replicate the complexity of real-world situations. Case studies provide practical examples for analysis and learning but may not always directly apply to every situation.

VILT Case Study: There Was Little Choice About Going To Virtual Training

eLearning Industry

MAY 4, 2021

Discover how the Director of an online learning company made the decision to switch to VILT and what she forecasts for the future of learning and development in a post-pandemic world. This post was first published on eLearning Industry.

[Case Study] Game Creators’ Odyssey – Ubisoft’s recipe on game design

KnowledgeOne

FEBRUARY 17, 2022

Multi pedagogical strategies have been implemented to overcome the different challenges posed by the training program’s transition to online learning . The post [ Case Study ] Game Creators’ Odyssey – Ubisoft’s recipe on game design appeared first on KnowledgeOne. Game Creator’s Odyssey. Visit Website.

Online Learning Summit, The University of Leeds

Webanywhere

JULY 12, 2023

The Online Learning Summit held at the University of Leeds in July, 2023, brought together educators, researchers, and experts to discuss the challenges and opportunities in online education. A key performance indicator for online learning discussed at the summit was the continuation rate.

Case Study: How Tesseract Learning Transformed COL’s C-DELTA Programme into a Gold Brandon Hall Winning Place to Learn

Tesseract Learning

OCTOBER 4, 2021

Commonwealth of Learning (COL) is an intergovernmental organization created by the Commonwealth Heads of Government in 1987 to promote the development and sharing of open learning and distance education knowledge, resources, and technologies.

4 Ways Online Learning Can Help You Grow Professionally | GoSkills

AUGUST 25, 2023

Explore four ways online learning fosters professional growth. To grow professionally doesn’t just mean climbing the corporate ladder.

Improving Online Readiness in Higher Education: A Case Study

NOVEMBER 20, 2018

To truly benefit from eLearning, we must increase students’ desire and ability to learn and perform in this environment. This case study discusses the course eConcordia created for enhancing students’ self-regulated learning , self-motivation, study skills, and technological self-efficacy. Amber Judge. Jihan Rabah.

5 Life-Changing Benefits of Online Learning | GoSkills

JUNE 6, 2023

Discover five life-changing benefits of online learning through the testimonial of one of GoSkills’ learners and most active social media fans.

Benefits of Online Learning: 27 of the Best Secrets You Need to Know Today

Spark Your Interest

JULY 16, 2020

Since Covid-19 people are now more familiar with some of the benefits of online learning . The global pandemic has forced organizations, their employees, and customers, to move to online learning . While most writers on the topic list 5 or 6 of the obvious benefits of online learning there are many more.

Innovate Your Teaching: Converting PowerPoint into eLearning for Impactful Learning Experiences

OCTOBER 31, 2023

By incorporating branching scenarios and adaptive learning , you can create a customized learning path for each student, ensuring that they get the most out of the material. Accessibility : Online learning can be accessed from anywhere, at any time.

eLearning ROI Case Study: How To Maximize Your Returns

JANUARY 9, 2019

I share an eLearning ROI case study that demonstrates how L&D teams can improve and maximize the ROI of corporate training. How To Maximize ROI In Corporate Training: A Case Study . Do read my eBook to get a complete insight into these practical approaches and tips and the eLearning ROI case study .

What does an instructional designer do?

Paradiso Solutions

JUNE 8, 2021

Online learning has transformed how learning is imparted to the learners; it makes learning simpler, easier, and more effective than the traditional teaching method. Today’s learners want personalized and self-paced learning content and access the content an unlimited number of times anytime, anywhere.

How to Pitch Online Learning in a Sales Meeting

DECEMBER 10, 2020

Speaking of which: every time you successfully complete a training program, you should develop your work into a case study to deliver to future clients. As part of the WordPress ecosystem, LearnDash plays well with other WordPress plugins, offering a lot of flexibility to online educators. Demonstrate your success.

How To Create Successful Video Training Content (A Case Study)

TechSmith Camtasia

APRIL 3, 2019

This impressed the Senior Leadership Team and helped outline the value of online learning in a very short time. Finally, I hope you found this case study helpful! Learn more about our software and try them free. They also constructed a User Portal to share training content across the organization.

Case Study: UPV Rapidly Scales Online Learning With edX Online Campus

AUGUST 12, 2020

Polytechnic University of Valencia | Public University | 55,000 students | Valencia, Spain.

Experiential Learning: Catering to varied custom eLearning needs

OCTOBER 26, 2020

As the global workforce continues to grow employees often have different learning needs. They could have varying levels of experience, learning preferences or live in completely different time zones. These challenges can be solved with online learning solutions that use custom content to engage learners.

Tips for Creating Effective Personalized Learning Experiences

OCTOBER 27, 2020

Most analytics tools are easy to operate and many are built into your learning management system. Learning analytics can analyze and report data that will help you evaluate employees online learning experiences and provide valuable insights. Include Social Learning Elements.

8 Best Online Learning Activities to Engage K-12 Students

NOVEMBER 30, 2021

Although there are several benefits that come with this virtual model of learning , one of the main issues that many teachers encounter is a general lack of student engagement online . This blog aims to explore some of the best online learning activities and practices that educators can use for better K-12 student engagement.

Creating Better Content with Lessons Learned from Brain Research

MAY 30, 2022

It’s important for instructional designers to create learning experiences for all—including designing materials that empower neurodivergent students. The best online learning programs and courses are brain-centrically designed. Join CourseArc for a learning science webinar on June 9, 2022. Learn More Schedule a Demo.

Q&A: Wellness for Educators

MARCH 9, 2023

CourseArc, a content authoring and management system (CAMS) was built to support organizations as they facilitate the collaborative creation of engaging and accessible online learning . Want to learn more about CourseArc? Check out our resource site to see how we can help your team.

Getting the Board behind Online Learning Part 2

Association eLearning

AUGUST 26, 2014

In a previous blog post I shared my tips for making an initial pitch for online learning to your association’s board of directors. Are you ready to move on to the “exploration phase” of your quest to implement online learning at your association? Have you ever made the case for eLearning to your association’s board?

MARCH 26, 2019

Why Compliance Training is a Must for Finance Industry?

MARCH 7, 2022

Online learning programs cater to specific job roles and learning interests. Financial organizations may adopt different training delivery modes, such as the online learning approach to provide role-specific content that educates about industry regulations. first appeared on Infopro Learning .

Best Online Learning Platform for Corporate Training

Professional Online Learning | ecadema

JUNE 1, 2022

A survey from Deloitte revealed that almost 75% of digital learning will completely replace the traditional printed textbooks in the coming 10 years. ecadema ecadema is an online platform that offers professional course certification with a simple mission - helping students and tutors find each other.

A case study in eLearning design to underpin and build on a crucial theme

JUNE 3, 2015

Recently, B Online Learning , completed the design and development of “On-Site Insight” for the Australian Medical Association (WA). This piece of eLearning content is primarily a resource for acknowledging and understanding mental health problems in a workplace environment.

Insights for Developing a Software Adoption Training Program

This is why we created 5 different student centered online learning paths. To learn more about software adoption training read the case study Digital Software Adoption Training: Online Learning for a Life Sciences Company The post Insights for Developing a Software Adoption Training Program first appeared on Infopro Learning .

Getting Your Association’s Board Behind Online Learning

APRIL 17, 2014

Understandably, some association executives have anxiety about the type of resistance they may face when presenting online learning to their board for the first time. Start with the purpose of offering online education rather than with the concept itself.

Stay Connected

Join 84,000+ Insiders by signing up for our newsletter

• Learn More about eLearning Learning
• Participate in eLearning Learning
• Selecting the right LMS
• 2019 eLearning Learning Summer Reading List
• Stay At Home Reading List
• Add a Source
• Add a Resource
• 2017 eLearning Learning MVP awards
• 2018 eLearning Learning MVP Awards
• 2019 eLearning Learning MVP Awards
• 2020 eLearning Learning MVP Awards
• 2021 eLearning Learning MVP Awards
• 2022 eLearning Learning MVP Awards
• Sun. Nov 12
• Sat. Nov 11
• Fri. Nov 10
• Thu. Nov 09
• Nov 04 - Nov 10
• Performance
• More Topics 

Input your email to sign up, or if you already have an account, log in here!

Enter your email address to reset your password. a temporary password will be e‑mailed to you., be in the know on.

eLearning Learning

Expert insights. Personalized for you.

We organize all of the trending information in your field so you don't have to. Join 84,000+ users and stay up to date on the latest articles your peers are reading.

Get the good stuff

Subscribe to the following eLearning Learning newsletters:

You must accept the Privacy Policy and Terms & Conditions to proceed.

You know about us, now we want to get to know you!

Check your mail, we've sent an email to . please verify that you have received the email..

We have resent the email to

Let's personalize your content

Use social media to find articles.

We can use your profile and the content you share to understand your interests and provide content that is just for you.

Turn this off at any time. Your social media activity always remains private.

Let's get even more personalized

Choose topics that interest you., so, what do you do.

Are you sure you want to cancel your subscriptions?

Cancel my subscriptions

Don't cancel my subscriptions

Changing Country?

Accept terms & conditions.

It looks like you are changing your country/region of residence. In order to receive our emails, you must expressly agree. You can unsubscribe at any time by clicking the unsubscribe link at the bottom of our emails.

You appear to have previously removed your acceptance of the Terms & Conditions.

We noticed that you changed your country/region of residence; congratulations! In order to make this change, you must accept the Aggregage Terms and Conditions and Privacy Policy. Once you've accepted, then you will be able to choose which emails to receive from each site .

You must choose one option

Please choose which emails to receive from each site .

• Update All Sites
• Update Each Site

Please verify your previous choices for all sites

Sites have been updated - click Submit All Changes below to save your changes.

Show other results

• EDOLOGY NEWS
• BUSINESS & MANAGEMENT
• COMPUTING & IT
• ACCOUNTING & FINANCE
• LAW & CRIMINOLOGY
• MARKETING & COMMUNICATIONS
• FASHION & MEDIA
• PSYCHOLOGY & SOCIAL SCIENCES

5 problems e-learning students experience, and how to overcome them

When historically established methods in any industry are challenged, reimagined, developed and redistributed, a period of acclimatisation for everyone involved is inevitable.

Even the first Industrial Revolution initially fell flat due to the opposition of sceptics, those who were fearful of change, and the machine-smashing luddites. However, there do exist irrepressible waves of change, and if the modernisation of traditional education is a requirement moving into the current fourth Industrial Revolution, prevalent distance education is something we should very much start getting used to.

Here are five problems new students enrolling on online courses might run into, but don’t worry, we’ve also included ways to overcome these concerns so you can make the very best of your online learning opportunities.

Struggles to adapt

Problem: The prospect of having an entire university experience compacted into a personal electronic device is unusual to say the least. It can be unnerving for students who have only ever known traditional classroom settings. Traditionally, a degree of passivity is expected during lectures, particularly when note-taking and listening, while discussion with tutors is allotted a limited time. Online learning demands springing into action, accepting course material in a variety of multimedia formats, and taking part in online discussions which can continue indefinitely.

Solution: First and foremost, an open heart and mind is necessary to be able to accept change and reap the many benefits of e-learning. Secondly, an understanding of the advantages of online education is essential. You can then utilise all of the benefits e-learning can afford you – empowering flexibility, personal tutor guidance, worldwide contact network, 24/7 access to course materials and student support. Once you get started on your online course, you may wonder how you ever learned any other way.

Credibility

Problem: Often, the worth of online learning is not fully trusted, nor given the respect it deserves. Though qualifications are accredited by esteemed university institutions, validity and credibility is met with scepticism because the format is relatively new, and the progress of students is not overseen in the flesh. The perceived value in attending a campus institution, the social education, is also viewed as something not to miss, which can lead to cynicism when considering enrolling in online degree programmes.

Solution: As far as the worth of online education is concerned, employers tend to see the benefit of hiring those who have succeeded online because of the implicit job skills e-learning requires – discipline, initiative, and time-management, as well as technological familiarity. Plus, accreditation from institutions like The University of Law, the UK’s oldest specialised legal training provider, or DeBroc School of Business, is gold-standard in terms of quality of materials and tuition, no matter if you’re on campus or studying online. In relation to the sociability of online learning, remember, you have the flexibility to study anywhere, anytime. Manage your time well, and you might end up with even more time to socialise than your on-campus counterparts. Plus, more money to socialise with.

Outdated hardware and software

Problem: Online learning makes standardised education accessible to students all around the globe. This is a spectacular advantage for online education providers, students in remote locations, and those without the funds to commence traditional campus study programmes where fees and student debt frequently eclipse the joy of learning. However, e-learning does require the necessary computer equipment to run online learning platforms. This can potentially pose problems for students and schools with old, outdated hardware and software.

Solution: While a host of obvious solutions may spring to mind – getting a newer computer, for example – for some that simply may not be possible. However, there are solutions offered by e-learning providers that tackle a wide range of problems faced by students. Course materials are downloadable, which means given time and a little organisation an entire course can be downloaded and the materials studied offline. Learning platforms may also employ a nifty video feature called “Dynamic Stream Switching”, which allows for varying strength of connection and bandwidth in real-time. That means your content always streams, no matter what speed of internet connection is available to you.

Managing time well

Problem: While learning online offers the ultimate freedom to organise your studies around your private and professional commitments, it can lead to complacency and a false sense of security if the appropriate dedication and time is not set aside for serious study. Online courses are every bit as detailed and demanding as their offline counterparts, though this realisation may not be fully formed yet in the Zeitgeist of our time. The intangible, digital nature of e-learning means that bad time-management could lead to failure.

Solution: Time management is something that can be practiced, and with a little discipline, can eventually become a vital asset in the overall skillset of a professional. Keeping prioritised to-do lists, making a study calendar, keeping a diary, using phone apps, are all good habits to nurture. Moreover, the learning platforms themselves are nearly always designed to help you do just this. With platform notifications on upcoming deadlines, email reminders, progress tracking on assignments, and handy tutor feedback direct to your inbox, you need only ever consult your mobile to find out exactly where you are in your course. Getting down to serious study is, as it has always been, down to you.

Discipline and motivation

Problem: Working towards any goal requires dedication and motivation and, on the face of it, studying online can seem fraught with opportunities to lose these qualities. For one, there is the unlimited distraction of already being on the internet; social media, YouTube and news websites are as present as your next assignment. A lot of valuable study time can pass if you don’t monitor your internet usage closely. Given the abstract nature of online learning, motivation in particular can take a hit, especially if you have already spent the day at work in front of a computer screen.

Solution: While advice for staying motivated and disciplined when working towards a goal can be extremely general, when it comes to online education, you can benefit from some specialised tips to keep your eyes on the prize. Firstly, whether you are already employed or not, while undertaking an online qualification it is prudent to treat your studies as if they were an extra part-time job; complete with working hours and repercussions for arriving late and underperformance. Secondly, because of the flexibility of online learning, you can choose a location that you love; a café, a park, a museum, or your favourite library, the choices are endless. Thirdly, keep in mind the very first pay check you’re going to receive. This is part of the real reward for your hard work now.

Explore a full range of undergraduate and postgraduate degree programmes, and professional qualifications right here.

Follow us on Social Media

Comment. Share. Inspire.

More insight from our blog

How to become an accountant – ....

Read more about how to get a head start in accountancy.

What can you do with a marketi...

You're thinking of doing a marketing degree, but what can it really do for you?

7 rewarding aspects of a career in HR

If you want more from your career, maybe a career in HR has what you need?

What can you do with a law degree?

Studying law online is great for launching a wide range of careers.

There’s more to a career in hospitalit...

Want a career that's worldly, and rich with experience?

Why studying while working may be the ...

Whether you are a school leaver, or an industry professional, there are plenty of rea...

Related Blogs

Healthy online studying: dos and don’t...

Fees and Admissions FAQ

Degree and Schedule FAQ

Online Study and Support FAQ

• 1 SELECT Papers & study options >
• 2 Personal details >
• 3 Payment >
• 4 Confirmation

Skolera LMS Blog Educational Technology Articles and News

5 Common Online Learning Problems and Solutions

Aida Elbanna Mobile Learning , Remote Learning , Skolera LMS , Virtual Classrooms Comments Off on 5 Common Online Learning Problems and Solutions 3,862 Views

Let’s discuss the most common online learning problems and solutions!

We get that adapting to an online learning environment can be a bit difficult for some people, especially students and teachers who have recently switched to this method of learning. 

That’s why we’ve decided to discuss, in this article, the most common problems of online learning that you might face and all the possible solutions to get the best outcomes for your students. 

Problems of Online Learning

Lack of interaction/motivation/ being bored.

One of the main difficulties of online learning is that teachers constantly find students lack the drive to learn something new. When students find no motivation to learn, study, or focus during the lesson, it can be frustrating and discouraging for teachers as well.

Lack of Student-Student Interaction

  In a traditional classroom, sometimes teachers would allow 5-minute breaks or extra time in between the lesson. This often encourages students to interact with one another and connect with their classroom peers. Usually, teachers also assign pair-based activities that let students work together towards a common goal. 

This social element is quite important for boosting class morale and building companionship. One of the issues with online learning is that teachers generally neglect this factor as they become too preoccupied with delivering the class content only. 

Internet Problems 

One of the challenges faced by students in online learning is the poor Internet connection. This can be an annoying and frustrating problem for students as they may find it difficult to keep up with the teacher with constant disturbances. 

E-learning School : A Huge Shift in Education

Lack of Discipline 

Many teachers face challenges with online learning as students are generally beyond their control and supervision. Teachers may feel frustrated when he/she realizes he is explaining something but students are not following or are easily distracted. 

So besides technical issues in online learning, teachers sometimes struggle with maintaining discipline in a virtual environment. 

Time Management for Students

With the sudden and swift transformation to online learning especially within the last couple of years, there have been quite a few issues for students. One issue that stands out is poor time management. When a student is learning from the comfort of his/her own bedroom or perhaps even from the bed, it can be difficult to feel productive. 

After 5 minutes of studying, a student may find himself/herself picking up the phone to scroll through social media platforms. At the end of the day, they realize that they haven’t studied properly or sufficiently. 

Solutions to Online Learning Problems

Here are some solutions to common online learning issues for both teachers and administrators:

Give digital literacy courses to teachers before the academic year

Because the shift to online learning can be difficult for many teachers, the school has to ensure the teachers are properly equipped and skilled in the digital sphere. This can take the form of digital learning courses prior to the academic year, and regular follow-ups with the teachers throughout the year for potential troubleshooting or system breakdowns. 

This might initially seem overwhelming for school administrators to implement. However, it only takes some planning and determination to make sure that the virtual learning journey goes smoothly till the end of the year.

Not only will teachers feel comfortable probing the virtual learning world, but they will also be more confident to come up with new techniques and methods to maximize efficiency year by year. 

Ever heard of Skolera’s solutions? Skolera learning management syste m can get all your school administrative information automated with fully equipped SMS and LMS platforms. Our platform will streamline all your school’s operations while giving you premium support. We offer the best features in our LMS solution that will guarantee you a seamless and effort-free management experience. 

Introduce different interactive learning strategies to boost engagement 

When students are feeling bored or demotivated to engage in class, it is the teachers’ duty to revive their interest in the learning material. Here are 3 different strategies to try out this semester:

Problem-Based Learning

Problem-Based Learning can be a great addition to your list of motivational classroom strategies. If included in the lesson, PBL may help learners acquire critical thinking, communication and problem-solving skills. Other benefits include: learning teamwork, becoming more literate in research methods and techniques, learning the importance of analytical skills, and working independently.

Learning by Teaching 

What is special about this technique is that it changes the whole teacher-student dynamic in the classroom. Instead of a traditional lesson in which the teacher talks more and explains things, students are encouraged to partake as well. Research has indicated that when students teach the learning material, they retain information more smoothly.

In your next online lesson, assign presentations to your students at the end of the class. Each pair/group of students may pick a topic and discuss it with the rest of the class. They can even prepare questions to encourage the rest of the students to participate. 

Virtual Field Trips 

If you are a teacher in the fields of biology, chemistry, or history, virtual field trips would be an immense addition to your online classes. Academic virtual trips may appeal to your students who regularly learn from books and written material.

This will be visually appealing and interesting to implement as well. Several websites offer this experience for historical places, artefacts, museums, etc.

Student Debates

Traditional classroom student debates are a great way to increase participation in classes that are lacking it. They not only teach students the value of speaking up, but also effective argumentation skills, and how to respect others’ viewpoints.

It may seem challenging to have an online student debate at first.

The secret, though, is to plan everything out carefully and take into account the occasional voice lags and unstable internet connections.

You can start by introducing a controversial question: conduct a poll to separate the pro-and con-siders, then allow each side time to present its case. The debate can then be led by a representative speaker from each team after that.

Unlike traditional lessons, which are typically mundane and one-sided, this approach will keep your students engaged and eager to participate.

Read more: The Role of a Teacher in Modular Distance Learning – 6 roles!

Collaborative learning

According to Barbara Leigh Smith and Jean T. MacGregor, 

“ Collaborative learning” is an umbrella term for a variety of educational approaches involving joint intellectual effort by students, or students and teachers together. Usually, students are working in groups of two or more, mutually searching for understanding, solutions, or meanings, or creating a product. Collaborative learning activities vary widely, but most center on students’ exploration or application of the course material, not simply the teacher’s presentation or explication of it.”

Bruce Tuckman, the American Psychological Researcher and Professor, devised Tuckman’s five-stage model to define the process of group development.

As seen in the image above, the process consists of Forming, Storming, Norming, Performing, and Adjourning. Therefore, it is clear that each stage is a prerequisite to the next one for a successful group project to work.

For collaborative learning to work, teachers need to guide every transition and ensure that students are on track and focused. 

Follow up with students after each lesson to make sure everything is clear 

To avoid the problem of students feeling lost or frustrated after online classes, try carrying out extra meetings for those students that need more class attention. It will help if you plan specified virtual office hours dedicated to addressing students’ queries and gaps in comprehension.

You need to remember that not all students function the same way; neither do they retain information at the same pace. Addressing this problem will definitely make the whole online learning journey much more productive for students.

Introduce an incentive system to indirectly assert discipline with students  

The problem of discipline in an online learning environment can be maintained by implementing an incentive system that will make each student think twice before disrupting class time.

It is a great way to build connections with the students and keep them constantly wanting to do their best.

This article is a great resource for teachers who want some inspiration on building a class incentive system. 

Finally, we discussed the most common online learning problems and solutions. This article was mainly concerned with the most common online learning problems that students and teachers face today.

We also discussed how teachers and school administrators can overcome these challenges using some tweaks in the system and different teaching approaches.

A teacher needs to pinpoint his/her students’ concerns and struggles and work on solving them accordingly. 

About Aida Elbanna

Related Articles

School Management Excellence – 30 Digital Tools to Empower School Administrators with Skolera

4 weeks ago

Skolera’s Solutions Preparing Teachers for the Digital Age

September 26, 2023

Transform Your Academic Year with Skolera

August 21, 2023

Microlearning: The Art of Learning in Bites

We’ve all been there: trying to study a chapter or topic but finding ourselves unable …

• Reference Manager
• Simple TEXT file

People also looked at

Original research article, prediction of groundwater level under the influence of groundwater exploitation using a data-driven method with the combination of time series analysis and long short-term memory: a case study of a coastal aquifer in rizhao city, northern china.

• 1 No.8 Institute of Geology and Mineral Resources Exploration of Shandong Province, Shandong Rizhao, China
• 2 Key Laboratory of Nonferrous Metal Ore Exploration and Resource Evaluation of Shandong Provincial Bureau of Geology and Mineral Resources, Shandong Rizhao, China
• 3 Rizhao Big Data Research Institute of Geology and Geographic Information, Shandong Rizhao, China
• 4 Rizhao Key Laboratory of land quality evaluation and pollution remediation, Rizhao, China
• 5 Chinese Academy of Geological Science, Beijing, China

The excessive exploitation of groundwater not only destroys the dynamic balance between coastal aquifer and seawater but also causes a series of geological and environmental problems. Groundwater level prediction provides an efficient way to solve these intractable ecological problems. Although several hydrological numerical models have been employed to conduct prediction, no study has accurately predicted the groundwater level change under the consideration of groundwater exploitation, especially in coastal aquifers. This is due to the characteristics of spatially and temporally complex hydrological processes. This study proposes a novel data-driven method based on the combination of time series analysis and a machine learning method for accurately predicting the variation of groundwater level in a coastal aquifer under the influence of groundwater exploitation. The partial autocorrelation function and continuous wavelet coherence were used to analyze the monitoring data of groundwater level at three wells, which indicated that the historical monitored data and the dataset of precipitation could be considered as the input variables to construct the hydrological model. Then, three models based on the different inputs were constructed, namely, the LSTM, PACF-LSTM, and PACF-WC-LSTM models. The performances of the three models were compared by the calculation of four error metrics. The results showed that the performance of the PACF-LSTM and PACF-WC-LSTM models was better than that of the LSTM model and that the PACF-WC-LSTM model achieved the best prediction performance. Accurately predicting the variation of groundwater level provides the basis for managing groundwater resources and preserving the ecological environment.

1 Introduction

Groundwater is the most important resource for supporting the demand from agriculture and industrial and domestic water supplies, and it also plays a crucial role in maintaining the stability of the ecosystem ( Hu et al., 2019 ; Xiao et al., 2022 ; Hikouei et al., 2023 ). Over a quarter of the population across the world depends on groundwater resources as their primary water resource, and more than half of irrigation water in agriculture is supplied by groundwater resources ( WWAP, 2015 ). However, due to rapid urbanization and intensified human activities, the overexploitation of groundwater resources causes a series of intractable environmental problems ( Long et al., 2020 ), such as geological disaster ( Hosono et al., 2019 ; Miyakoshi et al., 2020 ; Qu et al., 2020 ), land subsidence ( Wang et al., 2013 ; Xiao et al., 2022 ; Hikouei et al., 2023 ), and land desertification ( Daliakopoulos et al., 2005 ; Qu et al., 2021 ; Sun et al., 2022 ). Moreover, the intensive exploitation of groundwater induces an imbalance between the surface water, groundwater, and seawater in coastal regions, which causes saltwater intrusion and land salinization ( Tokunaga, 1999 ; Lee et al., 2013 ; Nourani et al., 2014 ; Wang et al., 2023 ). Thus, it is extremely urgent to solve these difficult environmental problems caused by the overexploitation of groundwater resources.

The scientific monitoring and accurate prediction of groundwater levels have focused on solving these intractable environmental problems and providing the basis for the implementation of effective management of groundwater resources ( Hosono et al., 2019 ; Mao et al., 2022 ; Mohammed et al., 2022 ). In general, the dynamic change of groundwater level is affected by external influencing factors, such as seismic activities, precipitation, and pumping activities ( Nourani et al., 2014 ; Shi et al., 2018 ; Wang et al., 2018 ; Gao et al., 2020 ; Vittecoq et al., 2020 ). All the external influence factors can be classified into three categories, namely, geological factors, meteorological factors, and anthropogenic factors, which cause groundwater levels to show non-linear dynamic changes in the time domain and frequency domain. The influence of many external factors on groundwater dynamic changes increases the difficulty of groundwater level prediction and decreases the accuracy of prediction. Previous studies have indicated that physical-based models, such as GMS, MODFLOW, and TOUGH, have a predominant advantage for the prediction of groundwater levels in complex hydrogeological conditions ( Chen et al., 2020 ; Tawara et al., 2020 ; Mohammed et al., 2022 ). However, these numerical models completely depend on hydrological information, such as stratigraphic structures, aquifer parameters, and boundary conditions. Due to the heterogeneity, discontinuity, and anisotropy of aquifer properties across different scales, hydrological parameters are difficult to obtain accurately. In recent years, data-driven methods have been shown to outperform numerical models in predicting the variation of groundwater level ( Kratzert et al., 2018 ; Bredy et al., 2020 ; Zhang et al., 2022 ; Hikouei et al., 2023 ). The greatest strength of data-driven methods is that these methods can build the relationship between the input variables and target variables without the need to explicitly define physical relationships between them.

Data-driven methods are needed to rebuild the relationship between external influence factors and the variation of groundwater level from a modern perspective. These types of methods have been widely considered to identify anomalous changes in groundwater levels and predict the variation of groundwater levels. Examples of data-driven methods include the decision tree model ( Bredy et al., 2020 ; Zhang et al., 2020 ), the Hilbert Huang Transform ( Zhang et al., 2019 ; Chien et al., 2020 ), and the artificial neural network method ( Wunsch et al., 2021 ; Hakim et al., 2022 ). However, these methods also have major drawbacks. For example, the decision tree method is easy to overfit during the training state, whereas the artificial neural network method cannot quantify how much historical data is used for prediction. n addition, these data-driven groundwater prediction methods do not consider the various external factors that influence the dynamic variation of groundwater level. More recently, the newly developed long short-term memory neural network (LSTM) method has provided an effective way to predict the variation of groundwater level based on valuable data from long-term monitoring and external influencing factors. However, in previous studies, groundwater withdrawal was less considered as the input variable to construct the LSTM model for predicting the variation of groundwater level. In order to improve the accuracy of the prediction model, in the present study, three modified LSTM models with the combination of time series analysis and machine learning methods based on the consideration of groundwater withdrawal were constructed and compared in terms of the performance of groundwater level prediction.

The core objective of this study is to accurately predict the variation of groundwater level under the influence of groundwater exploitation in a coastal area. In accordance with this objective, major research results were achieved by 1) using partial autocorrelation function and continuous wavelet coherence to identify the internal factors and external factors on the variation of groundwater level, and then to determine the input variables of data-driven models; 2) constructing and training three data-driven models, namely, the LSTM model, the PACF-LSTM model, and the PACF-WC-LSTM model, to predict groundwater level; 3) analyzing and comparing the model performance of groundwater level prediction in the validation and prediction stage under four error metrics, namely, R 2 , MAPE, RMSE, and NSE. Hydrogeologists analyze and predict the variation of groundwater levels, especially those changes under the influence of groundwater exploitation, which can provide the premise for water resource management. However, due to the non-linear and non-stationary characteristics of groundwater level monitored data, there is still no data-driven method to accurately predict groundwater level change under the influence of groundwater exploitation. The result of this study could provide a new data-driven method to simulate and predict groundwater level change under groundwater exploitation in a coastal aquifer.

2 Background of study area and data sources

2.1 regional hydrogeological setting.

Rizhao County is located in the Jiaonan uplift of the second-order structural unit, which belongs to the Ludong fault block of the first-order structural unit. Tectonically, it is located in the junction area of the Yishu Fault and the Wurong Fault. The terrain is generally low in the southeast and high in the northwest, and generally increases in height with the increasing distance between the coastline and inland. The length of the coastline in Rizhao County is 168 km. Geomorphic types are divided into three categories: mountain, hill, and plain. The area of hills occupies approximately 57.2% of the city’s territory, whereas the areas of mountains and plains occupy approximately 25.3% and 17.5% of the area, respectively. The main surface water bodies include the Futuan River, Chaobai River, Xiuzhen River, and Wei River, which flow into the Huanghai Sea.

Based on the difference in hydrogeological conditions, the aquifer types in the study area can be roughly classified into four categories: Quaternary loose rock aquifer, Bedrock fissure aquifer, Clastic rock aquifer, and Carbonate rock aquifer ( Figure 1 ). The Quaternary loose rock aquifer is mainly distributed on both sides of the Futuan River and Xiuzhen River and has a wide distribution range and strong water supply capacity. The hydrochemical type in the Quaternary loose rock aquifer is HCO 3 ·Cl-Ca·Na. The Bedrock fissure aquifer is the most widely distributed in the study area, however, its water supply capacity is inadequate. The hydrochemical type in the Bedrock fissure aquifer is HCO 3 -Ca·Mg. The rock types of the Clastic rock aquifer are composed of conglomerate, siltstone, and clastic rock. The hydrochemical type in the Clastic rock aquifer is HCO 3 —Ca·Na. The Carbonate rock aquifer is less distributed in the study area.

FIGURE 1 . The hydrogeological map of Rizhao City and the location of monitoring wells.

2.2 Data collection

Dongguan monitoring well (DG well), Jufeng monitoring well (JF well), and Kouguan monitoring well (KG well) are connected with the Carbonate rock aquifer, Quaternary loose rock aquifer, and Bedrock fissure aquifer, respectively. In this study, the monitored data of groundwater levels in the abovementioned three monitoring wells from 2003 to 2020 were collected to be used in the analysis and prediction of groundwater levels. In addition, precipitation and groundwater withdrawal are important sources of groundwater recharge and discharge, respectively. To accurately analyze and predict groundwater regimes, we also collected the dataset of precipitation and groundwater withdrawal in Rizhao City from 2003 to 2020.

The monitored interval of groundwater levels in these three monitoring wells was 5 days. The dynamic changes in groundwater levels in the three monitoring wells showed seasonal fluctuations, with a higher level during the summer wet season and lower levels during the drier winter season ( Figure 2 ). Due to the difference in hydrological conditions, the magnitude of annual change in groundwater levels showed significant differences: 2.6 m in the DG well, 1.7 m in the JF well, and 3.5 m in the KG well.

FIGURE 2 . The variation of groundwater levels at the three monitoring wells (DG well, KG well, and JF well) from 2003 to 2020.

The meteorological dataset was collected from the China Meteorological Administration ( http://data.cma.cn/ ). Rizhao County is characterized by the monsoon climate of medium latitudes, with a mean annual rainfall of 874 mm, of which approximately 70% falls from June to October. The average value of annual atmospheric temperature is approximately 12.7°C.

The data on groundwater withdrawal in Rizhao was collected from the Rizhao hydrological reports published by the Rizhao Water Resources Bureau ( http://slj.rizhao.gov.cn/ ). As shown in Figure 3 , the annual average value of groundwater withdrawal from 2003 to 2020 was 1.595 × 10 8  m 3 . The maximum value of groundwater withdrawal was 1.956 × 10 8  m 3 in 2006. Due to the change in water supply structure and effective administration, the amount of annual groundwater withdrawal in Rizhao was reduced to 1.363 × 10 8  m 3 in 2020.

FIGURE 3 . The variation of precipitation and groundwater withdrawal from 2003 to 2020.

In order to eliminate the effect of different monitored intervals, the monitored values of groundwater level, precipitation, and groundwater withdrawal were transferred to the monthly average value for analysis.

3.1 Partial autocorrelation function

The partial autocorrelation function (PACF) is an efficiency tool in time series analysis for analyzing the correlation between the X t and X t+k by eliminating the variables interference Y t-1 , Y t-2 , … , Y t− k+1 . Partial autocorrelation coefficients can be calculated by the Yule-Walker equation ( Tinungki and Iop, 2019 ; Yan et al., 2021 ; Zhang et al., 2022 ; He et al., 2022 ), as follows:

Where φ k k is the correlation coefficient between two variables X t and X t+k . It is defined by the following equation

Partial autocorrelation coefficients can be estimated by using the partial autocorrelation coefficients of the sample by changing the value ρ on the Yule-Walker equation with r, and counting for k = 1, 2, … to get the value Φ kk using Cramer rules. Several previous studies indicate that PACF provides an efficient way to analyze the correlation of time series in hydrogeological science ( Rodrigues et al., 2018 ; Yu et al., 2019 ; Bredy et al., 2020 ; Nelson et al., 2021 ; Yan et al., 2021 ). In this study, we also use this method to analyze the correlation of time series between groundwater level at time t and antecedent groundwater level.

3.2 Continuous wavelet coherence

Wavelet transform is an efficient tool to decompose the time series into various times and frequencies and analyze non-stationary time series with multi-time resolution, which mainly includes the discrete wavelet transform (DWT) and the continuous wavelet transform (CWT) ( Acworth et al., 2016 ; Yan et al., 2020 ; Zhang et al., 2020 ; Qu et al., 2021 ). Previous studies indicated that the latter method has been used to analyze the correlation of different hydrological time series ( Massei et al., 2006 ; Nourani et al., 2014 ; Yan et al., 2017 ; Lee and Kim, 2019 ; Zhang et al., 2021 ). In this study, continuous wavelet coherence was used to identify the external influencing factors of groundwater level change, which is the typical technology in continuous wavelet transform. Due to the perfect performance and good balance in the time and frequency domain, the Morlet wavelet was selected to conduct the continuous wavelet coherence ( Massei et al., 2006 ; Zhang et al., 2022 ; Gu et al., 2022 ). Monte Carlo methods were used to determine the statistical significance level of WTC. The Cone of Influence was used to evaluate the edge effects caused by discontinuities at endpoints.

Different from the definition of the correlation coefficient, continuous wavelet coherence is defined as follows:

where the W operator represents the continuous wavelet transform when it has one argument. The capital letter S and lowercase letter s represent the smoothing operator and the wavelet scale, respectively. R 2 is the correlation coefficient, which ranges from 0 to 1. The value of 1 means a high correlation between two time series, while the value of 0 means a low correlation between them.

3.3 Long short-term memory neural network

Data-driven artificial neural networks can simulate the data processing process of the human brain. Both recurrent neural networks and long short-term memory are typical artificial neural networks, which are widely used to analyze the non-linear characteristics between input and output variables. The long short-term memory proposed by Hochreiter and Schmidhuber (1997) is the special structure of a recurrent neural network (RNN). Similar to the typical structure of a recurrent neural network, the LSTM network is also composed of three layers: the input layer, the hidden layer, and the output layer. The obvious difference between recurrent neural networks and long short-term memory networks is the algorithm structure of the hidden layer ( Rodrigues et al., 2018 ; Zhang et al., 2022 ; Zhang et al., 2022 ; Sun et al., 2022 ). In recurrent neural networks, the unrolled loop cell is the medium of information transformation, which stores the historical information of time series and allows the historical information to conduct predicting ( Wunsch et al., 2021 ). However, the major drawback of a recurrent neural network is that the unrolled loop cell cannot identify how much historical information should be used to predict the time series. Meanwhile, it also causes vanishing gradients and gradient explosion during the back-propagation. The efficient structure of the hidden layer in the LSTM network, comprising three gates, namely, input gates, output gates, and forget gates, can solve the drawback of the RNN ( Zhang et al., 2018 ; Chen et al., 2021 ; Vu et al., 2021 ; Mohammed et al., 2022 ). The gate structure of the hidden layer controls which historical data in the time series is important to keep and protects the valuable information passed down in the process of information transfer. The distinct structure of the hidden layer can efficiently solve the problem of gradient explosion and gradient disappearance in the training stage ( Rodrigues et al., 2018 ; Yu and Ma, 2021 ).

Detailed information on the forget gate, input gate, and output gate is introduced as follows:

1) The forget gate can read the stored information in the previous hidden state h t − 1 and the current input varies x t . The sigmoid function in the forget gate is used to determine which information is stored or ignored from the previous cell ( Kratzert et al., 2018 ; Hakim et al., 2022 ; Mohammed et al., 2022 ). The forget gate is defined by the following equation:

Where σ is the sigmoid function which outputs a number ranging from 0 to 1. The value of 0 means that the historical information is ignored, while the value of 1 means that the historical information is valuable for prediction and should be kept. W f is the network matric of the forget gate, and b f is a bias vector.

2) The input gate identifies what information is to be retained and updated in the cell state, which consists of two layers: the tanh layer and the sigmoid layer. These two layers process the data simultaneously. The tanh layer calculates the update vector based on the last hidden state, and the sigmoid layer determines which historical information can be retained to update the cell state in the current time step. The abovementioned process is defined by the following equation:

Where C t is the cell state, C ∼ represents the update vector, and tanh is the hyperbolic tangent. W and b represent the weight matric and bias vector, respectively. The subscripts C and i define the tanh layer and input gate, respectively.

3) The output gate determines which historical information can be passed on to the new hidden layer. It is defined by the following equation:

Where W o is the weight matric and b o is the bias vector.

4 Model development

To improve the accuracy of model prediction, three data-driven models were constructed by machine learning methods: the LSTM model, the PACF-LSTM model, and the PACF-WC-LSTM model. In this section, we introduce how to split the dataset into different stages, how to normalize the dataset, and how to identify the input variables of each model. In the present study, all the modified LSTM models were programmed by MATLAB.

4.1 Splitting the dataset into different subsets

The splitting of the dataset is an important step to train the machine learning model. If the training subset takes up a small proportion, the machine learning model may not analyze and identify the mathematic characteristics of the time series, leading to a reduced accuracy of prediction. If the training and validation subsets take up a large proportion, the model may overfit and lead to data not being accurately predicted. However, there is no fixed ratio between the training dataset, validation dataset, and test dataset ( Rodrigues et al., 2018 ; Wunsch et al., 2021 ; Zhang et al., 2022 ). In general, the training dataset should comprise more than 50% of the whole dataset. In this study, the hydrological monitored dataset was split into three subsets: the training subset, the validation subset, and the prediction subset. The proportion of these subsets was 5:3:2. To accurately build the relationship between the input variables and the target variable, the training and validation process is aimed at optimizing the model parameters.

4.2 Data normalization and error metric

Due to the diversity of monitoring data, the min–max normalization approach was used to normalize all input variables into the range of [0,1], which can improve the learning and training efficiency, and eliminate external influences, especially the dimensional influence. The min–max normalization approach is defined by the following equation:

Where x norm is the normalized value and x , x max , and x min are the monitored value, the maximum monitored value, and the minimum monitored value, respectively. After training, the model output results can be retransformed through the contrary process of Eq. 10 .

Four error metrics were selected to evaluate the accuracy and predictive efficiency of the machine learning model, as follows:

The coefficient of determination:

The root mean square error (RMSE):

The mean absolute percentage error (MAPE):

The Nash-Sutcliffe efficiency (NSE):

Where y i is the observed value, y i * is the simulated value, y i _ is the mean of observed values, and N is the number of observations. R 2 is the typical error metric between the monitored values and simulated values ( An et al., 2020 ; Zhang et al., 2022 ). If R 2 is close to 1, it indicates that the model predictability is accurate. RMSE is defined as the deviation between the monitored values and the model-simulated values ( Nourani and Mousavi, 2016 ; Xiao et al., 2018 ). The smaller the value of RMSE, the better the model accuracy is. MAPE is used to evaluate model performance and accuracy as a percentage. If the value of MAPE is <10%, the accuracy of the model is considered excellent. If the value of MAPE is >50%, the model performance and prediction result are inaccurate. NSE is the traditional efficiency indicator for hydrologic models, which is used to evaluate the model accuracy in the training and test stage ( Hussein et al., 2013 ; Barzegar et al., 2017 ; Hakim et al., 2022 ). The value of NSE close to 1 means that the model predictability is satisfactory. In general, when the RMSE and MAPE are close to 0, and NSE and R 2 are close to 1, the model is regarded as a good fit between simulated and observed values.

4.3 Input variable selection

The appropriate input variables provide the basic hydrological information for constructing the hydrological models. Due to the complexity of hydrogeological conditions, there are no guidelines on how to select the input variables for the construction of a hydrological model. The regional groundwater regime is affected by meteorological factors and human activities. In the present study, partial autocorrelation function and continuous wavelet coherence were introduced to identify the influencing factors on the variation of groundwater level and help us select the input variables of the hydrological model. The determination of input variables in each model is introduced in detail in Section 5.3 .

5 Results and discussion

5.1 the mathematical characteristic of groundwater level time series.

The autocorrelation analysis was the efficiency tool used for analyzing the correlation relationship between the hydrological data ( d ) and the historical time series [ d (t-1), d (t-2), … d (t-p)] with p being the lag time. In order to determine whether the historical monitored data of groundwater level data could be considered as an input variable, the autocorrelation analysis was conducted based on the monthly monitored data of groundwater level during the interval of 2003–2020. The partial autocorrelation coefficient of monthly data is shown in Figure 4 . The result indicates that the 0th-order partial autocorrelation coefficients are constant at 1. In addition, it is easily found that the autocorrelation coefficients fluctuate around the 0-axis with the increase of lag time, which indicates that the time series of groundwater level in three monitoring wells show the stationary signal.

FIGURE 4 . Partial autocorrelation coefficient of groundwater level time series in the DG well, KG well, and JF well (Blue solid line is the 95% confidence bound).

For the DG well, the PACF result showed a significant correlation with up to 3 months of lag time for groundwater levels. Hence, the lag time, p, is equivalent to 3 months for groundwater level at the DG well. Similarly, the lag time, p, is also equivalent to 3 months for the groundwater level of the KG well and JF well. The PACF results of three wells indicate that a strong correlation relationship exists between the groundwater level data and the historical monitored data. Thus, the historical dataset of groundwater can be considered as the input variable to predict the target variable.

5.2 The external influencing factors on the variation of groundwater level

The wavelet coherence was used to analyze and examine the relationship between the change in groundwater level and the variation of precipitation, which is an efficient time series analysis tool. The coherence relationship between groundwater level and precipitation in three wells is shown in Figure 5 . The thick black contour indicates the 95% confidence level. The black arrows indicate the relative phase relationship. The in-phase points to the right, while the anti-phase points to the left. The phase-lagging by 90° points straight up while the phase-leading by 90° points straight down.

FIGURE 5 . Wavelet coherences WTC (1979–2015) between groundwater level and precipitation in (A) the DG well, (B) the JF well, and (C) the KG well.

For the DG well, the groundwater level and precipitation were highly coherent at a >95% point-wise confidence level within the band between 256 days and 512 days (about 1 year) during the interval of 2003–2015 and 2016–2017. Similarly, in the KG well and JF well, high coherence was evident for precipitation and groundwater levels within the band between 256 days and 512 days (about 1 year) throughout the whole monitoring period. In addition, the results of wavelet analysis also indicated that the variation of groundwater level in the three monitoring wells lagged behind precipitation change. The mean phase angles between groundwater level and precipitation were approximately 45°, 60°, and 75° at the DG well, KG well, and JF well, respectively. The lag time between groundwater level and precipitation was 45, 60, and 75 days, at the DG well, KG well, and JF well, respectively.

5.3 Comparisons of groundwater level prediction performance between the LSTM, PACF-LSTM, and PACF-WC- LSTM models

The result of partial autocorrelation analysis and wavelet coherence analysis indicated that the historical monitored data of groundwater level and the monitored data of precipitation could be considered as the input variables to construct a model for predicting the variation of groundwater level. The LSTM model was set up by the dataset of groundwater withdrawal. For the PACF-LSTM model, the historical monitored data of groundwater level were considered as the second input variables to construct the model. The input variables for training the PACF-WC-LSTM model included groundwater withdrawal, historical groundwater level, and precipitation. The input and output variables of the LSTM model, PACF-LSTM model, and PACF-WC-LSTM model are summarized in Figure 6A and Table 1 . Based on the abovementioned splitting strategy described in Section 4.1 , the percentage of training subset, validation subset, and prediction subset were 50%, 30%, and 20%, respectively. The training stage was from January 2003 to August 2011. The validation stage was from September 2011 to December 2016. The prediction stage was from January 2017 to June 2020.

FIGURE 6 . (A) A schematic flowchart for the LSTM model, PACF-LSTM model, and PACF-WC-LSTM model. (B) A schematic flowchart for an LSTM model Q14 modified from Yan et al. (2021) .

TABLE 1 . The input and output variables of different models.

The validation and prediction results of the DG well, JF well, and KG well calculated by the LSTM model, PACF-LSTM model, and PACF-WC-LSTM model are shown in Figures 7 – 9 , respectively. Their performance indexes are summarized in Table 2 . Although the variation of groundwater level under the effect of groundwater exploitation could be fitted by the three different models, different error indicators indicated that the performance of the three models was different.

FIGURE 7 . The training, validation, and prediction results of groundwater at (A) the DG well, (B) the JF well, and (C) the KG well by the LSTM method.

FIGURE 8 . The training, validation, and prediction results of groundwater at (A) the DG well, (B) the JF well, and (C) the KG well by the PACF-LSTM method.

FIGURE 9 . The training, validation, and prediction results of groundwater at (A) the DG well, (B) the JF well, and (C) the KG well by the PACF-WC-LSTM method.

TABLE 2 . The errors of three models in the validation and prediction stage in the DG well, the JF well, and the KG well.

NSE is the traditional efficiency indicator for evaluating the accuracy of hydrological models. For each monitoring well, the NSE value varied with different models. In the DG well, the NSE values of the validation and prediction stage in the LSTM model were 0.92 and 0.87, respectively, which were the smaller values in all three models. The smaller the NSE value is, the poorer the performance of the hydrological model. In comparison with the LSTM model, the PACF-LSTM model and PACF-WC-LSTM model increased the NSE values of the validation stage by 2% (from 0.92 to 0.94) and 4% (from 0.92 to 0.96), respectively ( Table 3 ). For the prediction stage, the NSE values in the PACF-LSTM model and PACF-WC-LSTM model increased by 5% (from 0.87 to 0.91) and 8% (from 0.87 to 0.94), respectively. In addition, for the JF well, the change ratio of the NSE value was 1% and 3%, respectively, in the validation stage, and 0% and 1%, respectively, in the prediction stage. For the KG well, the change ratio of the NSE value was 1% and 7%, respectively, in the validation stage, and 2% and 8%, respectively, in the prediction stage. The results indicated that the quality of validation and prediction of the PACF-LSTM model and the PACF-WC-LSTM model were better than the LSTM model. Based on the error metric of NSE, the quality of the validation and prediction based on the PACF-WC-LSTM was the best in the three monitoring wells.

TABLE 3 . The change ratio of NSE value in the three models at different monitoring wells.

RMSE is used to calculate the difference between the monitored value and the stimulated value. The smaller value of RMSE indicates that the model performance is perfect. Take the DG well as an example. For the LSTM model of the DG well, the RMSE values of the validation and prediction stages were 0.2080 and 0.2176, respectively. Compared with the LSTM model, the RMSE values of the PACF-LSTM and PACF-WC-LSTM models in the validation stage reduced by 11% (from 0.2080 to 0.1846) and 28% (from 0.2080 to 0.1502), respectively, and the values of the prediction stage reduced by 18% (from 0.2176 to 0.1792) and 35% (from 0.2176 to 0.1406), respectively ( Table 4 ). For the validation stage in the other monitoring wells, the change ratio of RMSE value in the PACF-LSTM model was 9% and 4%, respectively, and the change ratio was 23% and 41%, respectively, in the PACF-WC-LSTM model. For the prediction stage in the other monitoring wells, the change ratio of RMSE value in the PACF-LSTM model was 4% and 8%, respectively, and it was 8% and 38%, respectively, in the PACF-WC-LSTM model. The change ratio of RMSE value also indicated that the prediction performance of PACF-WC-LSTM was the best in the three data-driven models.

TABLE 4 . The change ratio of RMSE value in the three models at different monitoring wells.

The MAPE value was introduced to evaluate the difference between the model prediction result and the monitored value as a percentage. For the DG well, the MAPE values in the validation stage and prediction stage of the LSTM model were 0.0978 and 0.0923, respectively. Those values in the PACF-LSTM model were reduced by 19% (from 0.0978 to 0.0788) and 12% (from 0.0923 to 0.0809), respectively. In the PACF-WC-LSTM model, those values were reduced by 42% (from 0.0978 to 0.0569) and 33% (from 0.0923 to 0.0622), respectively ( Table 5 ). For the JF well, the change ratio of MAPE value in the validation stage was 9% for the PACF-LSTM model and 26% for the PACF-WC-LSTM model. Those values of the prediction stage were 10% and 20% for the PACF-LSTM model and PACF-WC-LSTM model, respectively. For the KG well, the change ratio of MAPE value in the PACF-LSTM model was 10% for the validation stage and 18% for the prediction stage. In the PACF-WC-LSTM model, those values were 52% and 32% in the validation stage and prediction stage, respectively.

TABLE 5 . The change ratio of MAPE value in the three models at different monitoring wells.

In order to analyze the prediction results by three models, the scatter plot of stimulated value and monitored value in the validation and prediction stage are displayed in Figures 10 – 12 , respectively. The X -axis and Y -axis represent the monitored value and simulated value, respectively. If the model has perfect performance, the prediction results should be distributed over X = Y or evenly distributed on both sides of the line. The closer the distribution of scatter to the 1:1 line, the smaller the model error. R 2 was introduced to evaluate the performance of different models. The results indicated that the performance of the PACF-WC-LSTM model was the best.

FIGURE 10 . Scatter plot of the monitored value vs. the simulated value calculated by the LSTM method in the validation stage and prediction stage. (A1–C1) represent the validation stage of the LSTM model at (a) the DG well, (b) the JF well, and (c) the KG well, respectively. (A2–C2) represent the prediction stage of the LSTM model at (a) the DG well, (b) the JF well, and (c) the KG well, respectively. The blue and red dashed lines represent the trend lines of validation and prediction, respectively. Gray dotted lines represent a 1:1 line.

FIGURE 11 . Scatter plot of the monitored value vs. the simulated value calculated by the PACF-LSTM method in the validation stage and prediction stage. (A1–C1) represent the validation stage of the PACF-LSTM model at (a) the DG well, (b) the JF well, and (c) the KG well, respectively. (A2–C2) represent the prediction stage of the PACF-LSTM model at (a) the DG well, (b) the JF well, and (c) the KG well, respectively. The blue and red dashed lines represent the trend lines of validation and prediction, respectively. Gray dotted lines represent a 1:1 line.

FIGURE 12 . Scatter plot of the monitored value vs. the simulated value calculated by the PACF-WC-LSTM method in the validation stage and prediction stage. (A1–C1) represent the validation stage of the PACF-WC-LSTM model at (a) the DG well, (b) the JF well, and (c) the KG well, respectively. (A2–C2) represent the prediction stage of the PACF-WC-LSTM model at (a) the DG well, (b) the JF well, and (c) the KG well, respectively. The blue and red dashed lines represent the trend lines of validation and prediction, respectively. Gray dotted lines represent a 1:1 line.

Based on the abovementioned analysis, the quantitative evaluation metric for the three models indicates that the prediction performance of the PACF-WC-LSTM model is superior to those of the LSTM model and PACF-LSTM model.

6 Conclusion

For improving the accuracy of simulation and prediction of non-linear and non-stationary groundwater level change under the influence of groundwater exploitation in a coastal aquifer, we proposed a novel data-driven method called PACF-WC-LSTM, based on the combination of time series analysis and machine learning method. The prediction performance of PACF-WC-LSTM was compared with the LSTM model and the PACF-LSTM model by the error metric of R 2 , RMSE, NSE, and MAPE. These three models were applied to predict the change in groundwater level in three monitoring wells that were connected with different aquifer types. This study draws the following conclusions:

1) The partial autocorrelation function results indicate that a strong correlation relationship exists between the groundwater level data and the historical monitored data of groundwater level at three monitoring wells. The lag time is approximately 3 months for the groundwater level at the DG well, JF well, and KG well. The historical monitored data of groundwater level can be considered as the input variables for constructing the prediction model.

2) The continuous wavelet coherence results indicate that the groundwater level at three monitoring wells and precipitation are highly coherent within the band of about 1 year during the whole monitoring period. The dataset of precipitation can be considered as the input variables to construct a hydrological model.

3) Based on the results of the partial autocorrelation function and continuous wavelet coherence, three data-driven models, namely, the LSTM model, PACF-LSTM model, and PACF-WC-LSTM model, were constructed, and four error metrics, namely, R 2 , RMSE, NSE, and MAPE, were introduced to evaluate the model performance. The results indicate that the PACF-WC-LSTM model yields a better prediction performance than the LSTM model and the PACF-LSTM model, and can be used to predict the variation of groundwater level under the influence of groundwater exploitation in the coastal aquifer.

Data availability statement

The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation.

Author contributions

BG: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Project administration, Writing–original draft. SZ: Conceptualization, Data curation, Formal Analysis, Methodology, Software, Writing–review and editing, Writing–original draft. KL: Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Writing–review and editing. PY: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Project administration, Writing–review and editing. HX: Investigation, Visualization, Writing–review and editing. QF: Investigation, Visualization, Writing–review and editing. WZ: Funding acquisition, Project administration, Writing–review and editing. YZ: Investigation, Visualization, Writing–review and editing. WJ: Investigation, Visualization, Writing–review and editing.

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This study was funded by the China Geological Survey (DD20221677-2), CGS Research (JKYQN202307), and the Rizhao City Geological Survey (SDGP371100202102000475).

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Acworth, R. I., Halloran, L. J., Rau, G. C., Cuthbert, M. O., and Bernardi, T. L. (2016). An objective frequency domain method for quantifying confined aquifer compressible storage using Earth and atmospheric tides. Geophys. Res. Lett. 43 (22), 11671–11678. doi:10.1002/2016GL071328

CrossRef Full Text | Google Scholar

An, L. X., Hao, Y. H., Yeh, T. C., Liu, Y., Liu, W. Q., and Zhang, B. J. (2020). Simulation of karst spring discharge using a combination of time-frequency analysis methods and long short-term memory neural networks. J. Hydrology 589, 125320. doi:10.1016/j.jhydrol.2020.125320

Barzegar, R., Fijani, E., Moghaddam, A. A., and Tziritis, E. (2017). Forecasting of groundwater level fluctuations using ensemble hybrid multi-wavelet neural network-based models. Sci. Total Environ. 599, 20–31. doi:10.1016/j.scitotenv.2017.04.189

PubMed Abstract | CrossRef Full Text | Google Scholar

Bredy, J., Gallichand, J., Celicourt, P., and Gumiere, S. J. (2020). Water table depth forecasting in cranberry fields using two decision-tree-modeling approaches. Agric. Water Manag. 233, 106090. doi:10.1016/j.agwat.2020.106090

Chen, C., He, W., Zhou, H., Xue, Y., and Zhu, M. D. (2020). A comparative study among machine learning and numerical models for simulating groundwater dynamics in the Heihe River Basin, northwestern China. Sci. Rep. 10 (1), 3904. doi:10.1038/s41598-020-60698-9

Chen, Z., Xu, H., Jiang, P., Yu, S. E., Lin, G., Bychkoy, I., et al. (2021). A transfer Learning-Based LSTM strategy for imputing Large-Scale consecutive missing data and its application in a water quality prediction system. J. Hydrology 602, 126573. doi:10.1016/j.jhydrol.2021.126573

Chien, S. H., Chi, W. C., and Ke, C. C. (2020). Precursory and coseismic groundwater temperature perturbation: an example from Taiwan. J. Hydrology 582, 124457. doi:10.1016/j.jhydrol.2019.124457

Daliakopoulos, I. N., Coulibaly, P., and Tsanis, I. K. (2005). Groundwater level forecasting using artificial neural networks. J. Hydrology 309 (1), 229–240. doi:10.1016/j.jhydrol.2004.12.001

Gao, X. H., Sato, K., and Horne, R. N. (2020). General solution for tidal behavior in confined and semiconfined aquifers considering skin and wellbore storage effects. Water Resour. Res. 56 (6). doi:10.1029/2020WR027195

Gu, X. F., Sun, H. G., Zhang, Y., Zhang, S. J., and Lu, C. P. (2022). Partial wavelet coherence to evaluate scale-dependent relationships between precipitation/surface water and groundwater levels in a groundwater system. Water Resour. Manag. 36 (7), 2509–2522. doi:10.1007/s11269-022-03157-6

Hakim, W. L., Nur, A. S., Rezaie, F., Panahi, M., Lee, C. W., and Lee, S. (2022). Convolutional neural network and long short-term memory algorithms for groundwater potential mapping in Anseong, South Korea. J. Hydrology Regional Stud. 39, 100990. doi:10.1016/j.ejrh.2022.100990

He, Y. Y., Zhou, Y., Wen, T., Zhang, S., Huang, F., Zou, X. Y., et al. (2022). A review of machine learning in geochemistry and cosmochemistry: method improvements and applications. Appl. Geochem. 140, 105273. doi:10.1016/j.apgeochem.2022.105273

Hikouei, I. S., Eshleman, K. N., Saharjo, B. H., Graham, L. B., Applegate, G., and Cohchrane, M. A. (2023). Using machine learning algorithms to predict groundwater levels in Indonesian tropical peatlands. Sci. Total Environ. 857, 159701. doi:10.1016/j.scitotenv.2022.159701

Hochreiter, S., and Schmidhuber, J. (1997). Long short-term memory. Neural Comput. 9 (8), 1735–1780. doi:10.1162/neco.1997.9.8.1735

Hosono, T., Shibata, Y. T., Wang, C. Y., Manga, M., Rahman, T. M. S., Shimada, J., et al. (2019). Coseismic groundwater drawdown along crustal ruptures during the 2016 Mw 7.0 kumamoto earthquake. Water Resour. Res. 55 (7), 5891–5903. doi:10.1029/2019WR024871

Hu, K. X., Awange, J. L., Kuhn, M., and Saleem, A. (2019). Spatio-temporal groundwater variations associated with climatic and anthropogenic impacts in South-West Western Australia. Sci. Total Environ. 696, 133599. doi:10.1016/j.scitotenv.2019.133599

Hussein, M. E. A., Odling, N. E., and Clark, R. A. (2013). Borehole water level response to barometric pressure as an indicator of aquifer vulnerability. Water Resour. Res. 49 (10), 7102–7119. doi:10.1002/2013WR014134

Kratzert, F., Klotz, D., Brenner, C., Schulz, K., and Herrnegger, M. (2018). Rainfall–runoff modelling using Long Short-Term Memory (LSTM) networks. Hydrology Earth Syst. Sci. 22 (11), 6005–6022. doi:10.5194/hess-22-6005-2018

Lee, E., and Kim, S. (2019). Wavelet analysis of soil moisture measurements for hillslope hydrological processes. J. Hydrology 575, 82–93. doi:10.1016/j.jhydrol.2019.05.023

Lee, S. H., Ha, K., Hamm, S. Y., and Ko, K. S. (2013). Groundwater responses to the 2011 tohoku earthquake on jeju island, korea. Hydrol. Process. 27 (8), 1147–1157. doi:10.1002/hyp.9287

Long, D., Yang, W. T., Scanlon, B. R., Zhao, J. S., Liu, D. G., Burek, P., et al. (2020). South-to-North Water Diversion stabilizing Beijing’s groundwater levels. Nat. Commun. 11 (1), 3665. doi:10.1038/s41467-020-17428-6

Mao, H. R., Wang, G. C., Liao, F., Shi, Z. M., Huang, X. J., Li, B., et al. (2022). Geochemical evolution of groundwater under the influence of human activities: a case study in the southwest of Poyang Lake Basin. Appl. Geochem. 140, 105299. doi:10.1016/j.apgeochem.2022.105299

Massei, N., Dupont, J. P., Mahler, B. J., Laignel, B., Fournier, M., Valdes, D., et al. (2006). Investigating transport properties and turbidity dynamics of a karst aquifer using correlation, spectral, and wavelet analyses. J. Hydrology 329 (1-2), 244–257. doi:10.1016/j.jhydrol.2006.02.021

Miyakoshi, A., Taniguchi, M., Ide, K., Kagabu, M., Hosono, T., and Shimada, J. (2020). Identification of changes in subsurface temperature and groundwater flow after the 2016 Kumamoto earthquake using long-term well temperature-depth profiles. J. Hydrology 582, 124530. doi:10.1016/j.jhydrol.2019.124530

Mohammed, H., Michel, H. T., and Seidu, R. (2022). Emulating process-based water quality modelling in water source reservoirs using machine learning. J. Hydrology 609, 127675. doi:10.1016/j.jhydrol.2022.127675

Nelson, D. B., Basler, D., and Kahmen, A. (2021). Precipitation isotope time series predictions from machine learning applied in Europe. Proc. Natl. Acad. Sci. 118 (26), e2024107118. doi:10.1073/pnas.2024107118

Nourani, V., Baghanam, A. H., Adamowski, J., and Kisi, O. (2014). Applications of hybrid wavelet-Artificial Intelligence models in hydrology: a review. J. Hydrology 514, 358–377. doi:10.1016/j.jhydrol.2014.03.057

Nourani, V., and Mousavi, S. (2016). Spatiotemporal groundwater level modeling using hybrid artificial intelligence-meshless method. J. Hydrology 536, 10–25. doi:10.1016/j.jhydrol.2016.02.030

Qu, S., Shi, Z. M., Wang, G. C., Xu, Q. Y., Han, J. Q., and Jiaqian, H. (2020). Using water-level fluctuations in response to Earth-tide and barometric-pressure changes to measure the in-situ hydrogeological properties of an overburden aquifer in a coalfield. Hydrogeology J. 28, 1465–1479. doi:10.1007/s10040-020-02134-w

Qu, S., Shi, Z. M., Wang, G. C., Zhang, H., Han, J. Q., Liu, T. X., et al. (2021). Detection of hydrological responses to longwall mining in an overburden aquifer. J. Hydrology 603, 126919. doi:10.1016/j.jhydrol.2021.126919

Rodrigues, E., Gomes, Á., Gaspar, A. R., and Henggeler, A. C. (2018). Estimation of renewable energy and built environment-related variables using neural networks – a review. Renew. Sustain. Energy Rev. 94, 959–988. doi:10.1016/j.rser.2018.05.060

Shi, Z. M., Zhang, S. C., Yan, R., and Wang, G. C. (2018). Fault zone permeability decrease following large earthquakes in a hydrothermal system. Geophys. Res. Lett. 45 (3), 1387–1394. doi:10.1002/2017GL075821

Sun, J. C., Hu, L. T., Li, D. D., Sun, K. N., and Yang, Z. Q. (2022). Data-driven models for accurate groundwater level prediction and their practical significance in groundwater management. J. Hydrology 608, 127630. doi:10.1016/j.jhydrol.2022.127630

Tawara, Y., Hosono, T., Fukuoka, Y., Yoshida, T., and Shimada, J. (2020). Quantitative assessment of the changes in regional water flow systems caused by the 2016 Kumamoto Earthquake using numerical modeling. J. Hydrology 583, 124559. doi:10.1016/j.jhydrol.2020.124559

Tinungki, G. M.Iop (2019). “The analysis of partial autocorrelation function in predicting maximum wind speed, 1st International Conference on Global Issue for Infrastructure, Environment and Socio-Economic Development,” in IOP Conference Series-Earth and Environmental Science , Bogar, West Java , 29 August 2019 .

Google Scholar

Tokunaga, T. (1999). Modeling of earthquake-induced hydrological changes and possible permeability enhancement due to the 17 January 1995 Kobe Earthquake, Japan. J. Hydrology 223 (3-4), 221–229. doi:10.1016/S0022-1694(99)00124-9

Vittecoq, B., Foritn, J., Maury, S., and Violette, S. (2020). Earthquakes and extreme rainfall induce long term permeability enhancement of volcanic island hydrogeological systems. Sci. Rep. 10 (1), 20231. doi:10.1038/s41598-020-76954-x

Vu, M. T., Jardani, A., Massei, N., and Fournier, M. (2021). Reconstruction of missing groundwater level data by using Long Short-Term Memory (LSTM) deep neural network. J. Hydrology 597, 125776. doi:10.1016/j.jhydrol.2020.125776

Wang, C. Y., Doan, M. L., Xue, L., and Barbour, A. J. (2018). Tidal response of groundwater in a leaky aquifer-application to Oklahoma. Water Resour. Res. 54 (10), 8019–8033. doi:10.1029/2018WR022793

Wang, C. Y., Liao, F., Wang, G. C., Qu, S., Mao, H. R., and Bai, Y. F. (2023). Hydrogeochemical evolution induced by long-term mining activities in a multi-aquifer system in the mining area. Sci. Total Environ. 854, 158806. doi:10.1016/j.scitotenv.2022.158806

Wang, C. Y., Wang, L. P., Manga, M., Wang, C. H., and Chen, C. H. (2013). Basin-scale transport of heat and fluid induced by earthquakes. Geophys. Res. Lett. 40 (15), 3893–3897. doi:10.1002/grl.50738

Wunsch, A., Liesch, T., and Broda, S. (2021). Groundwater level forecasting with artificial neural networks: a comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX). Hydrol. Earth Syst. Sci. 25 (3), 1671–1687. doi:10.5194/hess-25-1671-2021

WWAP (2015). “Water for a sustainable world, no. 6.2015,” in The united nations world water development report (Paris, France: United Nations World Water Assessment Programme – UNESCO ).

Xiao, Y., Hao, Q. C., Zhang, Y. H., Zhu, Y. C., Yin, S. Y., Qin, L. M., et al. (2022). Investigating sources, driving forces and potential health risks of nitrate and fluoride in groundwater of a typical alluvial fan plain. Sci. Total Environ. 802, 149909. doi:10.1016/j.scitotenv.2021.149909

Xiao, Y., Shao, J. L., Frape, S. K., Cui, Y. L., Dang, X. Y., Wang, S. B., et al. (2018). Groundwater origin, flow regime and geochemical evolution in arid endorheic watersheds: a case study from the Qaidam Basin, northwestern China. Hydrol. Earth Syst. Sci. 22 (8), 4381–4400. doi:10.5194/hess-22-4381-2018

Yan, R., Woith, H., Wang, R. J., and Wang, G. C. (2017). Decadal radon cycles in a hot spring. Sci. Rep. 7, 12120. doi:10.1038/s41598-017-12441-0

Yan, X., Shi, Z. M., Wang, G. C., Zhang, H., and Bi, E. P. (2021). Detection of possible hydrological precursor anomalies using long short-term memory: a case study of the 1996 Lijiang earthquake. J. Hydrology 599, 126369. doi:10.1016/j.jhydrol.2021.126369

Yan, X., Shi, Z. W., Zhou, P. P., Zhang, H., and Wang, G. C. (2020). Modeling earthquake-induced spring discharge and temperature changes in a fault zone hydrothermal system. J. Geophys. Res. Solid Earth 125 (7), e2020JB019344. doi:10.1029/2020JB019344

Yu, S. W., and Ma, J. W. (2021). Deep learning for geophysics: current and future trends. Rev. Geophys. 59 (3), e2021RG000742. doi:10.1029/2021RG000742

Yu, X., Wang, Y. H., Wu, L. F., Chen, G. H., and Qin, H. (2019). Comparison of support vector regression and extreme gradient boosting for decomposition-based data-driven 10-day streamflow forecasting. J. Hydrology 582, 124293. doi:10.1016/j.jhydrol.2019.124293

Zhang, H., Shi, Z. M., Wang, G. C., Yan, X., Liu, C. L., Sun, X. L., et al. (2021). Different sensitivities of earthquake-induced water level and hydrogeological property variations in two aquifer systems. Water Resour. Res. 57 (5), e2020WR028217. doi:10.1029/2020WR028217

Zhang, J. F., Zhu, Y., Zhang, X. P., Ye, M., and Yang, J. Z. (2018). Developing a Long Short-Term Memory (LSTM) based model for predicting water table depth in agricultural areas. J. Hydrology 561, 918–929. doi:10.1016/j.jhydrol.2018.04.065

Zhang, M. H., Li, X., and Wang, L. L. (2019). An adaptive outlier detection and processing approach towards time series sensor data. IEEE Access 7 (175192), 175192–175212. doi:10.1109/ACCESS.2019.2957602

Zhang, S. C., Shi, Z. M., Wang, G. C., Yan, R., and Zhang, Z. C. (2020). Groundwater radon precursor anomalies identification by decision tree method. Appl. Geochem. 121, 104696. doi:10.1016/j.apgeochem.2020.104696

Zhang, S. C., Shi, Z. M., Wang, G. C., Yan, R., and Zhang, Z. C. (2022a). Application of the extreme gradient boosting method to quantitatively analyze the mechanism of radon anomalous change in Banglazhang hot spring before the Lijiang Mw 7.0 earthquake. J. Hydrology 612, 128249. doi:10.1016/j.jhydrol.2022.128249

Zhang, X. Y., Dong, F., Chen, G. Q., and Dai, Z. X. (2022b). Advance prediction of coastal groundwater levels with temporal convolutional and long short-term memory networks. Hydrol. Earth Syst. Sci. 2022, 83–96. doi:10.5194/hess-27-83-2023

Keywords: long short term memory neural network, coastal groundwater levels, groundwater regime, groundwater withdrawal, machine learning

Citation: Guo B, Zhang S, Liu K, Yang P, Xing H, Feng Q, Zhu W, Zhang Y and Jia W (2023) Prediction of groundwater level under the influence of groundwater exploitation using a data-driven method with the combination of time series analysis and long short-term memory: a case study of a coastal aquifer in Rizhao City, Northern China. Front. Environ. Sci. 11:1253949. doi: 10.3389/fenvs.2023.1253949

Received: 07 July 2023; Accepted: 13 October 2023; Published: 08 November 2023.

Reviewed by:

Copyright © 2023 Guo, Zhang, Liu, Yang, Xing, Feng, Zhu, Zhang and Jia. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Shouchuan Zhang, [email protected] ; Kai Liu, [email protected]