Distractions towards the Use of Learning Management System (LMS): Evidence from Students during the Covid-19 Lockdown in a Developing Country Context

The study is one of the first of its type to investigate the distractions towards the use of Learning Management System (LMS) during the Covid-19 shutdown and how it affected students’ academic performance. The study sampled 456 learners who had returned to school following the Covid-19 lockout to resume their academic schedule. Data collected was evaluated using the SmartPLS tool for Structural Equation Modeling (SEM). Environmental distraction, technological distraction, and learning family distraction all had a detrimental impact on students' LMS usage, as found in the study. The use of the LMS was also found to have a detrimental impact on students' academic performance. The study's findings will assist lecturers, teachers or instructors in being courteous, as well as knowing the sort of evaluation to be utilized to assess students pursuing distance learning, as some of these distractions are likely to occur at home. As a result, the study contributes to the paucity of previous research on e-learning distractions in developing-country environments. use of the LMS during the Covid lockdown. quite some study on e-learning, this study explores the factors that interfere with students’ e-learning at home. The study proposed three destructive variables and validated one variable in the technology acceptance model. The study proposed model was seen to clarify 55.9% of the variability on students’ academic performance relating to E-learning on Covid-19 lockdown. The study findings would also help teachers to be considerate and also know the type of assessment to be used to assess students offering distance there are to encounter some of these distractions at home. The study provided a significant outcome but it is not limitation. The sample size was composed purposively to a specific target population. the findings might not be other and country contexts. Further studies can inculcate other factors to moderate or mediate the proposed model.

for online training for both teaching staff and administrative staff on effective use of the university LMS. Students were trained by their instructors using other social media applications to equip them on LMS usage. Learners were tasked to take quizzes, mid-semester examination as well as their final examinations online. Moodle is the name of the E-learning platform under research in this study. University instructors utilize Moodle to manage courses, assignments, class assessments, facilitate discussion forums and coordinate students' interim. Although the implementation and usage were successful, some distractions didn't enable some students to enjoy the successful usage of the Moodle. Students' academic performance using the LMS was seen to have declined as compared to the normal face to face traditional learning. The motivation behind this study is as a result of students' poor academic performance during the covid-19 lockdown using the Moodle LMS. A brief discussion with some students revealed that environmental noise, learning family issues, domestic problems and phone notifications (apps, network operators, incoming calls etc.) were some of the challenges they faced during the Covid-19 lockdown when using the LMS. The study, therefore decide to explore these concerns by students on a large scale of students to find out if indeed these distractions affected their LMS usage during the lockdown. The outcomes of this study will add to the body of knowledge and literature by presenting empirical facts that explain the characteristics that inhibit university students' successful adoption and usage of Moodle LMS at home.

Factors that influence E-learning Distraction 2.1 Technology Notification
Technological devices play a vital role in E-learning adoption. Many students access their online materials using either their laptops, iPads, mobile phones etc. In the Ghanaian economy, students have been observed to utilize their mobile phones more than any other device to access online content. They use their mobile phones to take quizzes, exams, and download homework. Students who use mobile phones receive notifications when new messages, emails, social media updates, and other events are posted. Many students use their mobile phones as a Hotspot to share the internet on their laptops or tablets. It is realized that notification from apps, SMS by the network operator as well as incoming calls has become a destructive motion that affects students' online quiz (e.g. Answering Multiple Choice Questions (MCQ) online). While the impact of notification at the workplace environment has been extensively researched (Kanjo et al. 2017;Ho & Intille 2005;Iqbal & Horvitz 2010;Leiva et al. 2012;Hernández-Reyes et al. 2020), nature of smartphone notification and their impact on students using Elearning are little understood. Pielot et al. (2014) emphasized the detrimental consequences of notifications on job efficiency. The authors reported dealing with an average of 63.5 alerts each day, with the majority coming from messengers and email. However, frequent notification is likely to affect job performance (Hernández-Reyes et al. 2020). Furthermore, Leiva et al. (2012) believe that phone calls interrupting app usage dramatically increase the time a user spends finishing a particular task. Ho & Intille (2005) studied how notifications may be given during the transition from one physical activity to another. Their findings show that when alerts are presented between two physical activities, such as either sitting or walking, they are regarded more favourably. However, it is clear that people often find it difficult to resume prior work after being stopped by phone calls, texts, or notifications, or a conversation with anyone (Czerwinski et al. 2004). To summarize, the majority of previous research has been on desktop and office environments. While previous research has looked at notifications in the setting of mobile devices, it appears that none of these studies has looked at how notifications affect students' e-learning utilization via their cell phones. What is lacking in the literature is how notifications on the phones distract students when taking their exams online, specifically answering test questions. Therefore, the study hypothesized that: H1: Technology notification had a negative effect on students Moodle usage during Covid-19 Lockdown

Learning-Family Distraction
The COVID-19 epidemic has created an exceptional scenario in which many individuals have been forced to learn from home (Oakman et al. 2020). Students who learn from home typically tend to learn fewer hours due to more family distractions than being in school. The study characterized the learning-family distraction as domestic duties, errands and other activities at home that impede and interfere with students' attention to study at home. In some African countries like Ghana, Togo, Nigeria etc., household duties are often undertaken by females in the family, whereas outside tasks are performed by males in the family (Kissi et al. 2018;Annor 2015). While these activities may appear to be normal in the family's daily routines, they are unplanned (Kissi et al. 2018) and so take up time that could be used for school activities. Furthermore, too many domestic actives have been considered as one of the hurdle that affect students in performing classroom activities as cited by Mordkowitz & Ginsburg (1987). The majority of families in our culture do not appear to prioritize their children's education. Some parents appear to have erroneous beliefs about their children's academic success. More often they feel reluctant to fulfil their duties and encouraged in their children's academic achievement. Some individuals believe that widespread failure or success in schools may be attributed to teachers and school administrators. During the covid-19 lockdown, a preliminary investigation declared that many students were not happy taking online tests or quizzes at home due to learning family conflict (e.g. too much domestic activities, too many errands, siblings playing and making loud noise etc.) caused by their family members. However, in most African communities, it is quite unusual to see an individual occupying his or her room in the family house. More often, you occupy a room with your siblings or with any other relatives. This level of occupancy sometimes distracts student learners when there is a nuisance. Sometimes children are being called by their parents to perform a task or send them for some errands, even though the parents are aware that their children are studying. Although learning from home allow you to "have more time with family and less commuting time, it also has several known demerits (Hoffman et al. 2020;Bergefurt et al. 2021). Some students saw this as a major challenge faced during the lockdown, which they believe affected their academic performance. Although earlier research has shown that distractions in the employee workplace can have a detrimental impact on employees' well-being (Bergefurt et al. 2021), little is known about distractions while learning at home and how it influences academic performance, since there is paucity of literature exploring how family distractions affect children learning at home. Hence, the following hypothesis was derived. H2: learning Family-Distraction had a negative effect on students Moodle usage during Covid-19 Lockdown

Environmental Distraction
The home environment is more crucial than anything else for students' academic performance (Younas et al. 2021;Bergefurt et al. 2021;Crawford & Zygouris-Coe 2006). However, encouragement from the environment may stimulate learning and improve a student's capabilities, whilst discouragement from the environment just depresses a student's talent. One of the most critical elements influencing student performance and academic accomplishment is the home environment. When students experience good and serene environments without distraction it encourages and motivates them to study much better. Moreover, some environmental noise can divert learners focus away from the focal task (Hughes 2014). Environmental distractions in this context are defined as the noise around and outside the learner's environment that distracts him or her when learning. This could be music played by neighbours, aircraft noise, road traffic noise, traders who move around the environment advertising their product verbally etc. Obeta (2014) emphasized that the location of the home of the learners have an impact on their academic performance. Wong et al. (2002) investigated the various noise sources that affect people in the community. It was realized in the authors' study that among the various environmental noise, the most annoying one was traffic noises, which is followed by construction, aircraft, neighbouring and industrial noise. This kind of noise sometimes affects students, when they are writing online quiz specifically multiple-choice questions (MCQ). Although there are studies that have explored some environmental distractions in schools and offices (Lundquist et al. 2000;Lee & Brand 2010;Woolner & Hall 2010), the environmental distraction towards e-learning has been seen to receive little concern. It is of interest to find out how this distraction affected students' e-learning adoption during the Covid-19 lockdown.

LMS Application Usage
Actual system usage is a measurement of an individual's behaviours or actions when utilizing a certain technology, such as online learning (Hassanzadeh et al. 2012). In accordance with Davis, Bagozzi, & Warshaw (1989), an "individual's behavioural intention leads to actual behaviour or system utilization". Mohammadi (2015) corroborated this in his research study on e-Learning systems. Suki (2011) have also corroborated this in various research. The Technology Adoption Model employs actual usage to reflect a self-report measure of time or frequency of adopting the system, whereas behavioural intention to use measures the chance that a person will adopt the system (Davis et al., 1989). Several studies have demonstrated that there is theoretical and empirical support for individual system usage (Opoku, Pobee & Okyerih 2020;Chong et al. 2012;Suki 2011;Vijayasarathy 2004). Actual system usage is a measurement of an individual's behaviours or actions when utilizing certain technology. Davis et al. (1989), emphasize that an individual's behavioural intention leads to real behaviour or actual system usage. Thus, an individual behavioural intention will lead to their actual behaviour. This has been confirmed in the studies of Mohammadi (2015), Chong et al. (2012) and Suki (2011). It is therefore hypothesized that; H4: Moodle usage had a negative effect on student performance during Covid-19 Lockdown

Student Academic Performance
Students' academic performance is a vital key to improving student learning. It is influenced by socioeconomic, personal factor and environmental factors. However, recognizing these variables and their influence on student performance might aid in their management (Shahibi et al. 2017;Baradwaj & Pal 2012). When students can perform very well, they obtain much desire in studying hard. However, the performance of students towards academic work may not only be affected by their hard work but also parent involvement (Kweon et al. 2017). Some studies have also indicated that the learning environment sometimes can contribute to low students' academic performance if such an environment is not conducive for students (e.g. Topor 2010; Kweon et al. 2017). Other studies have also indicated that student academic performance is influenced by both internal and external factors Journal of Education and Practice www.iiste.org ISSN 2222-1735 (Paper) ISSN 2222-288X (Online) Vol.12, No.35, 2021 23 (Mushtaq & Khan 2012). Internal factors constitute learning facilities, class environment, technology usage, instructors' role in the class and the kind of examination systems that exist whereas the external factors include family problems, financial, social and accommodation problems (Mushtaq & Khan 2012;Kweon et al. 2017;Baradwaj & Pal 2012). Although there are some studies on students' academic performance, it seems none of these studies has investigated covid-19 lockdown distractions on student performance. For example, in a study by Harb and El-Shaarawi (2006), the authors found English competence as an important factor that affects students' academic performance. In a study by Shahibi & Rusli (2017), the authors found that using internet media for instruction improves students' academic achievement. According to Adeoye et al. (2020), e-learning comes with some drawbacks with sometimes demote it implementation. To the best of my knowledge, there have been scanty reports presenting how e-learning has been adopted during the Covid-19 epidemic. This gives room for more academic research.

Study Method and Material
The study examined the distractions that students encountered during the covid-19 lock and how they influenced their academic performance. The Moodle is one of the most used E-learning applications by many institutions. The system allows lecturers or instructors to mount their courses, upload course materials, and conduct quizzes and exams as well. Students use this system to participate in group discussions, submit homework, and connect with their instructors or lecturers. To achieve the study's goal, a cross-sectional survey methodology was adopted. After students returned to school following the covid-19 lockdown, a sample of 480 questionnaires were distributed. However, 456 questionnaires were retrieved and used for the study analysis. Likert scale questions were used to illustrate how much participants agreed or disagreed with the statements used to test the variables. Items measuring the technology distraction and environmental distractions were developed based on students' experience and concerns. These items were modified by two scholars who have in-depth knowledge and has been practicing e-learning. Items measuring learning family distraction were adapted from Kissi et al. (2018) and modified. The gathered data was evaluated using the SmartPLS tool for Structural Equation Modeling (SEM). The SEM is a versatile and successful analytical tool for identifying structures and their interactions based on observational data. The study presented data on descriptive statistics of the demographic variable, evaluated both the measurement model (i.e. reliability, validity, normality and collinearity test) and the structural model (i.e. coefficient of determination, model predictive relevance, hypothesis). The items used were written to fit the setting of the study.

Results Presentation
The demographic information of the subjects was initially reported in the research. The study also tested the model's applicability by checking the data's normality, reliability and validity. Furthermore, the structural model was evaluated to assess the model's predictive ability, accuracy, and correlations among the components. The study consists of 456 university students as already presented in the methodology. According to Table 1, more than half of those who participated in this study (61.8%) were males, while the rest (38.2%) were females. Students aged 18 to 24 were found to be the most common responses (91.2%). In terms of academic Year, 38.4% were determined to be at the first year, 22.6 % at second year, 20.0 % at the third year, and 19.1 % at fourth year. www.iiste.org ISSN 2222-1735 (Paper) ISSN 2222-288X (Online) Vol.12, No.35, 2021

Normality Test and Collinearity
A normality test was run to see if the data warranted further investigation. The skewness-Kurtosis approach was used to determine the data set's normality (Byrne 2013). The findings were discovered to be on the predicted scales. As shown in Table 2, all skewness values between 2 and + 2 and kurtosis values between -7 and + 7 validated the data's normality (Byrne 2013). However, when data are collected from multiple sources, there can be a strong correlation between items or indicators, which can lead to the issue of multicollinearity. To prevent this, values for Variance Inflation Factor (VIF) should be less than 5 (VIF < 5) (Kim 2019). From the analysis, it was seen that all values of VIF were less than 5, which indicate no issue of multicollinearity. Table 2 presents the findings.

Measurement Model Evaluation
The study evaluated the measurement model using Confirmatory Factor Analysis. All reliability and validity values were calculated. Factor loadings larger than 0.6 were considered (Gefen & Straub 2005). Cronbach's alpha was used to calculate the internal consistency of the constructs, with suggested values greater than 0.5. ( Hu & Bentler 1998;Hasan & Boa 2020;Hair et al. 2010). The composite reliability of the constructs was also found to be more than the required level of 0.70, demonstrating exceptional construct dependability (Fornell & Larcker 1981). Convergent validity was also calculated using AVE, which implies that each item measures what it was designed to measure. The Average Variation Extracted (AVE) criterion was more than 0.50, indicating that the measurement error was smaller than the structural observed variance. Convergent validity was good, as indicated in table 3, because all AVE values were more than 0.5 according to Henseler et al. (2015). The Path Analysis Diagram is also shown in Figure 1.

Discriminant validity
Discriminant validity, which describes how one concept differs (is distinct) from the others, was also assessed. Fornell-Larcker measure was used to evaluate discriminant validity (Henseler et al. 2015). Discriminant validity is deemed to be reached if the square root of AVE is larger than the inner correlation of the factors. Table 4 shows acceptance of discriminant validity among components. The Heterotrait-Monotrait ratio (HTMT) criteria is another way to support discriminant validity. All correlation values must be less than 0.900 for HTMT to be accomplished (Henseler et al. 2016). Table 5 demonstrated that the discriminant validity was excellent.

Structural Model Evaluation
The coefficient of determination (R 2 ) is commonly used to assess the prediction capacity of a structural model. It illustrates how the independent variable predicts the variation in the dependent variable. The coefficient of determination was examined at two different levels ( Table 6). The first explains the variation in Moodle usage (R 2 MOU), while the second explains the variation in Student Academic Performance (R 2 ACP). R 2 MOU = 0.209 implies that technological distraction, environmental distraction, and learning family distraction account for 20.9% of the variation in student Moodle usage. R 2 ACP = 0.559 implies that Moodle usage explains 55.9% of the variation in student academic achievement. The Stone-Geisser Indicator (Q 2 ) was also used to measure the model's predictive significance (Henseler et al. 2015). An indicator value larger than zero indicates a high prediction quality (Henseler et al. 2015). The values of Q 2 in Table 6 suggest that the model is accurate and that the constructs are crucial for general model tuning. 0.587 0.392 To evaluate the significance level of the latent variables, the construct relationship was established using the Bootstrapping procedure in Partial Least Square. Each of the four hypotheses was examined (H1 to H4, see Table  7). During the Covid-19 Lockdown, technology distraction had a substantial influence on students' Moodle usage (β = -0.182, t-value = 2.231, P < 0.039). As a result, H1 is supported. Learning family distraction had a substantial detrimental influence on students' Moodle usage during the Covid-19 Lockdown (β = -0.210, t-value = 2.552, P < 0.011). As a result, H2 is supported. Environmental distraction had a negative significant effect on students Journal of Education and Practice www.iiste.org ISSN 2222-1735 (Paper) ISSN 2222-288X (Online) Vol.12, No.35, 2021 Moodle usage during Covid-19 lockdown (β = -0.252, t-value = 3.442, P < 0.001). Hence, H3 is supported. Finally, Moodle usage by students showed a negative influence on academic performance (β = -0.447, t-value = 6.063, P < 0.000). Hence, H4 is supported.  (Kanjo et al. 2017;Hernández-Reyes et al. 2020).
The study also verified the second hypothesis (H2), which stated that learning family distraction has a negative effect on students' use of Moodle LMS. It is often understood that the majority of families in several African societies do not appear to prioritize their children's education. Some parents appear to have incorrect assumptions about their children's academic achievement. More often than not, they are hesitant to carry out their responsibilities while being proud of their children's academic performance. Some students have been seen to engage in a variety of household tasks that consume a significant portion of their studying time at home. This data lends credence to Kissi et al. (2018) work on learning family conflict. The hypothesis (H3) of environmental distraction had a negative significant influence on students' Moodle usage. It was discovered that some of the students were living in an unsuitable environment for studying. They have no choice but to manage in such a setting. Encouragement from the environment, on the other hand, may accelerate learning and develop a student's skills, but discouragement from the environment just depresses a student's potential. Loud noise from neighbouring residences, traffic noise, and noise from vendors advertising their products while wandering in the area have all been seen to have an impact on students studying at home. These findings back with prior research indicating that the environment is more important than everything else for students' academic performance (Younas et al. 2021;Bergefurt et al. 2021). Finally, the findings validated hypothesis 4, which stated that students' use of Moodle had a negative effect on academic performance. External variables (e.g., environmental, learning family distractions, etc.) that inhibit continuous use of the Moodle LMS might explain the detrimental effect of system utilization on academic performance. This type of discouragement might undoubtedly have a negative effect on performance. Furthermore, there are worries that learners or students living in rural regions may struggle to fulfil the expectations of e-learning due to a lack of consistent internet connectivity, which plays a critical role in maintaining the proper operation of the LMS and may impair academic achievement.

Implication, Limitation and Future Study
This study is one of the first kind to investigate destructive factors that hinder the use of the LMS during the Covid -19 lockdown. Although there is quite some study on e-learning, this study explores the factors that interfere with students' e-learning at home. The study proposed three destructive variables and validated one variable in the technology acceptance model. The study proposed model was seen to clarify 55.9% of the variability on students' academic performance relating to E-learning on Covid-19 lockdown. The study findings would also help teachers or instructors to be considerate and also know the type of assessment to be used to assess students offering distance learning since there are likely to encounter some of these distractions at home. The study provided a significant outcome but it is not without limitation. The sample size was composed purposively to a specific target population. However, the findings might not be generalized to other developed and developing country contexts. Further studies can inculcate other factors to moderate or mediate the proposed model.

Conclusion
Today, E-learning solutions are crucial in assisting schools to complete their studies throughout the Covid-19 epidemic. Schools that previously did not provide online education have all recognized the benefits of online education and have enrolled in it. During the Covid-19 pandemic, several researchers took use of the opportunity to conduct more studies on E-learning adoption. This study looked at the distractions that hampered students' elearning during the Covid-19 lockout. The empirical investigation revealed students' low academic performance during the covid-19 lockout as a result of environmental, technological, and learning family distractions that influenced their use of the Moodle LMS. These characteristics should be taken into account when evaluating students' LMS usage because it has a detrimental impact on their academic performance.