In many computer-based learning environments (CBLEs), learners of all ages struggle to use critical cognitive and metacognitive self-regulation abilities (Azevedo, 2015). In this situation, choosing what to learn, when to learn it, and how long to study it becomes more crucial (You, 2015). Time usage, goal-setting, self-monitoring, self-reactions, self-efficacy, and motivation are the six self-regulatory processes that underlie all other activities (Zimmerman & Risemberg, 1997). Time management, motivation, and perceived self-efficacy play the most significant roles among these processes (Zimmerman, 1998). According to several studies (Kirk et al., 2013; Visser et al., 2015), it appears that time management plays a significant role in educational outcomes from K–12 to higher education. Given the intricacy of the self regulation construct, the researcher of this study concentrated on one of its dimensions in this paper: time management.

Numerous studies discuss the value of time management and learning, emphasizing both the quantity and quality of the learning time students spend on learning (Balkis, 2011). Several of these studies concentrate on academic procrastination, which is defined as the propensity to put off or even avoid performing an activity that is within one's control (Gafni & Geri, 2010). Procrastination, a minor but significant aspect of self-regulation, is especially examined in this study in an effort to understand its connection to student’s success in CBLEs. Specifically, this study made use of the existing model of Strunk (2012) for the study of procrastination and timely engagement because of the significance of the time that students spend learning and engaging in academic procrastination. According to model proposed by Strunk (2012), procrastination is on one side, with two distinct motivating inclinations. Procrastination for the sake of strategic advantage and an increase in work quality is defined as the procrastination approach. The avoidant coping type of procrastination is procrastination-avoidance. On the other hand, a time engagement approach is characterized by getting started on tasks right away to produce higher grades. To avoid the anxiety of failure that arises from delaying starting tasks, one would interact with them as soon as possible. This strategy is known as timely engagement-avoidance.

Although some research findings emphasize the negative effects of procrastination, others have identified a profile of active procrastination that corresponds to students who choose to delay work in order to achieve a superior performance (Choi & Moran, 2009; Kim & Seo, 2013). This contradiction makes it even more important to contextualize the research of this particular phenomenon in CBLEs.

CBLEs are prepared to gather significant amounts of data through user-machine interaction. Particularly, LMSs gather student data that, when examined properly, can give educators and researchers the knowledge they need to assist and continuously enhance the learning process (Paule-Ruiz et al., 2015). Modular Object Oriented Developmental Learning Environment (Moodle), a free LMS that enables the design of potent, adaptable, and interesting online courses and experiences, is one of the most popular (Rice, 2006).

This study aims to examine relationships among students course achievement and several time management-related features. In this regard, the following research questions are asked to answer:

1)     How are the undergraduate students grouped based on the variables such as students course grade,  time management-related features (the time differences between first access of students to course assignments and release dates of assignments,  first access and submission dates, and submission dates and due dates)?

2)     Are there statistically significant differences among different latent profiles regarding course grade and time management-related features?

Answers to these questions offer insight to scheduling and planning of the assessment methods in online courses.

Participants This study included data from two sections of an undergraduate course designed for pre-service teachers. Both sections (n1 = 44 and n2 = 44) were taught by the same instructor during the Fall 2021- 2022 semester. Due to COVID-19 pandemic, all course sections were delivered online. After the students completed the course, the log data and student grades from the LMS for each section were extracted. Extraction of the Variables Students' final course grade was determined based on the weighted average of three assignments. In this study, time-related features associated with students' course performance were extracted from the LMS log data. Specifically, the release date (i.e., when the assignment was made available to students), submission date, and due date of each assignment were used. Thus, the following time-related features from the LMS log data for each assignment were extracted: the time difference between first access and release dates, the time difference between first access and submission dates, and the time difference between submission dates and due dates. This feature extraction process yielded nine time-related features. Analysis were performed after the features were combined by calculating the means of features. Data Analysis After removing the missing and extreme cases, 58 students were included in this study. In terms of assumptions, the homogeneity of the variance assumption was violated. Regarding the first research question of this study, latent profile analysis (LPA) was conducted. LPA is a statistical procedure in which continuous latent indicators are utilized while performing latent class analysis (Muthén & Muthén, 1998-2017). Accordingly, Akaike Information Criterion (AIC), the Bootstrap Likelihood Ratio Test (BLRT), Bayesian information criterion (BIC), and the entropy value were used. Smaller values of AIC and BIC indicate a better model fit. Also, an insignificant (p > 0.05) BLRT indicates that adding more profiles into the model does not improve the model. Additionally, a value closer to 1.0 for the entropy values indicates a better decision on the number of profiles to include (Wang & Wang, 2020). Regarding second research question of this study, due to the violation of homogeneity of variance assumption, Kruskall Wallis Test which is a non-parametric version of the one-way ANOVA was performed to compare the profiles regarding the variables addressed in this study. LPA was conducted with “tidyLPA” package (Rosenberg et al., 2018) in R, and SPSS software was used for Kruskall Wallis Test.

After LPA performed, model fit statistics were obtained to determine the optimal number of classes. According to model-fit statistics, the model that best fit the data was found to be the four-class model. General patterns of the profiles were plotted. Accordingly, those who are classified in Profile 1 have “timely-engagement avoidance” according to the model proposed by Strunk (2012). Additionally, those students classified in Profile 2 are called as students who “have timely engagement approach”. It was also found that Profile 3 included the students who had procrastination avoidance. Lastly, students classified in Profile 4 were found to have procrastination approach. The Kruskall Wallis test results, which were conducted to determine whether the students gathered under these profiles differ in terms of course grade and time-related features were showed that the students in the four profiles differ significantly regarding course grade and time-related features. When the results of the Post-Hoc comparison made to determine which profiles caused this difference were examined, it was found that Profile 1 differed from Profile 2, Profile 2 differed from Profile 3 and lastly Profile 3 differed from Profile 4 regarding course grade. Additionally, regarding the interval between the date an assignment is released and the moment students have access to it, Profile 1 was different from Profiles 2, 3, and 4. Profiles 2, 3, and 4 were also different from Profile 1. Moreover, in terms of time difference between submission dates and due dates of the assignments, Profile 1 differed from Profile 4, Profile 2 differed from Profile 4, and Profile 3 differed from Profile 4. Also, regarding the time interval between first access to assignments and assignment submission deadlines, Profiles 1 and 2 and 3 differed from Profile 4. Lastly, it was found that Profile 2 and Profile 3 differed from Profile 4.

Azevedo, R. (2015). Defining and measuring engagement and learning in science: Conceptual, Theoretical, methodological, and analytical issues. Educational Psychologist, 50, 84–94. https://doi.org/10.1080/00461520.2015.1004069 Balkıs, M. (2011). Academic efficacy as a mediator and moderator variable in the relationship between academic procrastination and academic achievement. Eurasian Journal of Educational Research, 45, 1–16. Choi, J. N., & Moran, S. V. (2009). Why not procrastinate? Development and validation of a new active procrastination scale. The Journal of Social Psychology, 149(2), 195–212. https://doi.org/10.3200/SOCP.149.2.195-212 Gafni, R., & Geri, N. (2010). Time management: Procrastination tendency in individual and collaborative tasks. Interdisciplinary Journal of Information, Knowledge, and Management, 5, 115–125. https://doi.org/10.28945/1127 Kim, E., & Seo, E. H. (2013). The relationship of flow and self-regulated learning to active procrastination. Social Behavior and Personality An International Journal, 41(7), 1099–1113. https://doi.org/10.2224/sbp. 2013.41.7.1099 Kirk, D., Oettingen, G., & Gollwitzer, P. M. (2013). Promoting integrative bargaining: mental contrasting with implementation intentions. International Journal of Conflict Management, 24(2), 148–165. https://doi.org/10.1108/10444061311316771 Muthén, L. K., & Muthén, B. O. (1998-2017). Mplus user’s guide. (8th ed.). Muthén & Muthén. Paule-Ruiz, M. P., Riestra-Gonzalez, M., Sánchez-Santillan, M., & Pérez-Pérez, J. R. (2015). The Procrastination related indicators in e-learning platforms. Journal of Universal Computer Science, 21(1), 7–22. Rice, W. H. (2006). Moodle E-Learning Course Development: A Complete Guide to Successful Learning Using Moodle. Packt Publishing. Rosenberg, J. M., Beymer, P. N., Anderson, D. J., Van Lissa, C. J., & Schmidt, J. A. (2018). tidyLPA: An R Package to Easily Carry Out Latent Profile Analysis (LPA) Using Open-Source or Commercial Software. Journal of Open Source Software, 3(30), 978, https://doi.org/10.21105/joss.00978 Strunk, K. K. (2012). Investigating a new model of time-related academic behavior: Procrastination and timely engagement by motivational orientation (Unpublished Doctoral dissertation). Retrieved from ProQuest Dissertation Publishing. (3554954) Visser, L. B., Korthagen, F. A. J., & Schoonenboom, J. (2015). Influences on and consequences of academic procrastination of first-year student teachers. Pedagogische Studiën, 92, 394–412. You, J. W. (2015). Examining the effect of academic procrastination on achievement using LMS data in e-learning. Educational Technology & Society, 18(3), 64–74. Wang, J., & Wang, X. (2020). Structural equation modeling: Applications using Mplus. (2nd ed.). John Wiley & Sons. Zimmerman, B., & Risemberg, R. (1997). Self-regulatory dimensions of academic learning and motivation. In G.D. Phye (Eds.), Handbook of academic learning: Construction of knowledge (pp. 105-125). Academic Press.