Enhancing Academic Performance in Tertiary Education through Social Media: A Multi-Arm Randomized Controlled Trial

Enhancing Academic Performance in Tertiary Education through Social Media: A Multi-Arm Randomized Controlled Trial | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Enhancing Academic Performance in Tertiary Education through Social Media: A Multi-Arm Randomized Controlled Trial Aida Tarifa-Rodriguez, Javier Virues-Ortega, Ana Calero-Elvira This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7932578/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Apr, 2026 Read the published version in Journal of Behavioral Education → Version 1 posted 9 You are reading this latest preprint version Abstract Few randomized controlled trials have analyzed evidence-based educational practices delivered through a social media environment. This study used a multi-arm randomized controlled trial to evaluate the critical components of an educational intervention package: study self-management skills training delivered through video modeling, cooperative learning, and semi-immediate feedback. We evaluated social media engagement and academic performance among 141 students in a graduate-level applied psychology program. Students were randomly assigned to five groups: control ( n = 27); self-management ( n = 27); cooperative learning ( n = 33); self-management and cooperative learning ( n = 27); and self-management, cooperative learning, and semi-immediate instructor feedback ( n = 27). Results indicated that participants receiving the complete intervention package showed the highest levels of engagement and academic performance. The analysis revealed that instructor semi-immediate feedback was critical to the package's effectiveness, whereas the impact of the self-management skills training could not be verified. We discuss the conceptual, methodological, and practical implications of the study. Facebook cooperative learning higher education instructor feedback self-management video-modeling Figures Figure 1 Figure 2 Figure 3 1. Introduction Online-only instruction may undermine critical aspects of traditional education. For example, students attending online courses enjoy fewer opportunities for social interaction, which could be detrimental to student motivation and morale (Meşe & Sevilen, 2021 ). Incorporating social media platforms into blended and online courses can supplement online education's otherwise limited social milieu (Irwin et al., 2012 ). Moreover, several studies indicate that incorporating social media into university courses may provide robust communication and collaboration that may impact student experience and academic performance (Arteaga et al., 2014; Hamid et al., 2015 ; Manca, 2020 ). For example, Alshuabi et al. (2018) reported that graduate students in courses with attached social media groups showed greater cognitive engagement and academic performance. The authors suggested that cognitive engagement in the social media environment may be a mediating factor leading to increased academic performance. Similarly, a survey by Graham ( 2014 ) reported a positive correlation between increased social study group participation and curricular engagement. These and other findings obtained through self-reported surveys await verification with prospective experimental studies. 1.1 Social Media Platforms Facebook is one of the most popular social media channels among undergraduates and graduate students (Statista, 2024). Closed Facebook groups offer a private space to interact and share information instantly. Namely, receiving private and public messages and comments through push notifications, quickly identifying online members, commenting, and reacting to third-party posts, and receiving instructor moderation may prompt student engagement and participation (e.g., Mansholt et al., 2021 ; Wang & Lin, 2021 ). X (formerly Twitter) has also been evaluated as a social network for academic purposes among graduate students (see, for example, Martinez-Cardama et al., 2019). The facility for immediate feedback may be a key component of the potential impact of social media platforms on the teaching-learning process. Gopal et al. ( 2021 ) and Dhawan ( 2020 ) have observed that providing feedback facilitates peer-to-peer and student-instructor interaction and increases student engagement with academic content. Instant messaging has become a natural means of interaction, leading to the notion that student-instructor interaction shall be equally fast-paced (Giannikas, 2019 ). However, excessive interactions can cause stress or lead to problematic internet use, which can be deleterious to academic performance (Azizi et al., 2019 ) or produce resistance to using social media in education among both students and educators (Cloete & Villiers, 2009 ; Roblyer et al., 2010 ). The balanced and safe use of social media aids in tertiary education demands the assessment of evidence-based educational practices that are amenable to this channel. Studies in this area have traditionally emphasized content (academic curriculum) rather than the range of evidence-based educational procedures that may be amenable to a social media channel (Mahdiuon et al., 2019 ). A further limitation of this literature is the almost complete lack of experimental research, which prevents the direct evaluation of educational strategies compatible with social media channels and can have a measurable impact on student engagement and performance. In an attempt to fill this vacuum, Tarifa-Rodriguez et al. ( 2024 ) developed and evaluated the Behavioral Education and Social Media intervention package (the BE-Social Program) incorporating a collection of evidence-based behavioral intervention strategies selected based on their adaptability to the social media channel. For example, semi-immediate instructor feedback is available on any social network with real-time texting and commenting, whereas whole-class discussion may not. The authors also developed a taxonomy of quantitative engagement metrics to be integrated into the intervention outcomes (Tarifa-Rodriguez et al., 2023 ). 1.2 Elements of a Social Media-Based Behavioral Education Intervention The BE-Social program comprises self-management skills training delivered through video-modeling, cooperative learning environment, and semi-immediate instructor feedback (Tarifa-Rodriguez et al., 2024 ). 1.2.1 Self-Management and Video-Modeling The convenience and accessibility of online platforms allow instructors to teach critical self-management skills, including identifying short- and long-term objectives, prioritizing, and recording time devoted to key study behaviors (writing study notes, working on study questions, active memorization activities, etc.). The teaching objectives of the self-management component of the BE-Social program include: (a) select and define specific target behaviors, (b) observe the target behaviors and collect data, (c) chart the data and modify the definitions of the target behaviors accordingly, and (d) deliver self-reinforcement for small academic accomplishments (e.g., grade in a weekly test of 75% or higher) (Speidel et al., 1981 ). Teaching self-management strategies may empower students to assume a more active role, positively impacting personal satisfaction and academic performance (Rios et al., 2018 ). Self-management skills have been studied extensively in the context of supporting academic engagement and performance in people with learning disabilities (Bahri et al., 2016 ). However, there have been few applications in tertiary education settings. In a notable exception, Choi and Chung ( 2012 ) studied the effectiveness of self-control strategies for college-level students. We divided 84 participants into three experimental groups according to the intensity of the intervention or the number of strategies applied (high, medium, and low). Results indicated that students who had received a high-intensity intervention (i.e., several self-control strategies) demonstrated the best outcomes. Video-modeling consists of presenting a video of an expert performing a target skill with the expectation that the student will imitate the skill while or after viewing the video (Nikopoulos et al., 2016 , p. 187). A meta-analysis by Bellini and Akullian ( 2007 ) strongly indicated that video-modeling is an effective procedure for teaching communicative and functional skills to children and adolescents with autism spectrum disorders (ASD). In addition, video-modeling has been used to teach various motor skills. For example, Quinn et al. ( 2020 ) reported that professional dancers receiving video-modeling and video feedback showed more rapid skill acquisition than those in the control group. Social media functionalities such as live broadcast, video storage and playback, and video chat, among others, suggest that video-modeling is readily compatible with the use of social media in educational settings. Specifically, social media video functionalities can help to model study habits, demonstrate course content elaboration routines, model problem-solving strategies, and other complex repertoires. We delivered the self-management skills training component of the BE-Social program via video-modeling. 1.2.2 Cooperative Learning Cooperative learning typically involves elaborating academic content under the instructor's guidance, promoting student participation, and setting group expectations (Johnson & Johnson, 1982 ). Moreover, cooperative learning emphasizes group collaboration, peer-to-peer communication and assistance, and information and resource sharing (Roseth et al., 2008 ). In particular, cooperative learning environments often include discussion scenarios presented by instructors, interdependent contingencies, and social praise, strategies that are known to facilitate student participation (Popkin & Skinner, 2003 ). There is evidence to suggest that positive student interactions have a positive impact on academic performance (Van Ryzin et al., 2020 ). Moreover, a meta-analysis by Capar and Tarim ( 2015 ) concluded that cooperative learning environments significantly improve academic performance and student motivation. 1.2.3 Feedback Immediate feedback has shown a near-universal impact on skill acquisition and performance (Kluger & DeNisi, 1996 ). Immediate feedback may increase self-reported interest in academic content and lead to higher academic performance among college students (He et al., 2019 ). University students often use instant messaging applications to access academic content efficiently (Kaysi, 2021 ). These findings emphasize the importance of incorporating instant peer-to-peer and instructor-student communication into online education programs, including those using social media as educational aids. The BE-Social package includes a semi-immediate feedback component whereby instructors respond to content-related student comments within a 24-hour cycle. Tarifa-Rodriguez et al. ( 2024 ) randomly assigned 46 students to a default online program (control) or default online program plus BE-Social (intervention) group. Target outcomes included academic performance and social media engagement (reactions, comments). Results indicated that the intervention effectively produced a 20% increase in academic performance and significant increases in social media engagement. A parallel single-subject analysis revealed that intervention effects were, to a considerable extent, idiosyncratic. In the present study, we aim to replicate and expand the findings by Tarifa-Rodriguez et al. ( 2024 ) by incorporating additional methodological standards: (1) increased sample size, (2) balanced number of participants across control and experimental groups, (3) balanced number of sessions across study phases (baseline, treatment and posttreatment), (4) comparable exposure intensity to academic content across groups, and, (5) a multi-arm RCT to evaluate the incremental validity of discrete elements of the BE-Social intervention program (self-management training via video-modeling, cooperative learning environment, and semi-intermediate instructor feedback). 2. Methods 2.1 Participants and setting Participants were students enrolled in an online college-level course in applied psychology. Prospective participants received an email with a detailed research project description, an informed consent form, and a link to a sociodemographic information questionnaire. The curricular content was located on a Moodle platform, and students had weekly access to new materials. The course began in September 2020 and ended in May 2021. We invited 170 students who had relatively low performance in weekly academic tests over the first three months of the course (40% or lower on average). Of these, 141 students (mean age, 41.97; SD, 8.47) agreed to participate (see a summary of participant attrition in Fig. 1 ). Students were primarily immigrant working-class females with a mean of 16.45 ± 5.44 years of education. The selection of students is relevant because social background can impact the probability of graduating from an online program (Sánchez-Gelabert et al., 2023). Table 1 presents the personal information of participants (see also Table A in the Supplementary Online Information for the sociodemographic information of the intention-to-treat group). Participants experienced various lockdowns and other COVID-related restrictions throughout the study. The study was approved by the ethics committee of the Universidad Autónoma de Madrid (ethics approval number CEI 112–2204). Table 1 Sociodemographic Characteristics of Participants (n = 141) Total ( n = 141) G1 ( n = 27) G2 ( n = 33) G3 ( n = 27) G4 ( n = 27) G5 ( n = 27) Gender, % ( n ) Female 90.07 (127) 88.89 (24) 90.91 (30) 96.30 (26) 96.30 (26) 77.78 (21) Male 9.92 (14) 11.11 (3) 6.06 (3) 3.70 (1) 3.70 (1) 22.22 (6) Mean age, years ( SD ) 41.97 (8.47) 40.11 (8.66) 41.48 (8.34) 42.78 (7.12) 44.56 (9.47) 41.04 (8.52) Years of education ( SD ) 16.45 (5.44) 17.70 (5.09) 16.67 (5.61) 15.96 (5.27) 17.44 (5.25) 14.41 (5.65) Mean BL performance ( SD ) 12.33 (2.18) 12.35 (1.72) 12.46 (2.05) 12.58 (2.15) 12.57 (2.52) 11.67 (2.45) Country, % ( n ) U. S 88.65 (125) 88.89 (24) 87.88 (29) 92.59 (25) 96.30 (26) 81.48 (22) Others 11.35 (16) 11.11(3) 12.12 (4) 7.41 (2) 3.70 (1) 18.52 (5) Socioeconomic Status, % ( n ) Lower-middle 61.70 (87) 30.56 (11) 64.71 (23) 58.06 (18) 41.18 (14) 60 (21) Middle 2.84 (4) 2.78 (1) 2.94 (1) 3.23 (1) 3.23 (1) 0 (0) Upper-middle 31.91 (45) 36.11 (11) 23.53 (9) 16.13 (5) 32.35 (11) 17.14 (6) Upper class 2.13 (3) 5.56 (2) 0 (0) 3.23 (1) 0 (0) 0 (0) Notes . G1 = BE-Social; G2 = BE-Social without semi-immediate feedback; G3 = Cooperative learning; G4 = Self-management; G5 = Control group. 2.3. Experimental Design We randomly assigned 141 participants to the five arms of the study. To ensure the effectiveness of the randomization process, we verified that age, gender, educational level, and socioeconomic standing were not statistically different across groups. Participants were blind to their group assignment status. We completed the study over 12 successive weeks (January 18 to April 11). The baseline phase comprised the first four weeks of the study. The intervention phase followed over the next four weeks. To control for a potential instructor effect, we divided the intervention into two two-week periods, and assigned semi-randomly each period within each group to an instructor. Participants returned to baseline conditions during the last phase of the study for another four weeks. Individuals in the control group remained in baseline conditions for the complete duration of the study. 2.4. Procedure A 33-year-old male and a 32-year-old female, both doctoral students, created five closed Facebook groups with similar names (i.e., "we learn together," "together we learn," "study group," "study with me," and "learn with me") and served as instructors and moderators within these social media groups. Each group was assigned a different set of behavioral strategies: complete BE-Social program (Group 1), BE-Social without semi-immediate feedback (Group 2), cooperative learning (Group 3), and self-management (Group 4). The fifth group was the control group. 2.4.1. Baseline During the baseline weeks, four multiple-choice discussion questions were posted daily (Monday through Friday). All posted questions covered aspects of the course content taught during that week. The posts were identical for all groups. Instructors scheduled a weekend day to provide feedback on the twenty discussion questions posted over the week. Instructors provided feedback in the form of comments to the relevant posts. After providing weekly feedback, the instructor would then "turn off commenting" for each post. In addition, each instructor broadcasted one-hour weekly live videos featuring the instructor engaging in office-like activities (i.e., working on a computer, writing notes, and silently reading and writing flashcards). Students could watch these video broadcasts synchronously or asynchronously and imitate the instructor's behavior. Instructors did not interact with students during the video broadcasts. Baseline activities were intended to maintain student engagement (thereby preventing participant dropout) and equate the instructor inputs across control and intervention groups. The control group was under baseline conditions for the complete duration of the study. 2.4.2. Intervention Complete BE-Social program (Group 1). Participants received all elements of the multi-component package (self-management via video-modeling, cooperative learning, and semi-immediate feedback). Each weekday, the instructors created three posts using all program components. One post involved a self-management strategy delivered through video-modeling, a second post required cooperative learning, and a third post provided the opportunity to receive semi-immediate instructor feedback. For example, one post would involve a video broadcast with the instructor modeling self-recording skills. A second post would present a practical problem to prompt student discussion (cooperative learning). Finally, a third post would consist of a multiple-choice scenario for which the instructor provided feedback to student comments. Instructor feedback for this last post was individualized and semi-immediate, meaning that the delay of the instructor's response was, on average, shorter than one hour. We describe the procedures of each of these program components below. We used self-management to encourage students to assume an active role in the organization and performance of academic tasks by fostering self-reliance and independence. The instructor focused on the critical elements of goal setting, self-monitoring, self-evaluation, and self-reinforcement. The instructor presented and modeled self-management skills through video-modeling sessions. To teach goal-setting skills, instructors made 10-min video broadcasts presenting a step-by-step approach to creating a list of specific, measurable, and achievable goals using specific course topics from the program syllabus to ensure relevance. The instructors demonstrated how to make a study planning schedule considering the time available, amount of study material, reading time, and time needed to complete assignments and self-assessment tests. The students could ask questions during or after the video broadcast and share their study planning schedule to receive feedback from the instructor. The self-monitoring component encouraged students to assess their progress as they worked toward their study goals. The instructors used a video broadcast to present self-recording strategies, which required students to time the duration of academic activities defined during the goal-setting phase. Then, the instructor demonstrated how to transfer daily study time records into a spreadsheet for graphing. The instructor added a weekly test performance data line to the graph and asked students to follow suit. Instructors encouraged students to check their weekly graphs and correlate study time with performance. Finally, instructors encouraged students to take extra free time (self-reinforcement) when they had improved their study time and performance. We implemented self-management procedures as an antecedent-based intervention with minimal contingent feedback beyond answering questions or providing sporadic feedback to students who presented their goal-setting schedule during a video broadcast (see Appendices 1 and 2 in the Supplementary Online Information for examples of self-management training via video-modeling posts). The goal of the cooperative learning component was to establish positive interactions among students by encouraging praise and informative feedback as part of their communication style. Cooperative learning posts could contain an image, a short video, a text, a combination of text and video, or a combination of text and image. Each post involved one of two possible activities: (a) presenting an applied scenario where multiple answers could be correct to generate discussion among students, and (b) presenting the outline of a student study time plan with blank spaces to generate cooperation among students to complete the proposed plan. The instructor further facilitated interaction by (a) presenting demand-free hints and prompts when communication halted (e.g., "This scenario is so relevant to professional practice! Have you considered the potential ethical ramifications?"), (b) tagging three students to each student's comment, (c) tagging students who had been inactive for the last 24 hours, and (d) providing weekly general feedback on the level of student communication and publicly praising those students who had given the most feedback to the rest of their peers. The instructor did not provide semi-immediate feedback as part of the cooperative learning procedures. BE-Social program without semi-immediate feedback (Group 2). The intervention procedures for this group were identical to those described for Group 1, except for the semi-immediate feedback component. Specifically, instructors provided feedback weekly over the weekend (as opposed to providing feedback within a 24-cycle or sooner). Cooperative learning (Group 3). The intervention procedures for the cooperative learning group were identical to those described for the cooperative learning component of Group 1, with the exception that all three daily posts were cooperative learning posts. Therefore, they only received the cooperative learning component of the BE-Social program. Self-management training via video-modeling (Group 4). The intervention procedures for the self-management training via video-modeling group were identical to those described for this component of the BE-Social group (Group 1), except that all three daily posts involved study self-management skills training delivered via video-modeling. Therefore, they only received the self-management component of the BE-Social program. 2.4.3 Post-Intervention The post-intervention phase procedures were identical to those described for the baseline phase. 2.5 Dependent variables Participants took a total of twelve weekly tests throughout the study. Each test included 20 multiple-choice questions with three distractors and one correct answer. We calculated test scores by dividing correct responses by 20 and converting that ratio into a percentage. Ninety percent of the scenarios described practical applications of the course contents, while the remainder of the questions focused on conceptual information (e.g., definitions, characteristics, conceptual clarification, etc.). To minimize the effect of content difficulty on the average performance of participants, the online tests, published weekly, could be completed over a three-week time window. We recorded the number of students that had viewed any instructor post (views) and the total number of student comments and reactions (e.g., likes) generated by each instructor post. We summarized these data as combined engagement , defined as the sum of views, likes, and comments per post. 2.5. Interobserver agreement and procedural integrity A secondary observer obtained data on student engagement (views, reactions, and comments) and instructor latency in resolving or responding to students. Twenty percent of instructor posts were selected from each of the five groups to assess interobserver agreement (IOA). We computed IOA by dividing the number of posts with agreement by the total number of posts selected for evaluation. Overall, our IOA reached 99.6% (range, 59% to 100%), 99.9% (96% to 100%), 100%, and 100% for instructor feedback latency, views, comments, and likes, respectively. To document procedural integrity, an independent observer recorded critical aspects of the intervention procedures throughout all study phases, including the number of daily posts, number of weekly videos, latency to instructor feedback (weekly vs. semi-immediate), and video content (delivery of self-management strategies). The independent observer conducted a 60-s partial interval recording of 20% of instructor video broadcasts to monitor video content. The observation protocol helped to verify the amount of instructor time devoted to discussing self-management skills. Observations were conducted with the app Big Eye Observer, validated for video-based behavioral observation applications (Virues-Ortega et al., 2023 ). Instructors also recorded the following intervention-specific procedural integrity standards during the intervention phase: number of self-management posts, format of self-management posts (video vs. text), cooperative learning posts, and mean latency of instructor feedback. A comprehensive summary of the procedural integrity assessment is available in the online supplementary information (Table B). 2.6. Statistical analysis 2.6.1. Academic achievement We conducted a mixed-model ANOVA with group as a between-groups factor and time (Time 1: Baseline; Time 2: Intervention; Time 3: Post-Intervention) as a within-subjects factor. The dependent variable was the mean performance in weekly course content tests (range, 0 to 20). Achieved power and effect size metrics are reported for the omnibus tests. In addition, we computed t -test pairwise group and time comparisons and estimated Hedges g effect sizes for any effects identified through the omnibus mixed-model ANOVA. 2.6.2. Social media engagement The Facebook platform does not provide access to the user's identity visualizing a video post (only the aggregate number of visualizations is available). This constraint meant we should use posts as units of analysis (as opposed to subjects within a group) to study engagement responses. The number of engagement responses (i.e., views, likes, comments) across posts was not normally distributed. This, added to the longitudinal (i.e., successive posts within a group) and hierarchical structure (i.e., group-level predictors) of the engagement responses, favored the use of hierarchical linear models (HLM) in our analysis. We used combined engagement, defined as the sum of views, likes, and comments per post, as our dependent variable. We used the group identifier as the subject variable and the post number within a group as the time-based variable. A first-order autoregressive covariance structure rendered the best goodness of fit values during the model development process. We used group (BE-Social, BE-Social without semi-immediate feedback, self-management, cooperative learning, control), time (baseline, treatment, posttreatment), post type (textual, visual, video), and the interaction between group and post type as fixed-effect factors. In addition, we added the group-level mean academic performance as a group-nested random-effects factor. Akaike's information criterion (AIC) of the unconditional model was reduced gradually by the stepwise addition of each factor (Akaike, 1974 ; Burhan & Anderson, 2002). The unconditional and terminal models' AIC values were 6991 and 6700, respectively. The addition of a random intercept did not improve model fitness. Group-nested academic performance correlated significantly with phase and group and was the subject of a second model without apparent confounders. We used the metric of the dependent variable for all model coefficients (range, 0 to 1). We used SPSS® IBM® Statistics, version 27 (IBM Corporation, 2021) for all analyses. We used a p value of 0.05 with Bonferroni adjustments for multiple comparisons. 2.7. Social Validity and Acceptability At the end of the study, the instructors made a live farewell video for each group, thanking students for their participation and inviting comments about their experiences and concerns. We recorded the unsolicited comments of the live-streamed farewell videos. We imported these comments into a textual analysis tool to identify common semantic themes or categories from textual samples (Hunerberg, 2019 ). We report the percentage of unsolicited positive feedback and the themes resulting from the textual analysis as indirect indications of the intervention's social validity (i.e., acceptability). 3. Results 3.1. Academic achievement The Box test for the mixed-model ANOVA was not significant ( p = .179), suggesting that the assumption of the equality of covariance matrices was met. Multivariate tests following the Wilks' lambda distribution revealed a significant Time effect ( F [2, 135] = 66.11, p < .001, η 2 = 0.50, 1- β = 1.00) and Time by Group interaction ( F [8, 270] = 4.06, p < .001, η 2 = 0.12, 1- β = .99). The corrected within-subject effect for Time ( F [1.69, 229.11] = 42.92, p < .001, η 2 = 0.24, 1- β = 1.00) and Time by Group interaction ( F [6.74, 229.11] = 3.25, p = .003, η 2 = 0.09, 1- β = .95) were also significant. The main effect of Group was also significant, F (4, 136) = 4.63, p = .002, η 2 = 0.12, 1- β = .94. The homogeneity assumption for the equality of error variances across groups was met. There was a main effect of Group as well, F (4, 136) = 126.05, p = .002, η 2 = 0.12, 1- β = .94. As it seems apparent from Fig. 1 , within-subject contrasts indicated, the effect of time followed a quadratic rather than a linear trend ( F [1, 136] = 7.73, p = .006, η 2 = 0.05 vs. F [1, 136] = 130.11, p < .001, η 2 = 0.49) suggesting that the intervention effects were to a certain degree transient for the intervention groups. In contrast, the control group followed a deteriorating trend over time (Fig. 1 ). The pairwise t -test comparisons indicated that the baseline academic performance in any of the intervention groups was not statistically different from that of the control group, p < .01 (Table 2 ). The only intervention group that did not differ from the control group during the intervention phases was the self-management intervention group (Group 4). All other intervention groups showed statistically significant departures in academic performance, both during the intervention (Time 2) and post-intervention phases (Time 3) (Table 2 ). Specifically, the complete BE-Social group (Group 1) showed a statistically significant increase in performance during the intervention, t (1. 52) = 5.38, p < .001, and post-intervention phases, t (1. 52) = 3.53, p < .001 relative to the control group. The effect sizes were both within the large effect size range ( g = 1.44, 98.75% CI 0.68, 2.97 and g = 0.95, 95% CI 0.23, 1.65, respectively). The slightly lower post-intervention effect suggests that the effect lessened after the intervention withdrawal. Table 2 Pairwise t Test Comparisons for Academic Achievement across Groups Group Pairwise t tests Mean SD t df p Hedges g [98.75% CI ] G1: BE-Social Time 1 13.22 3.48 1.67 1, 52 .102 0.45 [-0.23, 1.12] Time 2 16.04 2.67 5.38 1, 52 < .001 1.44 [ 0.68, 2.97] Time 3 13.48 4.17 3.53 1, 52 < .001 0.95 [ 0.23, 1.65] G2: BE-Social* Time 1 12.07 3.09 0.49 1, 52 .63 0.13 [-0.54, 0.80] Time 2 14.86 3.23 3.67 1, 52 < .001 0.98 [ 0.27, 1.69] Time 3 11.99 4.08 2.29 1, 52 .03 0.61 [-0.08, 1.29] G3: Cooperative learning Time 1 12.76 2.87 1.38 1, 58 .174 0.35 [-0.29, 0.99] Time 2 13.55 2.88 2.48 1, 58 .016 0.64 [-0.02, 1.29] Time 3 11.90 3.71 2.44 1, 58 .018 0.63 [-0.03, 1.28] G4: Self-Management Time 1 12.21 3.09 0.65 1, 52 .52 0.17 [-0.49, 0.84] Time 2 13.02 3.15 1.61 1, 52 .103 0.45 [-0.24, 1.12] Time 3 11.15 3.42 1.69 1, 52 .097 0.45 [-0.23, 1.13] G5: Control Time 1 11.63 3.52 Time 2 11.53 3.43 Time 3 9.32 4.49 Note . All pairwise comparisons against the control group. Bonferroni adjusted confidence intervals. CI = confidence interval. * BE-Social group without semi-immediate feedback. Time 1 = Baseline; Time 2 = Intervention; Time 3 = Post-intervention. The BE-Social without semi-immediate feedback group (Group 2) showed a statistically significant increase in performance during the intervention ( t [1, 52] = 3.67, p < 0.001) and post-intervention phases ( t [1, 52] = 2.29, p = 0.03). The pairwise effect size of Group 2 against the control group for the intervention and post-intervention phases was 0.98 (95% CI 0.27, 1.69) and 0.61 (95% CI -0.08, 1.29), respectively. For the cooperative learning group (Group 3), we observed a slight increase in performance during the intervention and post-intervention phases ( t ([, 58] = 2.48, p = .016 and t [1, 58] = 2.44, p = .018, respectively). However, using the control group as a reference, the intervention effect for Group 3 was somewhat smaller than the one observed for Groups 1 and 2. Group 3 showed nearly identical effect sizes within the moderate effect size range for both the intervention and post-intervention phases ( g = 0.64, 98.75% CI -0.02, 1.29 and g = 0.63, 95% CI -0.03, 1.28, respectively). Therefore, the effect of the intervention endured during the post-intervention phase. However, for all other groups, performance deteriorated during the post-intervention phase (Fig. 2 ). Finally, the pairwise comparison of the self-management skills training group (Group 4) relative to the control group did not reveal statistically significant differences during the intervention and post-intervention phases ( t [1, 52] = 1.61, p = .097 and t [1, 58] = 1.69, p = .016, respectively). Overall, the multi-arm RCT analysis indicated that semi-immediate feedback was critical to the effect of the multi-component intervention. In contrast, the unimodal intervention could not verify the value of self-management skills training. 6.2 Social Media Engagement The final linear mixed model analysis for combined social media engagement (Table 3 ) confirmed a significant effect of Group ( F [1, 800] = 16.69, p < .001), Time ( F [2, 288] = 51.79, p < .001), Post type ( F [2, 859] = 36.28, p < .001), and Group by Post type interaction ( F [8, 880] = 8.48, p < .001). Pairwise comparisons indicated that only the complete BE-Social program (Group 1) and the BE-Social program without semi-immediate feedback (Group 2) had higher levels of social media engagement relative to the control group ( \(\:\stackrel{-}{D}\) = 7.78 ± 2.51, p = .008 and \(\:\stackrel{-}{D}\) = 7.66 ± 2.54, p = .01, respectively). Engagement levels across groups did not change during the intervention phase relative to baseline, whereas engagement decreased across groups during the post-intervention phase relative to baseline ( \(\:\stackrel{-}{D}\) = 8.46 ± 0.91, p < .001). Finally, textual posts induced greater interaction across groups than visual (e.g., image with text) or video posts. Still, only the mean difference between textual and video posts was statistically significant in the pairwise analysis ( \(\:\stackrel{-}{D}\) = 9.59 ± 1.14, p < .001). While a mediation analysis was not feasible, there seems to be an apparent correlation between social media engagement and academic performance, as expressed in the mean differences of both outcomes during the intervention phase (Fig. 3 ). Table 3 Linear Mixed Effects Model for Combined Engagement Fixed effects F df p (Intercept) 2127.37 1, 799.66 < .001 Group 19.69 4, 628.11 < .001 Time 51.79 2, 287.99 < .001 Post type 36.28 2, 858.97 < .001 Group by Post type 8.48 8, 880.13 < .001 Pairwise comparisons, i - j \(\:\stackrel{-}{D}\pm\:SE\) df p BE-Social - Control 7.78 ± 2.51 1, 909.92 .008 BE-Social* - Control 7.66 ± 2.54 1, 906.58 .010 Cooperative learning - Control -0.16 ± 2.66 1, 911.03 1.000 Self-management - Control -4.09 ± 2.46 1, 889.42 .389 Treatment - Baseline 1.11 ± 0.91 1, 308.16 .449 Posttreatment - Baseline -8.46 ± 0.91 1, 233.41 < .001 Text - Visual 2.78 ± 1.49 1, 870.87 .124 Text - Video 9.59 ± 1.14 1, 854.24 < .001 Notes . \(\:\stackrel{-}{D}\pm\:SE\) = mean differences ± standard errors. * BE-Social group without semi-immediate feedback. 3.3 Social Validity and Acceptability Forty-six participants provided unsolicited positive comments during the farewell video broadcasts. Therefore, 32.6% of participants produced unsolicited positive comments. The BE-Social without semi-immediate feedback group provided the most positive comments (51.8%), followed by the complete BE-Social program group (44.5%). Only 18% of participants in the self-management and cooperative learning groups volunteered positive comments. Finally, 33% of participants in the control group shared positive comments. We did not see negative or derogatory comments posted in any of the groups (see Supplementary Online Material, Table C). The sentence-by-sentence thematic analysis indicated that the five most prevalent themes in the feedback messages were knowledge acquisition (35 occurrences, 7.5% of the text), high-quality teaching (27 occurrences, 5.8% of the text), commendation for the video broadcasts (14 occurrences, 3.0% of the text), appreciation for the problem-solving strategies presented (10 occurrences, 2.2% of the text), and perceived value of the experience (9 occurrences, 1.9% of the text). 4. Discussion The present study expands the evidence base of social media platforms to aid tertiary education. We implemented the multi-component BE-Social program as an adjunct to an online applied psychology course to assess the efficacy of the key elements of this program individually and in combination. The results suggest that participants exposed to the complete intervention package (including semi-immediate feedback) demonstrated the highest academic performance. While semi-immediate feedback did not affect the total social media engagement, the current multi-arm RCT seems to suggest that feedback was a critical element of the program. Our results also indicate that part of the intervention gains may fade after the intervention is withdrawn, thereby suggesting that the program ought to be present for the duration of the course for optimal results. The isolated implementation of discrete components of the program (particularly self-management training) had relatively minor effects on performance. Our findings align with the empirical literature suggesting that fast turn-around feedback is conducive to student learning. For example, a systematic review by Liu ( 2021 ) found that the immediacy of teacher responses positively impacted student engagement with the academic material and, ultimately, on performance. In addition, immediacy has been identified as a critical dimension of effective feedback (Er et al., 2021 ; Henderson & Wen, 1976 ). Also, evidence suggests that students prefer receiving feedback from educators through instant messaging instead of email (Gopal et al., 2021 ). While our study does not demonstrate reward learning effects at the individual level, these findings are in line with a delay-discounting effect that is amply documented in the animal and human operant literatures (Reynolds, 2006 ). Specifically, a relatively immediate reward (e.g., immediate instructor feedback) may have a greater impact on engagement and other relevant academic behaviors than more delayed ones. Interestingly, students in the BE-Social group without semi-immediate feedback (Group 2) showed high levels of social media engagement. At the same time, their academic achievement was relatively lower than the one observed in the complete BE-Social group (Group 1). This could indicate that social media engagement may not accurately predict course content elaboration. An alternative interpretation may be that, although frequent student posting can still occur under delayed instructor feedback (for example, because of peer-to-peer interactions), it may be less likely to induce successful content acquisition and elaboration. Social media engagement seems to follow simple social reward optimization processes (Lindström et al., 2021). Such realization has obvious practical implications that we evaluated only indirectly. Specifically, our analysis suggests that cooperative learning and semi-immediate feedback strategies may be particularly effective in mobilizing student online behavior in study groups. While the mediating role of social media responses in academic achievement cannot be asserted, the study provides observational evidence that those groups that showed more active online behavior had relatively higher academic achievement. We also illustrated the integration of multiple forms of immediate social media interactions (e.g., push notifications, tagging, reacting, commenting, etc.) into evidence-based educational practices, including the design of a cooperative learning environment, the delivery of self-management skills training, and the effective use of instructor-mediated feedback. The analytical approach of the multi-arm RCT suggests that the elements of the proposed multi-component intervention are likely to be additive, with the cooperative learning and semi-immediate feedback manipulations being the most likely contributors to the compounded effect observed in the full implementation of the intervention package. The effect of self-management training on performance (Fig. 1 ) was not statistically significant. Given the current multi-arm RCT evaluated discrete components of the intervention, whether isolated or in combination, additional analyses would be needed to substantiate the potential additivity of the critical elements of the BE-Social package (i.e., self-management training, cooperative learning environment, instructor-mediated semi-immediate feedback). While an additive effect hypothesis can hardly be verified with between-groups datasets (see, for example, Van Iddekinge et al., 2017 ), we can tentatively examine the mismatch between empirical effect sizes and the theoretical additive effects. Specifically, a plausible additive effect would mean that the sum of the unimodal interventions without semi-immediate feedback would approach the effect size of the BE-Social group without semi-immediate feedback, $$\:{g}_{3}+{g}_{4}\approx\:{g}_{2}$$ 1 $$\:0.64\:+\:0.45\approx\:\:0.98$$ Where \(\:{g}_{2}\) , \(\:{g}_{3}\) , and \(\:{g}_{4}\) are the pairwise comparison Hedges g effect size for Groups 2, 3, and 4, respectively. Similarly, the incremental effect of semi-immediate feedback and the separate effects of the unimodal interventions (Group 3 and Group 4) should approach the effect size of the full multi-component intervention (Group 1), $$\:{g}_{3}+{g}_{4}+\:({g}_{1}-{g}_{2})\approx\:{g}_{1}$$ 2 $$\:0.64+0.45+\:(1.44-0.98)\approx\:1.44$$ Where \(\:{g}_{1},\:{g}_{2},\:{g}_{3},\:{g}_{4}\) , and \(\:{g}_{5}\:\) are the pairwise comparison Hedges g effect sizes for Groups 1 through 5, respectively. The observed empirical effect sizes match the theoretical additive relation within 0.1 effect size units in both vases (0.98 vs. 1.09 and 1.44 vs. 1.55, respectively). While this demonstration is anecdotal, it does suggest that the multi-arm approach may be an analytic tool to weigh the relative contribution of multi-component educational programs. The results show that combining all the selected educational strategies may have a synergic effect. When applied individually, each intervention component seems to detract from the effect of the complete program. However, the current analytical approach was not comprehensive. Specifically, we could not assess semi-immediate feedback without other intervention procedures. Moreover, the control group baseline condition involved minimal interactions deemed essential to maintain student morale and engagement. Critical comparisons involved adding elements to a background intervention rather than comparing the absence of intervention to a discrete intervention component. Therefore, we cannot discard complex interactions between the instructor's baseline behavior and subsequent performance. Interestingly, instructor textual posts generated more engagement than video broadcasts (i.e., self-management training via video-modeling posts) or mixed posts with text and images. While studies that quantitatively evaluate engagement are rare, some surveys suggest that text-based posting generates more interactive engagement. In contrast, video-based and mixed posts seem to impose a one-sided communication dynamic (see, for example, Swartzwelder et al., 2019 ). The nature of the content may also play a role. For example, text-based posts presented multiple-choice scenarios that frequently prompted students to guess and follow up with questions. We should note several limitations and future extensions to the current. First, we have examined the compatibility of an incidental collection of behavioral education strategies using one of the most popular social media platforms for tertiary education students. Future studies could replicate our findings with alternative combinations of evidence-based strategies and social media platforms. The specific parameters of the strategies utilized here could be adjusted (e.g., elements of the cooperative learning and self-management protocols). For example, adding performance-based feedback instead of relying primarily on video modeling could improve the effect of the self-management protocol. Second, our data collection strategy did not allow us to reconcile discrete online events (e.g., reactions, views, posts, comments) with individual students, which meant that the impact of the intervention could only be assessed at the group level, restricting the possibility of mediation analysis. This also meant that idiosyncratic patterns of treatment effects could not be studied in detail (for an analysis of the idiosyncratic effects of the BE-Social program, see Tarifa-Rodriguez et al., 2024 ). Future studies in this area could minimize these concerns by purposely developing application programming interfaces that would automate aspects of the data collection process in the social media environment. Additional practical enhancements include integrating text and theme analysis systems to characterize further social dynamics and emotional factors involved in social media exchanges in educational contexts (Drus & Khalid, 2019 ). Moreover, adding and evaluating AI-powered feedback could greatly minimize intervention costs and facilitate the program's deployment at scale (see, for example, Escalante et al., 2023 ). Third, our outcome measures were limited to ad hoc course content tests and standard social media responses (i.e., views, likes, comments, posts). These outcomes were important because they allowed frequent measurements across study phases and were ecologically valid and naturally integrated with the course. However, future analyses would benefit from adding standardized measurements of student satisfaction and end-of-course assessments. The latter was not practical in the current study owing to the multi-phase structure of the design, which had to fit within a one-year course. While frequent testing allowed the opportunity to monitor the intervention effects over time, it prevented a more comprehensive evaluation of the program's overall impact on student academic achievement. Fourth, the apparent deteriorating academic achievement trend observed in the control group may be due to selecting low-performance students (see inclusion criteria) or to the hierarchical nature of course contents (i.e., latter lessons relied on the foundational knowledge introduced in earlier ones). It is interesting to note that the elements of the BE-Social program were sufficient to offset this deteriorating trend, even though intervention gains were not evident during the post-intervention phase for those elements. 5. Conclusions Our study supports a few tentative conclusions: (1) the unique combination of self-management training via video-modeling, cooperative learning online environment, and semi-immediate instructor feedback included in the BE-Social program had a large positive impact on social media engagement and academic achievement, (2) the effect of the elements of the intervention seems to be additive, although we could only verify the incremental validity of semi-immediate feedback, (3) increased social media engagement may be a mediating factor to increased academic achievement, and (4) the multi-component intervention may have optimal effects on achievement when delivered for the complete duration of a semester or year-long course. The exponential growth of the casual use of social media in tertiary education calls for an extensive evaluation of social media as a medium for evidence-based educational practices. Integrating social media study groups in tertiary education courses may be a cost-effective approach to enhancing peer- and instructor-mediated interactions and improving student experience and academic achievement. The proposed program demonstrated that key evidence-based educational practices amenable to a social media study group can support student participation and performance. Declarations Declaration of generative AI and AI-assisted technologies in the writing process. The authors did not use AI-assisted technologies in the process of writing this manuscript. Availability of data and material. The complete databased use for all analyses will be made public through the platform Figshare upon the manuscript's acceptance for publication. Funding. This study received financial support from the research contracts RYC-2016-20706 (Ramon y Cajal Program, Spain) and CON02739 (The University of Auckland, New Zealand). Acknowledgements . We thank Agustín Perez-Bustamante for his assistance with data coding. Author Contribution ATR. This study was part of the requirements for the Doctor of Psychology degree of the first author at the Universidad Autónoma de Madrid (Spain). Research design development. Data collection and data curation. Data analysis design. Manuscript writing and editing (first draft). JVO. Research design development. Data analysis design. Logistics and resources. Funding procurement. Manuscript writing and editing. Doctoral supervision of ATR.ACE. Research design development. Manuscript writing and editing. Doctoral supervision of ATR. References Ali, W. (2020). Online and remote learning in higher education institutes: a necessity in light of COVID-19 pandemic. Higher Education Studies, 10 (3), 16. https://doi.org/10.5539/hes.v10n3p16 Alkatout, I., Dhanawat, J., Ackermann, J., Freytag, D., Peters, G., Maass, N., Mettler, L., & Pape, J. M. (2021). Video feedback and video-modeling in teaching laparoscopic surgery: a visionary concept from Kiel. Journal of Clinical Medicine, 10 (1), 163. https://doi.org/10.3390/jcm10010163 Alshuaibi, M. S. I., Alshuaibi, A. S. I., Shamsudin, F. M., & Arshad, D. A. (2018). Use of social media, student engagement, and academic performance of business students in Malaysia. International Journal of Educational Management, 32 (4), 625–640. https://doi.org/10.1108/IJEM-08-2016-0182 Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19 (6), 716–723. https://doi.org/10.1109/TAC.1974.1100705 Arteaga Sánchez, R., Cortijo, V., & Javed, U. (2014). Students' perceptions of Facebook for academic purposes. Computers & Education, 70 , 138–149. https://doi.org/10.1016/J.COMPEDU.2013.08.012 Azizi, S. M., Soroush, A., & Khatony, A. (2019). The relationship between social networking addiction and academic performance in Iranian students of medical sciences: a cross-sectional study. BMC Psychology, 7 (1), 28. https://doi.org/10.1186/s40359-019-0305-0 Bahri, L., Mirnasab, M., Noorazar, G., Fathi Azar, E., & Asadi, S. (2016). The effect of self-management strategies on improving task performance in students with learning disabilities. Journal of Analytical Research in Clinical Medicine, 4 (1), 53–65. https://doi.org/10.15171/jarcm.2016.009 Bellini, S., & Akullian, J. (2007). A meta-analysis of video-modeling and video self-modeling interventions for children and adolescents with autism spectrum disorders. Exceptional Children, 73 (3), 264–287. https://doi.org/10.1177/001440290707300301 Bertucci, A., Conte, S., Johnson, D. W., & Johnson, R. T. (2010). The impact of size of cooperative group on achievement, social support, and self-esteem. Journal of General Psychology, 137 (3), 256–272. https://doi.org/10.1080/00221309.2010.484448 Burnham, K. P., & Anderson, D. R. (2002). Model Selection and Multimodel Inference (2 nd ed.). Springer. Capar, G., & Tarim, K. (2015). Efficacy of the cooperative learning method on mathematics achievement and attitude: A meta-analysis research. Educational Sciences: Theory & Practice, 15 (6), 553–559. https://doi.org/10.12738/estp.2015.2.2098 Cloete, S., & Villiers, C. de. (2009). Facebook as an academic tool for ICT lecturers. Proceedings of the 2009 Annual Conference of the Southern African Computer Lecturers' Association, 16-22. https://doi.org/10.1145/1562741.1562743 Choi, J. H., & Chung, K. M. (2012). Effectiveness of a college-level self-management course on successful behavior change. Behavior Modification, 36 (1), 18–36. https://doi.org/10.1177/0145445511418102 Cooper, J. O., Heron, T. E., & Heward, W. L. (2019). Applied Behavior Analysis (3 rd ed.). Pearson. Dhawan, S. (2020). Online learning: a panacea in the time of COVID-19 crisis. Journal of Educational Technology Systems, 49 (1), 5–22. https://doi.org/10.1177/0047239520934018 Drus, Z., & Khalid, H. (2019). Sentiment analysis in social media and its application: Systematic literature review. Procedia Computer Science, 161 , 707–714. https://doi.org/10.1016/j.procs.2019.11.174 Er, E., Dimitriadis, Y., & Gašević, D. (2021). A collaborative learning approach to dialogic peer feedback: a theoretical framework. Assessment and Evaluation in Higher Education, 46 (4), 586–600. https://doi.org/10.1080/02602938.2020.1786497 Escalante, J., Pack, A., & Barrett, A. (2023). AI-generated feedback on writing: Insights into efficacy and ENL student preference. International Journal of Educational Technology in Higher Education, 20 , 57. https://doi.org/10.1186/s41239-023-00425-2 Giannikas, C. (2019). Facebook in tertiary education: the impact of social media in e-learning. Journal of University Teaching and Learning Practice, 17 (1). https://doi.org/10.53761/1.17.1.3 Gopal, R., Singh, V., & Aggarwal, A. (2021). Impact of online classes on the satisfaction and performance of students during the pandemic period of COVID 19. Education and Information Technologies, 26 , 6923–6947. https://doi.org/10.1007/s10639-021-10523-1 Graham, M. (2014). Social media as a tool for increased student participation and engagement outside the classroom in higher education. Journal of Perspectives in Applied Academic Practice, 2 (3), 16–24. https://doi.org/10.14297/jpaap.v2i3.113 Hamid, S., Waycott, J., Kurnia, S., & Chang, S. (2015). Understanding students' perceptions of the benefits of online social networking use for teaching and learning. The Internet and Higher Education, 26 , 1-9. https://doi.org/10.1016/j.iheduc.2015.02.004 He, J., Hao, W., & Kim, J.-W. (2019). The effects of instant feedback system on course interest and academic achievement in gamification learning. Educational Innovations and Applications, 467–470. https://doi.org/10.35745/ecei2019v2.119 Henderson, W. T., & Wen, S. S. (1976). Effects of immediate positive reinforcement on undergraduates' course achievement. Psychological Reports, 39 (2), 568-570. https://doi.org/10.2466/pr0.1976.39.2.568 Hunerberg, E. (2019 ). 3rd Eye Theme Analysis [Online Application]. https://3rdeyeinformation.com/home Irwin, C., Ball, L., Desbrow, B., & Leveritt, M. (2012). Students' perceptions of using Facebook as an interactive learning resource at university. Australasian Journal of Educational Technology, 28 (7), 1221–1232. https://doi.org/10.14742/ajet.798 Johnson, D. W., & Johnson, R. T. (1982). The effects of cooperative and individualistic instruction on handicapped and nonhandicapped students. The Journal of Social Psychology, 118 (2), 257–268. https://doi.org/10.1080/00224545.1982.9922805 Kaysi, F. (2021). Mobile instant messaging application habits among university students. Interactive Learning Environments, 1–19. https://doi.org/10.1080/10494820.2021.1922460 Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119 (2), 254–284. https://doi.org/10.1037/0033-2909.119.2.254 Kusumawati, A. J. (2020). Redesigning face-to-face into online learning for speaking competence during covid-19: ESP for higher education in Indonesia. International Journal of Language Education, 4 (2), 276–288. https://doi.org/10.26858/ijole.v4i2.14745 Li, S., Zhu, Y., & Ellis, R. (2016). The effects of the timing of corrective feedback on the acquisition of a new linguistic structure. Modern Language Journal, 100 (1), 276–295. https://doi.org/10.1111/modl.12315 Liu, W. (2021). Does teacher immediacy affect students? A systematic review of the association between teacher verbal and non-verbal immediacy and student motivation. Frontiers in Psychology, 12 , 713978. https://doi.org/10.3389/fpsyg.2021.713978 Mah, E., Yu, J., Deck, M., Lyster, K., Kawchuk, J., Turnquist, A., & Thoma, B. (2021). Immersive video-modeling versus traditional video-modeling for teaching central venous catheter insertion to medical residents. Cureus, 13 (3) https://doi.org/10.7759/cureus.13661 Mahdiuon, R., Salimi, G., & Raeisy, L. (2019). Effect of social media on academic engagement and performance: Perspective of graduate students. Education and Information Technologies, 2427–2446. https://doi.org/10.1007/s10639-019-10032-2 Manca, S. (2020). Snapping, pinning, liking or texting: Investigating social media in higher education beyond Facebook. The Internet and Higher Education, 44 , 100707. https://doi.org/10.1016/j.iheduc.2019.100707 Mansholt, H., Götz, N. A., & Babitsch, B. (2021). Chances and barriers of online problem-based learning (ePBL) for advanced training in the healthcare sector. Studies in Health Technology and Informatics, 281 , 822–823. https://doi.org/10.3233/SHTI210292 Martínez-Cardama, S., & Caridad-Sebastián, M. (2019). Social media and new visual literacies: Proposal based on an innovative teaching project. Education for Information, 35 (3), 337–352. https://doi.org/10.3233/EFI-180214 Means, B., Toyama, Y., Murphy, R., & Baki, M. (2013). The effectiveness of online and blended learning: A meta-analysis of the empirical literature. Teachers College Record, 115 (3). https://doi.org/10.1177/016146811311500307 Nikopoulos, C., Luiselli, J. K., & Fischer, A. J. (2016). Video-modeling. In J. K. Luiselli, & Aaron J. Fischer (Eds.), Computer-Assisted and Web-Based Innovations in Psychology, Special Education, and Health (pp. 1 87–210). Academic Press. https://doi.org/10.1016/B978-0-12-802075-3.00008-5 Meşe, E. & Sevilen, Ç. (2021). Factors influencing EFL students' motivation in online learning: A qualitative case study. Journal of Educational Technology and Online Learning, 4 (1), 11-22. Retrieved from https://dergipark.org.tr/en/pub/jetol/issue/60134/817680 Popkin, J., & Skinner, C. H. (2003). Enhancing academic performance in a classroom serving students with serious emotional disturbance: Interdependent group contingencies with randomly selected components. School Psychology Review, 32 (2), 282–295. https://doi.org/10.1080/02796015.2003.12086199 Quinn, M., Narozanick, T., Miltenberger, R., Greenberg, L., & Schenk, M. (2020). Evaluating video-modeling and video-modeling with video feedback to enhance the performance of competitive dancers. Behavioral Interventions, 35 (1), 76–83. https://doi.org/10.1002/bin.1691 Rega, A., Somma, F., & Iovino, L. (2020). Development of emotional skills through video-modeling: A case study with a non-verbal participant. Technology and Disability, 32 (3), 211–218. https://doi.org/10.3233/TAD-180222 Reynolds B. (2006). A review of delay-discounting research with humans: relations to drug use and gambling. Behavioural Pharmacology, 17 (8), 651–667. https://doi.org/10.1097/FBP.0b013e3280115f99 Rios, D., Kazemi, E., & Peterson, S. M. (2018). Best practices and considerations for effective service provision via remote technology. Behavior Analysis: Research and Practice, 18 (3), 277–287. https://doi.org/10.1037/bar0000072 Roblyer, M. D., McDaniel, M., Webb, M., Herman, J., & Witty, J. V. (2010). Findings on Facebook in higher education: A comparison of college faculty and student uses and perceptions of social networking sites. Internet and Higher Education, 13 (3), 134–140. https://doi.org/10.1016/j.iheduc.2010.03.002 Roseth, C. J., Johnson, D. W., & Johnson, R. T. (2008). Promoting early adolescents' achievement and peer relationships: The effects of cooperative, competitive, and individualistic goal structures. Psychological Bulletin, 134 (2), 223–246. https://doi.org/10.1037/0033-2909.134.2.223 Sánchez-Gelabert, A., & Elias, M. (2023). Going online? Does transferring to online university increase the likelihood of graduation among students from lower social background? International Journal of Educational Technology in Higher Education, 20 , 39. https://doi.org/10.1186/s41239-023-00407-4 Speidel, G. E., Tharp, R. G., & Speidel, G. E. (1981). What does self-reinforcement reinforce? An empirical analysis of the contingencies in self-determined reinforcement. Child Behavior Therapy, 2 (4), 1–22. https://doi.org/10.1300/J473V02N04_01 Statista (2024, April). Distribution of Facebook users worldwide as of April 2024, by age and gender. https://www.statista.com/statistics/376128/facebook-global-user-age-distribution/ Swartzwelder, K., Murphy, J., & Murphy, G. (2019). The impact of text-based and video discussions on student engagement and interactivity in an online course. Journal of Educators Online, 16 (1). https://doi.org/10.9743/jeo.2019.16.1.13 Tarifa-Rodriguez, A., Virues-Ortega, J., Perez-Bustamante Pereira, A., Calero-Elvira, A., & Cowie, S. (2023). Quantitative indices of student social media engagement in tertiary education: a systematic review and a taxonomy. Journal of Behavioral Education . https://doi.org/10.1007/s10864-023-09516-6 Tarifa-Rodriguez, A., Virues-Ortega, J., & Calero-Elvira, A. (2024). A behavioral education in social media program for postgraduate academic achievement: a randomized controlled trial. Journal of Behavioral Education . https://doi.org/10.1007/s10864-024-09545-9 Tran, P. (2016). Training learners to use quizlet vocabulary activities on mobile phones in Vietnam with Facebook. JALT CALL Journal, 12 (1), 43–56. https://doi.org/10.29140/jaltcall.v12n1.201 Van Iddekinge, C. H., Aguinis, H., Mackey, J. D., & DeOrtentiis, P. S. (2017). A meta-analysis of the interactive, additive, and relative effects of cognitive ability and motivation on performance. Journal of Management, 20 (10), 1-31. https://doi.org/10.1177/0149206317702220 Van Ryzin, M. J., Roseth, C. J., & Biglan, A. (2020). Mediators of effects of cooperative learning on prosocial behavior in middle school. International Journal of Applied Positive Psychology, 5 (1–2), 37–52. https://doi.org/10.1007/s41042-020-00026-8 Virues-Ortega, J., Delgado Casas, C., Martin, N., Tarifa-Rodriguez, A., Reina Hidalgo, A. J., & Cox, A. D., & Navarro Guzmán, J. I. (2023) Interobserver agreement in software-aided and paper-and-pencil behavioral observation. Behavior Research Methods, 55, 855–866. https://doi.org/10.3758/s13428-022-01861-0 Wang, W. T., & Lin, Y. L. (2021). Evaluating factors influencing knowledge-sharing behavior of students in online problem-based learning. Frontiers in Psychology, 12, 691755. https://doi.org/10.3389/fpsyg.2021.691755 Additional Declarations No competing interests reported. Supplementary Files JBESupplementarymaterial.docx Cite Share Download PDF Status: Published Journal Publication published 21 Apr, 2026 Read the published version in Journal of Behavioral Education → Version 1 posted Editorial decision: Revision requested 19 Jan, 2026 Reviews received at journal 15 Jan, 2026 Reviews received at journal 15 Jan, 2026 Reviewers agreed at journal 29 Dec, 2025 Reviewers agreed at journal 12 Dec, 2025 Reviewers invited by journal 05 Nov, 2025 Editor assigned by journal 23 Oct, 2025 Submission checks completed at journal 23 Oct, 2025 First submitted to journal 23 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7932578","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":545593248,"identity":"cefc0d65-5ad5-4440-aeaa-b9f3d58116ec","order_by":0,"name":"Aida Tarifa-Rodriguez","email":"","orcid":"","institution":"Universidad Autónoma de Madrid","correspondingAuthor":false,"prefix":"","firstName":"Aida","middleName":"","lastName":"Tarifa-Rodriguez","suffix":""},{"id":545593249,"identity":"92875159-6757-4578-b056-82052de77754","order_by":1,"name":"Javier Virues-Ortega","email":"data:image/png;base64,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","orcid":"","institution":"The University of Auckland","correspondingAuthor":true,"prefix":"","firstName":"Javier","middleName":"","lastName":"Virues-Ortega","suffix":""},{"id":545593250,"identity":"c5b27dca-8898-48e5-8028-ac5211f4bda0","order_by":2,"name":"Ana Calero-Elvira","email":"","orcid":"","institution":"Universidad Autónoma de Madrid","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"","lastName":"Calero-Elvira","suffix":""}],"badges":[],"createdAt":"2025-10-23 13:23:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7932578/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7932578/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10864-026-09629-8","type":"published","date":"2026-04-21T15:59:43+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":96049629,"identity":"aac8f532-7ae4-446b-8cf1-d3f3b082bd1f","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":295142,"visible":true,"origin":"","legend":"","description":"","filename":"BlindedManuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/7c6024e60b852b21acc44f87.docx"},{"id":96049630,"identity":"3ab52da9-e8ea-4034-86c6-03723a3d993d","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5129,"visible":true,"origin":"","legend":"","description":"","filename":"f8bae1ccd1f6468280dd315846741eba.json","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/81a1cd026983f026bd27ec4b.json"},{"id":96245705,"identity":"b1631daa-0351-457e-a961-cd9d1e59a9fb","added_by":"auto","created_at":"2025-11-19 07:22:00","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28582,"visible":true,"origin":"","legend":"","description":"","filename":"JBESupplementarymaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/feca862c47f2892a7e361161.docx"},{"id":96049633,"identity":"c2c0da09-bb66-4d05-8231-95a2a961c2b4","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":177004,"visible":true,"origin":"","legend":"","description":"","filename":"f8bae1ccd1f6468280dd315846741eba1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/8b2c5c561f551c002c1eae41.xml"},{"id":96049635,"identity":"6a5107c6-1d84-4ee8-abf8-49eeac7a86dc","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"emf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":283000,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.emf","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/1f5b197fd0d4d04c696e2656.emf"},{"id":96246490,"identity":"1f0c11f5-64cf-4012-b1d9-7636c96d5c56","added_by":"auto","created_at":"2025-11-19 07:26:06","extension":"emf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":83164,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.emf","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/6ed09001e406e3707c77d327.emf"},{"id":96049638,"identity":"3120db80-fc4f-4c6c-b2d8-7d3d2e9cf072","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":237247,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/af2eca0357b39b1efb8db474.jpeg"},{"id":96049636,"identity":"7e4078a8-5aca-4341-9cb9-b40337082b83","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":172351,"visible":true,"origin":"","legend":"","description":"","filename":"f8bae1ccd1f6468280dd315846741eba1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/dced131d4a6f4fced19b7514.xml"},{"id":96247131,"identity":"a5fcb5bd-a414-4c9b-ab65-fef32faf4efb","added_by":"auto","created_at":"2025-11-19 07:27:11","extension":"html","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":188695,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/780661ad4cf646c5eba8095b.html"},{"id":96049626,"identity":"a0ded7f5-3fdd-45ca-a613-f994997f18c8","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":18025,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eParticipant Attrition Flowchart\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/3024b854b26b8cd5c8db8cf4.png"},{"id":96049627,"identity":"18d83ff6-d87c-456c-ba03-980a5275c07a","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":12276,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eChanges in Academic Achievement during Pre-Intervention, Intervention, and Post-Intervention\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNote.\u003c/em\u003e Time 1 = Baseline; Time 2 = Intervention; Time 3 = Post-intervention. * BE-Social group without semi-immediate feedback.\u003c/p\u003e","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/34095e5abbdf22b7a9805377.png"},{"id":96049632,"identity":"99578bcd-6a56-4002-9261-23d6b896a13b","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":54502,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAcademic Performance and Social Media Engagement across Intervention Groups\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNotes.\u003c/em\u003e All mean differences and standard errors during the intervention phase. Broken line denotes linear regression. * BE-Social group without semi-immediate feedback.\u003c/p\u003e","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/e7833d85cf574443548a6634.png"},{"id":107929067,"identity":"8534d64f-5423-4365-83a9-417ca7d3148d","added_by":"auto","created_at":"2026-04-27 16:13:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":707816,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/c7f11363-104b-45de-bd8c-9a857da792eb.pdf"},{"id":96049628,"identity":"fd53ccaf-1e3b-463a-82ca-32d65a577aa0","added_by":"auto","created_at":"2025-11-17 06:34:45","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":28582,"visible":true,"origin":"","legend":"","description":"","filename":"JBESupplementarymaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7932578/v1/4d5dde2cace369a83fef3c6c.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Enhancing Academic Performance in Tertiary Education through Social Media: A Multi-Arm Randomized Controlled Trial","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eOnline-only instruction may undermine critical aspects of traditional education. For example, students attending online courses enjoy fewer opportunities for social interaction, which could be detrimental to student motivation and morale (Meşe \u0026amp; Sevilen, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Incorporating social media platforms into blended and online courses can supplement online education's otherwise limited social milieu (Irwin et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Moreover, several studies indicate that incorporating social media into university courses may provide robust communication and collaboration that may impact student experience and academic performance (Arteaga et al., 2014; Hamid et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Manca, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). For example, Alshuabi et al. (2018) reported that graduate students in courses with attached social media groups showed greater cognitive engagement and academic performance. The authors suggested that cognitive engagement in the social media environment may be a mediating factor leading to increased academic performance. Similarly, a survey by Graham (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) reported a positive correlation between increased social study group participation and curricular engagement. These and other findings obtained through self-reported surveys await verification with prospective experimental studies.\u003c/p\u003e\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e\u003ch2\u003e1.1 Social Media Platforms\u003c/h2\u003e\u003cp\u003eFacebook is one of the most popular social media channels among undergraduates and graduate students (Statista, 2024). Closed Facebook groups offer a private space to interact and share information instantly. Namely, receiving private and public messages and comments through push notifications, quickly identifying online members, commenting, and reacting to third-party posts, and receiving instructor moderation may prompt student engagement and participation (e.g., Mansholt et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Wang \u0026amp; Lin, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). X (formerly Twitter) has also been evaluated as a social network for academic purposes among graduate students (see, for example, Martinez-Cardama et al., 2019).\u003c/p\u003e\u003cp\u003eThe facility for immediate feedback may be a key component of the potential impact of social media platforms on the teaching-learning process. Gopal et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and Dhawan (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) have observed that providing feedback facilitates peer-to-peer and student-instructor interaction and increases student engagement with academic content. Instant messaging has become a natural means of interaction, leading to the notion that student-instructor interaction shall be equally fast-paced (Giannikas, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). However, excessive interactions can cause stress or lead to problematic internet use, which can be deleterious to academic performance (Azizi et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) or produce resistance to using social media in education among both students and educators (Cloete \u0026amp; Villiers, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Roblyer et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe balanced and safe use of social media aids in tertiary education demands the assessment of evidence-based educational practices that are amenable to this channel. Studies in this area have traditionally emphasized content (academic curriculum) rather than the range of evidence-based educational procedures that may be amenable to a social media channel (Mahdiuon et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). A further limitation of this literature is the almost complete lack of experimental research, which prevents the direct evaluation of educational strategies compatible with social media channels and can have a measurable impact on student engagement and performance. In an attempt to fill this vacuum, Tarifa-Rodriguez et al. (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) developed and evaluated the Behavioral Education and Social Media intervention package (the BE-Social Program) incorporating a collection of evidence-based behavioral intervention strategies selected based on their adaptability to the social media channel. For example, semi-immediate instructor feedback is available on any social network with real-time texting and commenting, whereas whole-class discussion may not. The authors also developed a taxonomy of quantitative engagement metrics to be integrated into the intervention outcomes (Tarifa-Rodriguez et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e1.2 Elements of a Social Media-Based Behavioral Education Intervention\u003c/h2\u003e\u003cp\u003eThe BE-Social program comprises self-management skills training delivered through video-modeling, cooperative learning environment, and semi-immediate instructor feedback (Tarifa-Rodriguez et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cdiv id=\"Sec4\" class=\"Section3\"\u003e\u003ch2\u003e1.2.1 Self-Management and Video-Modeling\u003c/h2\u003e\u003cp\u003eThe convenience and accessibility of online platforms allow instructors to teach critical self-management skills, including identifying short- and long-term objectives, prioritizing, and recording time devoted to key study behaviors (writing study notes, working on study questions, active memorization activities, etc.). The teaching objectives of the self-management component of the BE-Social program include: (a) select and define specific target behaviors, (b) observe the target behaviors and collect data, (c) chart the data and modify the definitions of the target behaviors accordingly, and (d) deliver self-reinforcement for small academic accomplishments (e.g., grade in a weekly test of 75% or higher) (Speidel et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e1981\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTeaching self-management strategies may empower students to assume a more active role, positively impacting personal satisfaction and academic performance (Rios et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Self-management skills have been studied extensively in the context of supporting academic engagement and performance in people with learning disabilities (Bahri et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). However, there have been few applications in tertiary education settings. In a notable exception, Choi and Chung (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) studied the effectiveness of self-control strategies for college-level students. We divided 84 participants into three experimental groups according to the intensity of the intervention or the number of strategies applied (high, medium, and low). Results indicated that students who had received a high-intensity intervention (i.e., several self-control strategies) demonstrated the best outcomes.\u003c/p\u003e\u003cp\u003eVideo-modeling consists of presenting a video of an expert performing a target skill with the expectation that the student will imitate the skill while or after viewing the video (Nikopoulos et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2016\u003c/span\u003e, p. 187). A meta-analysis by Bellini and Akullian (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) strongly indicated that video-modeling is an effective procedure for teaching communicative and functional skills to children and adolescents with autism spectrum disorders (ASD). In addition, video-modeling has been used to teach various motor skills. For example, Quinn et al. (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) reported that professional dancers receiving video-modeling and video feedback showed more rapid skill acquisition than those in the control group.\u003c/p\u003e\u003cp\u003eSocial media functionalities such as live broadcast, video storage and playback, and video chat, among others, suggest that video-modeling is readily compatible with the use of social media in educational settings. Specifically, social media video functionalities can help to model study habits, demonstrate course content elaboration routines, model problem-solving strategies, and other complex repertoires. We delivered the self-management skills training component of the BE-Social program via video-modeling.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section3\"\u003e\u003ch2\u003e1.2.2 Cooperative Learning\u003c/h2\u003e\u003cp\u003eCooperative learning typically involves elaborating academic content under the instructor's guidance, promoting student participation, and setting group expectations (Johnson \u0026amp; Johnson, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1982\u003c/span\u003e). Moreover, cooperative learning emphasizes group collaboration, peer-to-peer communication and assistance, and information and resource sharing (Roseth et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). In particular, cooperative learning environments often include discussion scenarios presented by instructors, interdependent contingencies, and social praise, strategies that are known to facilitate student participation (Popkin \u0026amp; Skinner, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). There is evidence to suggest that positive student interactions have a positive impact on academic performance (Van Ryzin et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Moreover, a meta-analysis by Capar and Tarim (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) concluded that cooperative learning environments significantly improve academic performance and student motivation.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section3\"\u003e\u003ch2\u003e1.2.3 Feedback\u003c/h2\u003e\u003cp\u003eImmediate feedback has shown a near-universal impact on skill acquisition and performance (Kluger \u0026amp; DeNisi, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). Immediate feedback may increase self-reported interest in academic content and lead to higher academic performance among college students (He et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). University students often use instant messaging applications to access academic content efficiently (Kaysi, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These findings emphasize the importance of incorporating instant peer-to-peer and instructor-student communication into online education programs, including those using social media as educational aids. The BE-Social package includes a semi-immediate feedback component whereby instructors respond to content-related student comments within a 24-hour cycle.\u003c/p\u003e\u003cp\u003eTarifa-Rodriguez et al. (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) randomly assigned 46 students to a default online program (control) or default online program plus BE-Social (intervention) group. Target outcomes included academic performance and social media engagement (reactions, comments). Results indicated that the intervention effectively produced a 20% increase in academic performance and significant increases in social media engagement. A parallel single-subject analysis revealed that intervention effects were, to a considerable extent, idiosyncratic. In the present study, we aim to replicate and expand the findings by Tarifa-Rodriguez et al. (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) by incorporating additional methodological standards: (1) increased sample size, (2) balanced number of participants across control and experimental groups, (3) balanced number of sessions across study phases (baseline, treatment and posttreatment), (4) comparable exposure intensity to academic content across groups, and, (5) a multi-arm RCT to evaluate the incremental validity of discrete elements of the BE-Social intervention program (self-management training via video-modeling, cooperative learning environment, and semi-intermediate instructor feedback).\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.1 \u003cem\u003eParticipants and setting\u003c/em\u003e\u003c/h2\u003e\u003cp\u003eParticipants were students enrolled in an online college-level course in applied psychology. Prospective participants received an email with a detailed research project description, an informed consent form, and a link to a sociodemographic information questionnaire. The curricular content was located on a Moodle platform, and students had weekly access to new materials. The course began in September 2020 and ended in May 2021. We invited 170 students who had relatively low performance in weekly academic tests over the first three months of the course (40% or lower on average). Of these, 141 students (mean age, 41.97; SD, 8.47) agreed to participate (see a summary of participant attrition in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Students were primarily immigrant working-class females with a mean of 16.45\u0026thinsp;\u0026plusmn;\u0026thinsp;5.44 years of education. The selection of students is relevant because social background can impact the probability of graduating from an online program (S\u0026aacute;nchez-Gelabert et al., 2023). Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the personal information of participants (see also Table A in the Supplementary Online Information for the sociodemographic information of the intention-to-treat group). Participants experienced various lockdowns and other COVID-related restrictions throughout the study. The study was approved by the ethics committee of the Universidad Aut\u0026oacute;noma de Madrid (ethics approval number CEI 112\u0026ndash;2204).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003e\u003cem\u003eSociodemographic Characteristics of Participants (n\u0026thinsp;=\u0026thinsp;141)\u003c/em\u003e\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTotal (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;141)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eG1 (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eG2 (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;33)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eG3 (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eG4 (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eG5 (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGender, % (\u003cem\u003en\u003c/em\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e90.07 (127)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e88.89 (24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e90.91 (30)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e96.30 (26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e96.30 (26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e77.78 (21)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e9.92 (14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e11.11 (3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6.06 (3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.70 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3.70 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e22.22 (6)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMean age, years (\u003cem\u003eSD\u003c/em\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e41.97 (8.47)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e40.11 (8.66)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e41.48 (8.34)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e42.78 (7.12)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e44.56 (9.47)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e41.04 (8.52)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYears of education (\u003cem\u003eSD\u003c/em\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e16.45 (5.44)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e17.70 (5.09)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16.67 (5.61)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e15.96 (5.27)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e17.44 (5.25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e14.41 (5.65)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMean BL performance (\u003cem\u003eSD\u003c/em\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e12.33 (2.18)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e12.35 (1.72)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e12.46 (2.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e12.58 (2.15)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e12.57 (2.52)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e11.67 (2.45)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCountry, % (\u003cem\u003en\u003c/em\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eU. S\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e88.65 (125)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e88.89 (24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e87.88 (29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e92.59 (25)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e96.30 (26)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e81.48 (22)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOthers\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e11.35 (16)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e11.11(3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e12.12 (4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e7.41 (2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3.70 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e18.52 (5)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSocioeconomic Status, % (\u003cem\u003en\u003c/em\u003e)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLower-middle\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e61.70 (87)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e30.56 (11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e64.71 (23)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e58.06 (18)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.18 (14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e60 (21)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMiddle\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2.84 (4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2.78 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.94 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.23 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e3.23 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0 (0)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUpper-middle\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e31.91 (45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e36.11 (11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e23.53 (9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e16.13 (5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e32.35 (11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e17.14 (6)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUpper class\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2.13 (3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e5.56 (2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0 (0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.23 (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0 (0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0 (0)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003e\u003cem\u003eNotes\u003c/em\u003e. G1\u0026thinsp;=\u0026thinsp;BE-Social; G2\u0026thinsp;=\u0026thinsp;BE-Social without semi-immediate feedback; G3\u0026thinsp;=\u0026thinsp;Cooperative learning; G4\u0026thinsp;=\u0026thinsp;Self-management; G5\u0026thinsp;=\u0026thinsp;Control group.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Experimental Design\u003c/h2\u003e\u003cp\u003eWe randomly assigned 141 participants to the five arms of the study. To ensure the effectiveness of the randomization process, we verified that age, gender, educational level, and socioeconomic standing were not statistically different across groups. Participants were blind to their group assignment status. We completed the study over 12 successive weeks (January 18 to April 11). The baseline phase comprised the first four weeks of the study. The intervention phase followed over the next four weeks. To control for a potential instructor effect, we divided the intervention into two two-week periods, and assigned semi-randomly each period within each group to an instructor. Participants returned to baseline conditions during the last phase of the study for another four weeks. Individuals in the control group remained in baseline conditions for the complete duration of the study.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.4. Procedure\u003c/h2\u003e\u003cp\u003eA 33-year-old male and a 32-year-old female, both doctoral students, created five closed Facebook groups with similar names (i.e., \"we learn together,\" \"together we learn,\" \"study group,\" \"study with me,\" and \"learn with me\") and served as instructors and moderators within these social media groups. Each group was assigned a different set of behavioral strategies: complete BE-Social program (Group 1), BE-Social without semi-immediate feedback (Group 2), cooperative learning (Group 3), and self-management (Group 4). The fifth group was the control group.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section3\"\u003e\u003ch2\u003e2.4.1. Baseline\u003c/h2\u003e\u003cp\u003eDuring the baseline weeks, four multiple-choice discussion questions were posted daily (Monday through Friday). All posted questions covered aspects of the course content taught during that week. The posts were identical for all groups. Instructors scheduled a weekend day to provide feedback on the twenty discussion questions posted over the week. Instructors provided feedback in the form of comments to the relevant posts. After providing weekly feedback, the instructor would then \"turn off commenting\" for each post. In addition, each instructor broadcasted one-hour weekly live videos featuring the instructor engaging in office-like activities (i.e., working on a computer, writing notes, and silently reading and writing flashcards). Students could watch these video broadcasts synchronously or asynchronously and imitate the instructor's behavior. Instructors did not interact with students during the video broadcasts. Baseline activities were intended to maintain student engagement (thereby preventing participant dropout) and equate the instructor inputs across control and intervention groups. The control group was under baseline conditions for the complete duration of the study.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section3\"\u003e\u003ch2\u003e2.4.2. Intervention\u003c/h2\u003e\u003cp\u003e\u003cem\u003eComplete BE-Social program\u003c/em\u003e (Group 1). Participants received all elements of the multi-component package (self-management via video-modeling, cooperative learning, and semi-immediate feedback). Each weekday, the instructors created three posts using all program components. One post involved a self-management strategy delivered through video-modeling, a second post required cooperative learning, and a third post provided the opportunity to receive semi-immediate instructor feedback. For example, one post would involve a video broadcast with the instructor modeling self-recording skills. A second post would present a practical problem to prompt student discussion (cooperative learning). Finally, a third post would consist of a multiple-choice scenario for which the instructor provided feedback to student comments. Instructor feedback for this last post was individualized and semi-immediate, meaning that the delay of the instructor's response was, on average, shorter than one hour. We describe the procedures of each of these program components below.\u003c/p\u003e\u003cp\u003eWe used self-management to encourage students to assume an active role in the organization and performance of academic tasks by fostering self-reliance and independence. The instructor focused on the critical elements of goal setting, self-monitoring, self-evaluation, and self-reinforcement. The instructor presented and modeled self-management skills through video-modeling sessions. To teach goal-setting skills, instructors made 10-min video broadcasts presenting a step-by-step approach to creating a list of specific, measurable, and achievable goals using specific course topics from the program syllabus to ensure relevance. The instructors demonstrated how to make a study planning schedule considering the time available, amount of study material, reading time, and time needed to complete assignments and self-assessment tests. The students could ask questions during or after the video broadcast and share their study planning schedule to receive feedback from the instructor. The self-monitoring component encouraged students to assess their progress as they worked toward their study goals. The instructors used a video broadcast to present self-recording strategies, which required students to time the duration of academic activities defined during the goal-setting phase. Then, the instructor demonstrated how to transfer daily study time records into a spreadsheet for graphing. The instructor added a weekly test performance data line to the graph and asked students to follow suit. Instructors encouraged students to check their weekly graphs and correlate study time with performance. Finally, instructors encouraged students to take extra free time (self-reinforcement) when they had improved their study time and performance. We implemented self-management procedures as an antecedent-based intervention with minimal contingent feedback beyond answering questions or providing sporadic feedback to students who presented their goal-setting schedule during a video broadcast (see Appendices 1 and 2 in the Supplementary Online Information for examples of self-management training via video-modeling posts).\u003c/p\u003e\u003cp\u003eThe goal of the cooperative learning component was to establish positive interactions among students by encouraging praise and informative feedback as part of their communication style. Cooperative learning posts could contain an image, a short video, a text, a combination of text and video, or a combination of text and image. Each post involved one of two possible activities: (a) presenting an applied scenario where multiple answers could be correct to generate discussion among students, and (b) presenting the outline of a student study time plan with blank spaces to generate cooperation among students to complete the proposed plan. The instructor further facilitated interaction by (a) presenting demand-free hints and prompts when communication halted (e.g., \"This scenario is so relevant to professional practice! Have you considered the potential ethical ramifications?\"), (b) tagging three students to each student's comment, (c) tagging students who had been inactive for the last 24 hours, and (d) providing weekly general feedback on the level of student communication and publicly praising those students who had given the most feedback to the rest of their peers. The instructor did not provide semi-immediate feedback as part of the cooperative learning procedures.\u003c/p\u003e\u003cp\u003e\u003cem\u003eBE-Social program without semi-immediate feedback\u003c/em\u003e (Group 2). The intervention procedures for this group were identical to those described for Group 1, except for the semi-immediate feedback component. Specifically, instructors provided feedback weekly over the weekend (as opposed to providing feedback within a 24-cycle or sooner).\u003c/p\u003e\u003cp\u003e\u003cem\u003eCooperative learning\u003c/em\u003e (Group 3). The intervention procedures for the cooperative learning group were identical to those described for the cooperative learning component of Group 1, with the exception that all three daily posts were cooperative learning posts. Therefore, they only received the cooperative learning component of the BE-Social program.\u003c/p\u003e\u003cp\u003e\u003cem\u003eSelf-management training via video-modeling\u003c/em\u003e (Group 4). The intervention procedures for the self-management training via video-modeling group were identical to those described for this component of the BE-Social group (Group 1), except that all three daily posts involved study self-management skills training delivered via video-modeling. Therefore, they only received the self-management component of the BE-Social program.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\u003ch2\u003e2.4.3 Post-Intervention\u003c/h2\u003e\u003cp\u003eThe post-intervention phase procedures were identical to those described for the baseline phase.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Dependent variables\u003c/h2\u003e\u003cp\u003eParticipants took a total of twelve weekly tests throughout the study. Each test included 20 multiple-choice questions with three distractors and one correct answer. We calculated test scores by dividing correct responses by 20 and converting that ratio into a percentage. Ninety percent of the scenarios described practical applications of the course contents, while the remainder of the questions focused on conceptual information (e.g., definitions, characteristics, conceptual clarification, etc.). To minimize the effect of content difficulty on the average performance of participants, the online tests, published weekly, could be completed over a three-week time window.\u003c/p\u003e\u003cp\u003eWe recorded the number of students that had viewed any instructor post (views) and the total number of student comments and reactions (e.g., likes) generated by each instructor post. We summarized these data as \u003cem\u003ecombined engagement\u003c/em\u003e, defined as the sum of views, likes, and comments per post.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e2.5. Interobserver agreement and procedural integrity\u003c/h2\u003e\u003cp\u003eA secondary observer obtained data on student engagement (views, reactions, and comments) and instructor latency in resolving or responding to students. Twenty percent of instructor posts were selected from each of the five groups to assess interobserver agreement (IOA). We computed IOA by dividing the number of posts with agreement by the total number of posts selected for evaluation. Overall, our IOA reached 99.6% (range, 59% to 100%), 99.9% (96% to 100%), 100%, and 100% for instructor feedback latency, views, comments, and likes, respectively.\u003c/p\u003e\u003cp\u003eTo document procedural integrity, an independent observer recorded critical aspects of the intervention procedures throughout all study phases, including the number of daily posts, number of weekly videos, latency to instructor feedback (weekly vs. semi-immediate), and video content (delivery of self-management strategies). The independent observer conducted a 60-s partial interval recording of 20% of instructor video broadcasts to monitor video content. The observation protocol helped to verify the amount of instructor time devoted to discussing self-management skills. Observations were conducted with the app Big Eye Observer, validated for video-based behavioral observation applications (Virues-Ortega et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Instructors also recorded the following intervention-specific procedural integrity standards during the intervention phase: number of self-management posts, format of self-management posts (video vs. text), cooperative learning posts, and mean latency of instructor feedback. A comprehensive summary of the procedural integrity assessment is available in the online supplementary information (Table B).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e2.6. Statistical analysis\u003c/h2\u003e\u003cdiv id=\"Sec17\" class=\"Section3\"\u003e\u003ch2\u003e2.6.1. Academic achievement\u003c/h2\u003e\u003cp\u003eWe conducted a mixed-model ANOVA with group as a between-groups factor and time (Time 1: Baseline; Time 2: Intervention; Time 3: Post-Intervention) as a within-subjects factor. The dependent variable was the mean performance in weekly course content tests (range, 0 to 20). Achieved power and effect size metrics are reported for the omnibus tests. In addition, we computed \u003cem\u003et\u003c/em\u003e-test pairwise group and time comparisons and estimated Hedges \u003cem\u003eg\u003c/em\u003e effect sizes for any effects identified through the omnibus mixed-model ANOVA.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section3\"\u003e\u003ch2\u003e2.6.2. Social media engagement\u003c/h2\u003e\u003cp\u003eThe Facebook platform does not provide access to the user's identity visualizing a video post (only the aggregate number of visualizations is available). This constraint meant we should use posts as units of analysis (as opposed to subjects within a group) to study engagement responses. The number of engagement responses (i.e., views, likes, comments) across posts was not normally distributed. This, added to the longitudinal (i.e., successive posts within a group) and hierarchical structure (i.e., group-level predictors) of the engagement responses, favored the use of hierarchical linear models (HLM) in our analysis. We used combined engagement, defined as the sum of views, likes, and comments per post, as our dependent variable. We used the group identifier as the subject variable and the post number within a group as the time-based variable. A first-order autoregressive covariance structure rendered the best goodness of fit values during the model development process. We used group (BE-Social, BE-Social without semi-immediate feedback, self-management, cooperative learning, control), time (baseline, treatment, posttreatment), post type (textual, visual, video), and the interaction between group and post type as fixed-effect factors. In addition, we added the group-level mean academic performance as a group-nested random-effects factor. Akaike's information criterion (AIC) of the unconditional model was reduced gradually by the stepwise addition of each factor (Akaike, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1974\u003c/span\u003e; Burhan \u0026amp; Anderson, 2002). The unconditional and terminal models' AIC values were 6991 and 6700, respectively. The addition of a random intercept did not improve model fitness. Group-nested academic performance correlated significantly with phase and group and was the subject of a second model without apparent confounders. We used the metric of the dependent variable for all model coefficients (range, 0 to 1).\u003c/p\u003e\u003cp\u003eWe used SPSS\u0026reg; IBM\u0026reg; Statistics, version 27 (IBM Corporation, 2021) for all analyses. We used a \u003cem\u003ep\u003c/em\u003e value of 0.05 with Bonferroni adjustments for multiple comparisons.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003e2.7. Social Validity and Acceptability\u003c/h2\u003e\u003cp\u003eAt the end of the study, the instructors made a live farewell video for each group, thanking students for their participation and inviting comments about their experiences and concerns. We recorded the unsolicited comments of the live-streamed farewell videos. We imported these comments into a textual analysis tool to identify common semantic themes or \u003cem\u003ecategories\u003c/em\u003e from textual samples (Hunerberg, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). We report the percentage of unsolicited positive feedback and the themes resulting from the textual analysis as indirect indications of the intervention's social validity (i.e., acceptability).\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Academic achievement\u003c/h2\u003e\u003cp\u003eThe Box test for the mixed-model ANOVA was not significant (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.179), suggesting that the assumption of the equality of covariance matrices was met. Multivariate tests following the Wilks' lambda distribution revealed a significant Time effect (\u003cem\u003eF\u003c/em\u003e [2, 135]\u0026thinsp;=\u0026thinsp;66.11, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.50, 1-\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.00) and Time by Group interaction (\u003cem\u003eF\u003c/em\u003e [8, 270]\u0026thinsp;=\u0026thinsp;4.06, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.12, 1-\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.99). The corrected within-subject effect for Time (\u003cem\u003eF\u003c/em\u003e [1.69, 229.11]\u0026thinsp;=\u0026thinsp;42.92, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.24, 1-\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.00) and Time by Group interaction (\u003cem\u003eF\u003c/em\u003e [6.74, 229.11]\u0026thinsp;=\u0026thinsp;3.25, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.003, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.09, 1-\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.95) were also significant. The main effect of Group was also significant, \u003cem\u003eF\u003c/em\u003e (4, 136)\u0026thinsp;=\u0026thinsp;4.63, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.002, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.12, 1-\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.94. The homogeneity assumption for the equality of error variances across groups was met. There was a main effect of Group as well, \u003cem\u003eF\u003c/em\u003e (4, 136)\u0026thinsp;=\u0026thinsp;126.05, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.002, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.12, 1-\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.94. As it seems apparent from Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, within-subject contrasts indicated, the effect of time followed a quadratic rather than a linear trend (\u003cem\u003eF\u003c/em\u003e [1, 136]\u0026thinsp;=\u0026thinsp;7.73, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.006, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.05 vs. \u003cem\u003eF\u003c/em\u003e [1, 136]\u0026thinsp;=\u0026thinsp;130.11, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001, \u003cem\u003eη\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.49) suggesting that the intervention effects were to a certain degree transient for the intervention groups. In contrast, the control group followed a deteriorating trend over time (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe pairwise \u003cem\u003et\u003c/em\u003e-test comparisons indicated that the baseline academic performance in any of the intervention groups was not statistically different from that of the control group, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.01 (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The only intervention group that did not differ from the control group during the intervention phases was the self-management intervention group (Group 4). All other intervention groups showed statistically significant departures in academic performance, both during the intervention (Time 2) and post-intervention phases (Time 3) (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Specifically, the complete BE-Social group (Group 1) showed a statistically significant increase in performance during the intervention, \u003cem\u003et\u003c/em\u003e(1. 52)\u0026thinsp;=\u0026thinsp;5.38, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001, and post-intervention phases, \u003cem\u003et\u003c/em\u003e(1. 52)\u0026thinsp;=\u0026thinsp;3.53, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001 relative to the control group. The effect sizes were both within the large effect size range (\u003cem\u003eg\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.44, 98.75% CI 0.68, 2.97 and \u003cem\u003eg\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.95, 95% CI 0.23, 1.65, respectively). The slightly lower post-intervention effect suggests that the effect lessened after the intervention withdrawal.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003e\u003cem\u003ePairwise t Test Comparisons for Academic Achievement across Groups\u003c/em\u003e\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eGroup\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"4\" nameend=\"c7\" namest=\"c4\"\u003e\u003cp\u003ePairwise \u003cem\u003et\u003c/em\u003e tests\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cem\u003eMean\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cem\u003eSD\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003et\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003edf\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eHedges \u003cem\u003eg\u003c/em\u003e\u003c/p\u003e\u003cp\u003e[98.75% \u003cem\u003eCI\u003c/em\u003e]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eG1: BE-Social\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.102\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.45 [-0.23, 1.12]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e5.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1.44 [ 0.68, 2.97]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.95 [ 0.23, 1.65]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eG2: BE-Social*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e12.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.13 [-0.54, 0.80]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14.86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.98 [ 0.27, 1.69]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.61 [-0.08, 1.29]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eG3: Cooperative learning\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e12.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.174\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.35 [-0.29, 0.99]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13.55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.64 [-0.02, 1.29]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.90\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.018\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.63 [-0.03, 1.28]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eG4: Self-Management\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e12.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.17 [-0.49, 0.84]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e13.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.103\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.45 [-0.24, 1.12]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1, 52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e.097\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.45 [-0.23, 1.13]\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eG5: Control\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9.32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003e\u003cem\u003eNote\u003c/em\u003e. All pairwise comparisons against the control group. Bonferroni adjusted confidence intervals. CI\u0026thinsp;=\u0026thinsp;confidence interval. * BE-Social group without semi-immediate feedback. Time 1\u0026thinsp;=\u0026thinsp;Baseline; Time 2\u0026thinsp;=\u0026thinsp;Intervention; Time 3\u0026thinsp;=\u0026thinsp;Post-intervention.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe BE-Social without semi-immediate feedback group (Group 2) showed a statistically significant increase in performance during the intervention (\u003cem\u003et\u003c/em\u003e[1, 52]\u0026thinsp;=\u0026thinsp;3.67, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and post-intervention phases (\u003cem\u003et\u003c/em\u003e[1, 52]\u0026thinsp;=\u0026thinsp;2.29, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03). The pairwise effect size of Group 2 against the control group for the intervention and post-intervention phases was 0.98 (95% CI 0.27, 1.69) and 0.61 (95% CI -0.08, 1.29), respectively.\u003c/p\u003e\u003cp\u003eFor the cooperative learning group (Group 3), we observed a slight increase in performance during the intervention and post-intervention phases (\u003cem\u003et\u003c/em\u003e([, 58]\u0026thinsp;=\u0026thinsp;2.48, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.016 and \u003cem\u003et\u003c/em\u003e[1, 58]\u0026thinsp;=\u0026thinsp;2.44, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.018, respectively). However, using the control group as a reference, the intervention effect for Group 3 was somewhat smaller than the one observed for Groups 1 and 2. Group 3 showed nearly identical effect sizes within the moderate effect size range for both the intervention and post-intervention phases (\u003cem\u003eg\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.64, 98.75% CI -0.02, 1.29 and \u003cem\u003eg\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.63, 95% CI -0.03, 1.28, respectively). Therefore, the effect of the intervention endured during the post-intervention phase. However, for all other groups, performance deteriorated during the post-intervention phase (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFinally, the pairwise comparison of the self-management skills training group (Group 4) relative to the control group did not reveal statistically significant differences during the intervention and post-intervention phases (\u003cem\u003et\u003c/em\u003e[1, 52]\u0026thinsp;=\u0026thinsp;1.61, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.097 and \u003cem\u003et\u003c/em\u003e[1, 58]\u0026thinsp;=\u0026thinsp;1.69, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.016, respectively).\u003c/p\u003e\u003cp\u003eOverall, the multi-arm RCT analysis indicated that semi-immediate feedback was critical to the effect of the multi-component intervention. In contrast, the unimodal intervention could not verify the value of self-management skills training.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003e6.2 Social Media Engagement\u003c/h2\u003e\u003cp\u003eThe final linear mixed model analysis for combined social media engagement (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) confirmed a significant effect of Group (\u003cem\u003eF\u003c/em\u003e[1, 800]\u0026thinsp;=\u0026thinsp;16.69, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), Time (\u003cem\u003eF\u003c/em\u003e[2, 288]\u0026thinsp;=\u0026thinsp;51.79, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), Post type (\u003cem\u003eF\u003c/em\u003e[2, 859]\u0026thinsp;=\u0026thinsp;36.28, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001), and Group by Post type interaction (\u003cem\u003eF\u003c/em\u003e[8, 880]\u0026thinsp;=\u0026thinsp;8.48, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). Pairwise comparisons indicated that only the complete BE-Social program (Group 1) and the BE-Social program without semi-immediate feedback (Group 2) had higher levels of social media engagement relative to the control group (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{D}\\)\u003c/span\u003e\u003c/span\u003e = 7.78\u0026thinsp;\u0026plusmn;\u0026thinsp;2.51, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.008 and\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{D}\\)\u003c/span\u003e\u003c/span\u003e = 7.66\u0026thinsp;\u0026plusmn;\u0026thinsp;2.54, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.01, respectively). Engagement levels across groups did not change during the intervention phase relative to baseline, whereas engagement decreased across groups during the post-intervention phase relative to baseline (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{D}\\)\u003c/span\u003e\u003c/span\u003e = 8.46\u0026thinsp;\u0026plusmn;\u0026thinsp;0.91, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). Finally, textual posts induced greater interaction across groups than visual (e.g., image with text) or video posts. Still, only the mean difference between textual and video posts was statistically significant in the pairwise analysis (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{D}\\)\u003c/span\u003e\u003c/span\u003e = 9.59\u0026thinsp;\u0026plusmn;\u0026thinsp;1.14, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.001). While a mediation analysis was not feasible, there seems to be an apparent correlation between social media engagement and academic performance, as expressed in the mean differences of both outcomes during the intervention phase (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003e\u003cem\u003eLinear Mixed Effects Model for Combined Engagement\u003c/em\u003e\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFixed effects\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cem\u003eF\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cem\u003edf\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e(Intercept)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2127.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 799.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGroup\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e19.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4, 628.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTime\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e51.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2, 287.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePost type\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e36.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2, 858.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGroup by Post type\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8, 880.13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePairwise comparisons, \u003cem\u003ei - j\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{D}\\pm\\:SE\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cem\u003edf\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBE-Social - Control\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.78\u0026thinsp;\u0026plusmn;\u0026thinsp;2.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 909.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e.008\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBE-Social* - Control\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.66\u0026thinsp;\u0026plusmn;\u0026thinsp;2.54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 906.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e.010\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCooperative learning - Control\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.16\u0026thinsp;\u0026plusmn;\u0026thinsp;2.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 911.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSelf-management - Control\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-4.09\u0026thinsp;\u0026plusmn;\u0026thinsp;2.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 889.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e.389\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTreatment - Baseline\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.11\u0026thinsp;\u0026plusmn;\u0026thinsp;0.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 308.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e.449\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePosttreatment - Baseline\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-8.46\u0026thinsp;\u0026plusmn;\u0026thinsp;0.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 233.41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eText - Visual\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.78\u0026thinsp;\u0026plusmn;\u0026thinsp;1.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 870.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e.124\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eText - Video\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9.59\u0026thinsp;\u0026plusmn;\u0026thinsp;1.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1, 854.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003cem\u003eNotes\u003c/em\u003e. \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{D}\\pm\\:SE\\)\u003c/span\u003e\u003c/span\u003e = mean differences \u0026plusmn; standard errors. * BE-Social group without semi-immediate feedback.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec23\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Social Validity and Acceptability\u003c/h2\u003e\u003cp\u003eForty-six participants provided unsolicited positive comments during the farewell video broadcasts. Therefore, 32.6% of participants produced unsolicited positive comments. The BE-Social without semi-immediate feedback group provided the most positive comments (51.8%), followed by the complete BE-Social program group (44.5%). Only 18% of participants in the self-management and cooperative learning groups volunteered positive comments. Finally, 33% of participants in the control group shared positive comments. We did not see negative or derogatory comments posted in any of the groups (see Supplementary Online Material, Table C). The sentence-by-sentence thematic analysis indicated that the five most prevalent themes in the feedback messages were \u003cem\u003eknowledge acquisition\u003c/em\u003e (35 occurrences, 7.5% of the text), \u003cem\u003ehigh-quality teaching\u003c/em\u003e (27 occurrences, 5.8% of the text), \u003cem\u003ecommendation for the video broadcasts\u003c/em\u003e (14 occurrences, 3.0% of the text), \u003cem\u003eappreciation for the problem-solving strategies presented\u003c/em\u003e (10 occurrences, 2.2% of the text), and \u003cem\u003eperceived value of the experience\u003c/em\u003e (9 occurrences, 1.9% of the text).\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eThe present study expands the evidence base of social media platforms to aid tertiary education. We implemented the multi-component BE-Social program as an adjunct to an online applied psychology course to assess the efficacy of the key elements of this program individually and in combination. The results suggest that participants exposed to the complete intervention package (including semi-immediate feedback) demonstrated the highest academic performance. While semi-immediate feedback did not affect the total social media engagement, the current multi-arm RCT seems to suggest that feedback was a critical element of the program. Our results also indicate that part of the intervention gains may fade after the intervention is withdrawn, thereby suggesting that the program ought to be present for the duration of the course for optimal results. The isolated implementation of discrete components of the program (particularly self-management training) had relatively minor effects on performance.\u003c/p\u003e\u003cp\u003eOur findings align with the empirical literature suggesting that fast turn-around feedback is conducive to student learning. For example, a systematic review by Liu (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) found that the immediacy of teacher responses positively impacted student engagement with the academic material and, ultimately, on performance. In addition, immediacy has been identified as a critical dimension of effective feedback (Er et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Henderson \u0026amp; Wen, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1976\u003c/span\u003e). Also, evidence suggests that students prefer receiving feedback from educators through instant messaging instead of email (Gopal et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). While our study does not demonstrate reward learning effects at the individual level, these findings are in line with a delay-discounting effect that is amply documented in the animal and human operant literatures (Reynolds, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Specifically, a relatively immediate reward (e.g., immediate instructor feedback) may have a greater impact on engagement and other relevant academic behaviors than more delayed ones.\u003c/p\u003e\u003cp\u003eInterestingly, students in the BE-Social group without semi-immediate feedback (Group 2) showed high levels of social media engagement. At the same time, their academic achievement was relatively lower than the one observed in the complete BE-Social group (Group 1). This could indicate that social media engagement may not accurately predict course content elaboration. An alternative interpretation may be that, although frequent student posting can still occur under delayed instructor feedback (for example, because of peer-to-peer interactions), it may be less likely to induce successful content acquisition and elaboration.\u003c/p\u003e\u003cp\u003eSocial media engagement seems to follow simple social reward optimization processes (Lindstr\u0026ouml;m et al., 2021). Such realization has obvious practical implications that we evaluated only indirectly. Specifically, our analysis suggests that cooperative learning and semi-immediate feedback strategies may be particularly effective in mobilizing student online behavior in study groups. While the mediating role of social media responses in academic achievement cannot be asserted, the study provides observational evidence that those groups that showed more active online behavior had relatively higher academic achievement. We also illustrated the integration of multiple forms of immediate social media interactions (e.g., push notifications, tagging, reacting, commenting, etc.) into evidence-based educational practices, including the design of a cooperative learning environment, the delivery of self-management skills training, and the effective use of instructor-mediated feedback.\u003c/p\u003e\u003cp\u003eThe analytical approach of the multi-arm RCT suggests that the elements of the proposed multi-component intervention are likely to be additive, with the cooperative learning and semi-immediate feedback manipulations being the most likely contributors to the compounded effect observed in the full implementation of the intervention package. The effect of self-management training on performance (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) was not statistically significant. Given the current multi-arm RCT evaluated discrete components of the intervention, whether isolated or in combination, additional analyses would be needed to substantiate the potential additivity of the critical elements of the BE-Social package (i.e., self-management training, cooperative learning environment, instructor-mediated semi-immediate feedback). While an \u003cem\u003eadditive effect hypothesis\u003c/em\u003e can hardly be verified with between-groups datasets (see, for example, Van Iddekinge et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), we can tentatively examine the mismatch between empirical effect sizes and the theoretical additive effects. Specifically, a plausible additive effect would mean that the sum of the unimodal interventions without semi-immediate feedback would approach the effect size of the BE-Social group without semi-immediate feedback,\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:{g}_{3}+{g}_{4}\\approx\\:{g}_{2}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:0.64\\:+\\:0.45\\approx\\:\\:0.98$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{g}_{2}\\)\u003c/span\u003e\u003c/span\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{g}_{3}\\)\u003c/span\u003e\u003c/span\u003e, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{g}_{4}\\)\u003c/span\u003e\u003c/span\u003e are the pairwise comparison Hedges \u003cem\u003eg\u003c/em\u003e effect size for Groups 2, 3, and 4, respectively. Similarly, the incremental effect of semi-immediate feedback and the separate effects of the unimodal interventions (Group 3 and Group 4) should approach the effect size of the full multi-component intervention (Group 1),\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$$\\:{g}_{3}+{g}_{4}+\\:({g}_{1}-{g}_{2})\\approx\\:{g}_{1}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:0.64+0.45+\\:(1.44-0.98)\\approx\\:1.44$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{g}_{1},\\:{g}_{2},\\:{g}_{3},\\:{g}_{4}\\)\u003c/span\u003e\u003c/span\u003e, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{g}_{5}\\:\\)\u003c/span\u003e\u003c/span\u003eare the pairwise comparison Hedges \u003cem\u003eg\u003c/em\u003e effect sizes for Groups 1 through 5, respectively. The observed empirical effect sizes match the theoretical additive relation within 0.1 effect size units in both vases (0.98 vs. 1.09 and 1.44 vs. 1.55, respectively). While this demonstration is anecdotal, it does suggest that the multi-arm approach may be an analytic tool to weigh the relative contribution of multi-component educational programs.\u003c/p\u003e\u003cp\u003eThe results show that combining all the selected educational strategies may have a synergic effect. When applied individually, each intervention component seems to detract from the effect of the complete program. However, the current analytical approach was not comprehensive. Specifically, we could not assess semi-immediate feedback without other intervention procedures. Moreover, the control group baseline condition involved minimal interactions deemed essential to maintain student morale and engagement. Critical comparisons involved adding elements to a background intervention rather than comparing the absence of intervention to a discrete intervention component. Therefore, we cannot discard complex interactions between the instructor's baseline behavior and subsequent performance.\u003c/p\u003e\u003cp\u003eInterestingly, instructor textual posts generated more engagement than video broadcasts (i.e., self-management training via video-modeling posts) or mixed posts with text and images. While studies that quantitatively evaluate engagement are rare, some surveys suggest that text-based posting generates more interactive engagement. In contrast, video-based and mixed posts seem to impose a one-sided communication dynamic (see, for example, Swartzwelder et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The nature of the content may also play a role. For example, text-based posts presented multiple-choice scenarios that frequently prompted students to guess and follow up with questions.\u003c/p\u003e\u003cp\u003eWe should note several limitations and future extensions to the current. First, we have examined the compatibility of an incidental collection of behavioral education strategies using one of the most popular social media platforms for tertiary education students. Future studies could replicate our findings with alternative combinations of evidence-based strategies and social media platforms. The specific parameters of the strategies utilized here could be adjusted (e.g., elements of the cooperative learning and self-management protocols). For example, adding performance-based feedback instead of relying primarily on video modeling could improve the effect of the self-management protocol.\u003c/p\u003e\u003cp\u003eSecond, our data collection strategy did not allow us to reconcile discrete online events (e.g., reactions, views, posts, comments) with individual students, which meant that the impact of the intervention could only be assessed at the group level, restricting the possibility of mediation analysis. This also meant that idiosyncratic patterns of treatment effects could not be studied in detail (for an analysis of the idiosyncratic effects of the BE-Social program, see Tarifa-Rodriguez et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Future studies in this area could minimize these concerns by purposely developing application programming interfaces that would automate aspects of the data collection process in the social media environment. Additional practical enhancements include integrating text and theme analysis systems to characterize further social dynamics and emotional factors involved in social media exchanges in educational contexts (Drus \u0026amp; Khalid, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Moreover, adding and evaluating AI-powered feedback could greatly minimize intervention costs and facilitate the program's deployment at scale (see, for example, Escalante et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThird, our outcome measures were limited to \u003cem\u003ead hoc\u003c/em\u003e course content tests and standard social media responses (i.e., views, likes, comments, posts). These outcomes were important because they allowed frequent measurements across study phases and were ecologically valid and naturally integrated with the course. However, future analyses would benefit from adding standardized measurements of student satisfaction and end-of-course assessments. The latter was not practical in the current study owing to the multi-phase structure of the design, which had to fit within a one-year course. While frequent testing allowed the opportunity to monitor the intervention effects over time, it prevented a more comprehensive evaluation of the program's overall impact on student academic achievement.\u003c/p\u003e\u003cp\u003eFourth, the apparent deteriorating academic achievement trend observed in the control group may be due to selecting low-performance students (see inclusion criteria) or to the hierarchical nature of course contents (i.e., latter lessons relied on the foundational knowledge introduced in earlier ones). It is interesting to note that the elements of the BE-Social program were sufficient to offset this deteriorating trend, even though intervention gains were not evident during the post-intervention phase for those elements.\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eOur study supports a few tentative conclusions: (1) the unique combination of self-management training via video-modeling, cooperative learning online environment, and semi-immediate instructor feedback included in the BE-Social program had a large positive impact on social media engagement and academic achievement, (2) the effect of the elements of the intervention seems to be additive, although we could only verify the incremental validity of semi-immediate feedback, (3) increased social media engagement may be a mediating factor to increased academic achievement, and (4) the multi-component intervention may have optimal effects on achievement when delivered for the complete duration of a semester or year-long course.\u003c/p\u003e\u003cp\u003eThe exponential growth of the casual use of social media in tertiary education calls for an extensive evaluation of social media as a medium for evidence-based educational practices. Integrating social media study groups in tertiary education courses may be a cost-effective approach to enhancing peer- and instructor-mediated interactions and improving student experience and academic achievement. The proposed program demonstrated that key evidence-based educational practices amenable to a social media study group can support student participation and performance.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eDeclaration of generative AI and AI-assisted technologies in the writing process.\u0026nbsp;\u003c/strong\u003eThe authors did not use\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eAI-assisted technologies in the process of writing this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material.\u003c/strong\u003e The complete databased use for all analyses will be made public through the platform \u003cem\u003eFigshare\u003c/em\u003e upon the manuscript\u0026apos;s acceptance for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding.\u003c/strong\u003e This study received financial support from the research contracts RYC-2016-20706 (Ramon y Cajal Program, Spain) and CON02739 (The University of Auckland, New Zealand).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e. We thank Agust\u0026iacute;n Perez-Bustamante for his assistance with data coding.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eATR. This study was part of the requirements for the Doctor of Psychology degree of the first author at the Universidad Aut\u0026oacute;noma de Madrid (Spain). Research design development. Data collection and data curation. Data analysis design. Manuscript writing and editing (first draft). JVO. Research design development. Data analysis design. Logistics and resources. Funding procurement. Manuscript writing and editing. Doctoral supervision of ATR.ACE. Research design development. Manuscript writing and editing. Doctoral supervision of ATR.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAli, W. (2020). Online and remote learning in higher education institutes: a necessity in light of COVID-19 pandemic. \u003cem\u003eHigher Education Studies, 10\u003c/em\u003e(3), 16. https://doi.org/10.5539/hes.v10n3p16\u003c/li\u003e\n\u003cli\u003eAlkatout, I., Dhanawat, J., Ackermann, J., Freytag, D., Peters, G., Maass, N., Mettler, L., \u0026amp; Pape, J. M. (2021). Video feedback and video-modeling in teaching laparoscopic surgery: a visionary concept from Kiel. \u003cem\u003eJournal of Clinical Medicine, 10\u003c/em\u003e(1), 163. https://doi.org/10.3390/jcm10010163\u003c/li\u003e\n\u003cli\u003eAlshuaibi, M. S. I., Alshuaibi, A. S. I., Shamsudin, F. M., \u0026amp; Arshad, D. A. (2018). Use of social media, student engagement, and academic performance of business students in Malaysia. \u003cem\u003eInternational Journal of Educational Management, 32\u003c/em\u003e(4), 625\u0026ndash;640. https://doi.org/10.1108/IJEM-08-2016-0182\u003c/li\u003e\n\u003cli\u003eAkaike, H. (1974). A new look at the statistical model identification. \u003cem\u003eIEEE Transactions on Automatic Control, 19\u003c/em\u003e(6), 716\u0026ndash;723. https://doi.org/10.1109/TAC.1974.1100705\u003c/li\u003e\n\u003cli\u003eArteaga S\u0026aacute;nchez, R., Cortijo, V., \u0026amp; Javed, U. (2014). Students\u0026apos; perceptions of Facebook for academic purposes. \u003cem\u003eComputers \u0026amp; Education, 70\u003c/em\u003e, 138\u0026ndash;149. https://doi.org/10.1016/J.COMPEDU.2013.08.012\u003c/li\u003e\n\u003cli\u003eAzizi, S. M., Soroush, A., \u0026amp; Khatony, A. (2019). The relationship between social networking addiction and academic performance in Iranian students of medical sciences: a cross-sectional study. \u003cem\u003eBMC Psychology, 7\u003c/em\u003e(1), 28. https://doi.org/10.1186/s40359-019-0305-0 \u003c/li\u003e\n\u003cli\u003eBahri, L., Mirnasab, M., Noorazar, G., Fathi Azar, E., \u0026amp; Asadi, S. (2016). The effect of self-management strategies on improving task performance in students with learning disabilities. \u003cem\u003eJournal of Analytical Research in Clinical Medicine, 4\u003c/em\u003e(1), 53\u0026ndash;65. https://doi.org/10.15171/jarcm.2016.009\u003c/li\u003e\n\u003cli\u003eBellini, S., \u0026amp; Akullian, J. (2007). A meta-analysis of video-modeling and video self-modeling interventions for children and adolescents with autism spectrum disorders. \u003cem\u003eExceptional Children, 73\u003c/em\u003e(3), 264\u0026ndash;287. https://doi.org/10.1177/001440290707300301\u003c/li\u003e\n\u003cli\u003eBertucci, A., Conte, S., Johnson, D. W., \u0026amp; Johnson, R. T. (2010). The impact of size of cooperative group on achievement, social support, and self-esteem. \u003cem\u003eJournal of General Psychology, 137\u003c/em\u003e(3), 256\u0026ndash;272. https://doi.org/10.1080/00221309.2010.484448\u003c/li\u003e\n\u003cli\u003eBurnham, K. P., \u0026amp; Anderson, D. R. (2002). \u003cem\u003eModel Selection and Multimodel Inference\u003c/em\u003e (2\u003csup\u003end\u003c/sup\u003e ed.). Springer.\u003c/li\u003e\n\u003cli\u003eCapar, G., \u0026amp; Tarim, K. (2015). Efficacy of the cooperative learning method on mathematics achievement and attitude: A meta-analysis research. \u003cem\u003eEducational Sciences: Theory \u0026amp; Practice, 15\u003c/em\u003e(6), 553\u0026ndash;559. https://doi.org/10.12738/estp.2015.2.2098\u003c/li\u003e\n\u003cli\u003eCloete, S., \u0026amp; Villiers, C. de. (2009). Facebook as an academic tool for ICT lecturers. \u003cem\u003eProceedings of the 2009 Annual Conference of the Southern African Computer Lecturers\u0026apos; Association,\u003c/em\u003e 16-22. https://doi.org/10.1145/1562741.1562743\u003c/li\u003e\n\u003cli\u003eChoi, J. H., \u0026amp; Chung, K. M. (2012). Effectiveness of a college-level self-management course on successful behavior change. \u003cem\u003eBehavior Modification, 36\u003c/em\u003e(1), 18\u0026ndash;36. https://doi.org/10.1177/0145445511418102\u003c/li\u003e\n\u003cli\u003eCooper, J. O., Heron, T. E., \u0026amp; Heward, W. L. (2019). \u003cem\u003eApplied Behavior Analysis\u003c/em\u003e (3\u003csup\u003erd\u003c/sup\u003e ed.). Pearson.\u003c/li\u003e\n\u003cli\u003eDhawan, S. (2020). Online learning: a panacea in the time of COVID-19 crisis. \u003cem\u003eJournal of Educational Technology Systems, 49\u003c/em\u003e(1), 5\u0026ndash;22. https://doi.org/10.1177/0047239520934018\u003c/li\u003e\n\u003cli\u003eDrus, Z., \u0026amp; Khalid, H. (2019). Sentiment analysis in social media and its application: Systematic literature review. \u003cem\u003eProcedia Computer Science, 161\u003c/em\u003e, 707\u0026ndash;714. https://doi.org/10.1016/j.procs.2019.11.174\u003c/li\u003e\n\u003cli\u003eEr, E., Dimitriadis, Y., \u0026amp; Ga\u0026scaron;ević, D. (2021). A collaborative learning approach to dialogic peer feedback: a theoretical framework. \u003cem\u003eAssessment and Evaluation in Higher Education, 46\u003c/em\u003e(4), 586\u0026ndash;600. https://doi.org/10.1080/02602938.2020.1786497\u003c/li\u003e\n\u003cli\u003eEscalante, J., Pack, A., \u0026amp; Barrett, A. (2023). AI-generated feedback on writing: Insights into efficacy and ENL student preference. \u003cem\u003eInternational Journal of Educational Technology in Higher Education, 20\u003c/em\u003e, 57. https://doi.org/10.1186/s41239-023-00425-2\u003c/li\u003e\n\u003cli\u003eGiannikas, C. (2019). Facebook in tertiary education: the impact of social media in e-learning. \u003cem\u003eJournal of University Teaching and Learning Practice, 17\u003c/em\u003e(1). https://doi.org/10.53761/1.17.1.3\u003c/li\u003e\n\u003cli\u003eGopal, R., Singh, V., \u0026amp; Aggarwal, A. (2021). Impact of online classes on the satisfaction and performance of students during the pandemic period of COVID 19. \u003cem\u003eEducation and Information Technologies,\u003c/em\u003e \u003cem\u003e26\u003c/em\u003e, 6923\u0026ndash;6947. https://doi.org/10.1007/s10639-021-10523-1\u003c/li\u003e\n\u003cli\u003eGraham, M. (2014). Social media as a tool for increased student participation and engagement outside the classroom in higher education. \u003cem\u003eJournal of Perspectives in Applied Academic Practice, 2\u003c/em\u003e(3), 16\u0026ndash;24. https://doi.org/10.14297/jpaap.v2i3.113\u003c/li\u003e\n\u003cli\u003eHamid, S., Waycott, J., Kurnia, S., \u0026amp; Chang, S. (2015). Understanding students\u0026apos; perceptions of the benefits of online social networking use for teaching and learning. \u003cem\u003eThe Internet and Higher Education, 26\u003c/em\u003e, 1-9. https://doi.org/10.1016/j.iheduc.2015.02.004\u003c/li\u003e\n\u003cli\u003eHe, J., Hao, W., \u0026amp; Kim, J.-W. (2019). The effects of instant feedback system on course interest and academic achievement in gamification learning. \u003cem\u003eEducational Innovations and Applications,\u003c/em\u003e 467\u0026ndash;470. https://doi.org/10.35745/ecei2019v2.119\u003c/li\u003e\n\u003cli\u003eHenderson, W. T., \u0026amp; Wen, S. S. (1976). Effects of immediate positive reinforcement on undergraduates\u0026apos; course achievement. \u003cem\u003ePsychological Reports, 39\u003c/em\u003e(2), 568-570. https://doi.org/10.2466/pr0.1976.39.2.568\u003c/li\u003e\n\u003cli\u003eHunerberg, E. (2019\u003cem\u003e). 3rd Eye Theme Analysis\u003c/em\u003e [Online Application]. https://3rdeyeinformation.com/home\u003c/li\u003e\n\u003cli\u003eIrwin, C., Ball, L., Desbrow, B., \u0026amp; Leveritt, M. (2012). Students\u0026apos; perceptions of using Facebook as an interactive learning resource at university. \u003cem\u003eAustralasian Journal of Educational Technology, 28\u003c/em\u003e(7), 1221\u0026ndash;1232. https://doi.org/10.14742/ajet.798\u003c/li\u003e\n\u003cli\u003eJohnson, D. W., \u0026amp; Johnson, R. T. (1982). The effects of cooperative and individualistic instruction on handicapped and nonhandicapped students. \u003cem\u003eThe Journal of Social Psychology, 118\u003c/em\u003e(2), 257\u0026ndash;268. https://doi.org/10.1080/00224545.1982.9922805\u003c/li\u003e\n\u003cli\u003eKaysi, F. (2021). Mobile instant messaging application habits among university students. \u003cem\u003eInteractive Learning Environments,\u003c/em\u003e 1\u0026ndash;19. https://doi.org/10.1080/10494820.2021.1922460\u003c/li\u003e\n\u003cli\u003eKluger, A. N., \u0026amp; DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis, and a preliminary feedback intervention theory. \u003cem\u003ePsychological Bulletin, 119\u003c/em\u003e(2), 254\u0026ndash;284. https://doi.org/10.1037/0033-2909.119.2.254\u003c/li\u003e\n\u003cli\u003eKusumawati, A. J. (2020). Redesigning face-to-face into online learning for speaking competence during covid-19: ESP for higher education in Indonesia. \u003cem\u003eInternational Journal of Language Education, 4\u003c/em\u003e(2), 276\u0026ndash;288. https://doi.org/10.26858/ijole.v4i2.14745\u003c/li\u003e\n\u003cli\u003eLi, S., Zhu, Y., \u0026amp; Ellis, R. (2016). The effects of the timing of corrective feedback on the acquisition of a new linguistic structure. \u003cem\u003eModern Language Journal, 100\u003c/em\u003e(1), 276\u0026ndash;295. https://doi.org/10.1111/modl.12315\u003c/li\u003e\n\u003cli\u003eLiu, W. (2021). Does teacher immediacy affect students? A systematic review of the association between teacher verbal and non-verbal immediacy and student motivation.\u003cem\u003e Frontiers in Psychology, 12\u003c/em\u003e, 713978. https://doi.org/10.3389/fpsyg.2021.713978\u003c/li\u003e\n\u003cli\u003eMah, E., Yu, J., Deck, M., Lyster, K., Kawchuk, J., Turnquist, A., \u0026amp; Thoma, B. (2021). Immersive video-modeling versus traditional video-modeling for teaching central venous catheter insertion to medical residents. \u003cem\u003eCureus, 13\u003c/em\u003e(3) https://doi.org/10.7759/cureus.13661\u003c/li\u003e\n\u003cli\u003eMahdiuon, R., Salimi, G., \u0026amp; Raeisy, L. (2019). Effect of social media on academic engagement and performance: Perspective of graduate students. \u003cem\u003eEducation and Information Technologies,\u003c/em\u003e 2427\u0026ndash;2446. https://doi.org/10.1007/s10639-019-10032-2\u003c/li\u003e\n\u003cli\u003eManca, S. (2020). Snapping, pinning, liking or texting: Investigating social media in higher education beyond Facebook. \u003cem\u003eThe Internet and Higher Education, 44\u003c/em\u003e, 100707. https://doi.org/10.1016/j.iheduc.2019.100707\u003c/li\u003e\n\u003cli\u003eMansholt, H., G\u0026ouml;tz, N. A., \u0026amp; Babitsch, B. (2021). Chances and barriers of online problem-based learning (ePBL) for advanced training in the healthcare sector. \u003cem\u003eStudies in Health Technology and Informatics, 281\u003c/em\u003e, 822\u0026ndash;823. https://doi.org/10.3233/SHTI210292\u003c/li\u003e\n\u003cli\u003eMart\u0026iacute;nez-Cardama, S., \u0026amp; Caridad-Sebasti\u0026aacute;n, M. (2019). Social media and new visual literacies: Proposal based on an innovative teaching project. \u003cem\u003eEducation for Information, 35\u003c/em\u003e(3), 337\u0026ndash;352. https://doi.org/10.3233/EFI-180214\u003c/li\u003e\n\u003cli\u003eMeans, B., Toyama, Y., Murphy, R., \u0026amp; Baki, M. (2013). The effectiveness of online and blended learning: A meta-analysis of the empirical literature. \u003cem\u003eTeachers College Record, 115\u003c/em\u003e(3). https://doi.org/10.1177/016146811311500307\u003c/li\u003e\n\u003cli\u003eNikopoulos, C., Luiselli, J. K., \u0026amp; Fischer, A. J. (2016). Video-modeling. In J. K. Luiselli, \u0026amp; Aaron J. Fischer (Eds.), \u003cem\u003eComputer-Assisted and Web-Based Innovations in Psychology, Special Education, and Health \u003c/em\u003e(pp. \u003cem\u003e1\u003c/em\u003e87\u0026ndash;210). Academic Press. https://doi.org/10.1016/B978-0-12-802075-3.00008-5\u003c/li\u003e\n\u003cli\u003eMeşe, E. \u0026amp; Sevilen, \u0026Ccedil;. (2021). Factors influencing EFL students\u0026apos; motivation in online learning: A qualitative case study. \u003cem\u003eJournal of Educational Technology and Online Learning, 4\u003c/em\u003e(1), 11-22. Retrieved from https://dergipark.org.tr/en/pub/jetol/issue/60134/817680\u003c/li\u003e\n\u003cli\u003ePopkin, J., \u0026amp; Skinner, C. H. (2003). Enhancing academic performance in a classroom serving students with serious emotional disturbance: Interdependent group contingencies with randomly selected components. \u003cem\u003eSchool Psychology Review, 32\u003c/em\u003e(2), 282\u0026ndash;295. https://doi.org/10.1080/02796015.2003.12086199\u003c/li\u003e\n\u003cli\u003eQuinn, M., Narozanick, T., Miltenberger, R., Greenberg, L., \u0026amp; Schenk, M. (2020). Evaluating video-modeling and video-modeling with video feedback to enhance the performance of competitive dancers. \u003cem\u003eBehavioral Interventions, 35\u003c/em\u003e(1), 76\u0026ndash;83. https://doi.org/10.1002/bin.1691\u003c/li\u003e\n\u003cli\u003eRega, A., Somma, F., \u0026amp; Iovino, L. (2020). Development of emotional skills through video-modeling: A case study with a non-verbal participant. \u003cem\u003eTechnology and Disability, 32\u003c/em\u003e(3), 211\u0026ndash;218. https://doi.org/10.3233/TAD-180222\u003c/li\u003e\n\u003cli\u003eReynolds B. (2006). A review of delay-discounting research with humans: relations to drug use and gambling. \u003cem\u003eBehavioural Pharmacology, 17\u003c/em\u003e(8), 651\u0026ndash;667. https://doi.org/10.1097/FBP.0b013e3280115f99\u003c/li\u003e\n\u003cli\u003eRios, D., Kazemi, E., \u0026amp; Peterson, S. M. (2018). Best practices and considerations for effective service provision via remote technology. \u003cem\u003eBehavior Analysis: Research and Practice, 18\u003c/em\u003e(3), 277\u0026ndash;287. https://doi.org/10.1037/bar0000072\u003c/li\u003e\n\u003cli\u003eRoblyer, M. D., McDaniel, M., Webb, M., Herman, J., \u0026amp; Witty, J. V. (2010). Findings on Facebook in higher education: A comparison of college faculty and student uses and perceptions of social networking sites. \u003cem\u003eInternet and Higher Education, 13\u003c/em\u003e(3), 134\u0026ndash;140. https://doi.org/10.1016/j.iheduc.2010.03.002\u003c/li\u003e\n\u003cli\u003eRoseth, C. J., Johnson, D. W., \u0026amp; Johnson, R. T. (2008). Promoting early adolescents\u0026apos; achievement and peer relationships: The effects of cooperative, competitive, and individualistic goal structures. \u003cem\u003ePsychological Bulletin, 134\u003c/em\u003e(2), 223\u0026ndash;246. https://doi.org/10.1037/0033-2909.134.2.223\u003c/li\u003e\n\u003cli\u003eS\u0026aacute;nchez-Gelabert, A., \u0026amp; Elias, M. (2023). Going online? Does transferring to online university increase the likelihood of graduation among students from lower social background? \u003cem\u003eInternational Journal of Educational Technology in Higher Education, 20\u003c/em\u003e, 39. https://doi.org/10.1186/s41239-023-00407-4\u003c/li\u003e\n\u003cli\u003eSpeidel, G. E., Tharp, R. G., \u0026amp; Speidel, G. E. (1981). What does self-reinforcement reinforce? An empirical analysis of the contingencies in self-determined reinforcement. \u003cem\u003eChild Behavior Therapy, 2\u003c/em\u003e(4), 1\u0026ndash;22. https://doi.org/10.1300/J473V02N04_01\u003c/li\u003e\n\u003cli\u003eStatista (2024, April). \u003cem\u003eDistribution of Facebook users worldwide as of April 2024, by age and gender.\u003c/em\u003e https://www.statista.com/statistics/376128/facebook-global-user-age-distribution/\u003c/li\u003e\n\u003cli\u003eSwartzwelder, K., Murphy, J., \u0026amp; Murphy, G. (2019). The impact of text-based and video discussions on student engagement and interactivity in an online course. \u003cem\u003eJournal of Educators Online, 16\u003c/em\u003e(1). https://doi.org/10.9743/jeo.2019.16.1.13\u003c/li\u003e\n\u003cli\u003eTarifa-Rodriguez, A., Virues-Ortega, J., Perez-Bustamante Pereira, A., Calero-Elvira, A., \u0026amp; Cowie, S. (2023). Quantitative indices of student social media engagement in tertiary education: a systematic review and a taxonomy. \u003cem\u003eJournal of Behavioral Education\u003c/em\u003e. https://doi.org/10.1007/s10864-023-09516-6\u003c/li\u003e\n\u003cli\u003eTarifa-Rodriguez, A., Virues-Ortega, J., \u0026amp; Calero-Elvira, A. (2024). A behavioral education in social media program for postgraduate academic achievement: a randomized controlled trial. \u003cem\u003eJournal of Behavioral Education\u003c/em\u003e. https://doi.org/10.1007/s10864-024-09545-9\u003c/li\u003e\n\u003cli\u003eTran, P. (2016). Training learners to use quizlet vocabulary activities on mobile phones in Vietnam with Facebook. \u003cem\u003eJALT CALL Journal, 12\u003c/em\u003e(1), 43\u0026ndash;56. https://doi.org/10.29140/jaltcall.v12n1.201\u003c/li\u003e\n\u003cli\u003eVan Iddekinge, C. H., Aguinis, H., Mackey, J. D., \u0026amp; DeOrtentiis, P. S. (2017). A meta-analysis of the interactive, additive, and relative effects of cognitive ability and motivation on performance. \u003cem\u003eJournal of Management, 20\u003c/em\u003e(10), 1-31. https://doi.org/10.1177/0149206317702220\u003c/li\u003e\n\u003cli\u003eVan Ryzin, M. J., Roseth, C. J., \u0026amp; Biglan, A. (2020). Mediators of effects of cooperative learning on prosocial behavior in middle school. \u003cem\u003eInternational Journal of Applied Positive Psychology, 5\u003c/em\u003e(1\u0026ndash;2), 37\u0026ndash;52. https://doi.org/10.1007/s41042-020-00026-8\u003c/li\u003e\n\u003cli\u003eVirues-Ortega, J., Delgado Casas, C., Martin, N., Tarifa-Rodriguez, A., Reina Hidalgo, A. J., \u0026amp; Cox, A. D., \u0026amp; Navarro Guzm\u0026aacute;n, J. I. (2023) Interobserver agreement in software-aided and paper-and-pencil behavioral observation. \u003cem\u003eBehavior Research Methods, 55, \u003c/em\u003e855\u0026ndash;866.\u003cem\u003e \u003c/em\u003ehttps://doi.org/10.3758/s13428-022-01861-0\u003c/li\u003e\n\u003cli\u003eWang, W. T., \u0026amp; Lin, Y. L. (2021). Evaluating factors influencing knowledge-sharing behavior of students in online problem-based learning. \u003cem\u003eFrontiers in Psychology, 12,\u003c/em\u003e 691755. https://doi.org/10.3389/fpsyg.2021.691755\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-behavioral-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jobe","sideBox":"Learn more about [Journal of Behavioral Education](http://link.springer.com/journal/10864)","snPcode":"10864","submissionUrl":"https://submission.springernature.com/new-submission/10864/3","title":"Journal of Behavioral Education","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Facebook, cooperative learning, higher education, instructor feedback, self-management, video-modeling","lastPublishedDoi":"10.21203/rs.3.rs-7932578/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7932578/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eFew randomized controlled trials have analyzed evidence-based educational practices delivered through a social media environment. This study used a multi-arm randomized controlled trial to evaluate the critical components of an educational intervention package: study self-management skills training delivered through video modeling, cooperative learning, and semi-immediate feedback. We evaluated social media engagement and academic performance among 141 students in a graduate-level applied psychology program. Students were randomly assigned to five groups: control (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27); self-management (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27); cooperative learning (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;33); self-management and cooperative learning (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27); and self-management, cooperative learning, and semi-immediate instructor feedback (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;27). Results indicated that participants receiving the complete intervention package showed the highest levels of engagement and academic performance. The analysis revealed that instructor semi-immediate feedback was critical to the package's effectiveness, whereas the impact of the self-management skills training could not be verified. We discuss the conceptual, methodological, and practical implications of the study.\u003c/p\u003e","manuscriptTitle":"Enhancing Academic Performance in Tertiary Education through Social Media: A Multi-Arm Randomized Controlled Trial","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-17 06:34:40","doi":"10.21203/rs.3.rs-7932578/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-19T19:54:52+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-16T03:25:28+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-15T15:26:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"319410127835780330303706809715618605168","date":"2025-12-29T17:31:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"155768217921576573457990971394225602191","date":"2025-12-12T17:12:58+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-05T16:48:00+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-23T14:24:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-23T14:23:46+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Behavioral Education","date":"2025-10-23T13:11:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-behavioral-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jobe","sideBox":"Learn more about [Journal of Behavioral Education](http://link.springer.com/journal/10864)","snPcode":"10864","submissionUrl":"https://submission.springernature.com/new-submission/10864/3","title":"Journal of Behavioral Education","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"15272c17-2af7-4b66-b059-15408c3c495b","owner":[],"postedDate":"November 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T16:11:27+00:00","versionOfRecord":{"articleIdentity":"rs-7932578","link":"https://doi.org/10.1007/s10864-026-09629-8","journal":{"identity":"journal-of-behavioral-education","isVorOnly":false,"title":"Journal of Behavioral Education"},"publishedOn":"2026-04-21 15:59:43","publishedOnDateReadable":"April 21st, 2026"},"versionCreatedAt":"2025-11-17 06:34:40","video":"","vorDoi":"10.1007/s10864-026-09629-8","vorDoiUrl":"https://doi.org/10.1007/s10864-026-09629-8","workflowStages":[]},"version":"v1","identity":"rs-7932578","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7932578","identity":"rs-7932578","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}