Author | Method | Sample | Dimension | Selected findings |
---|---|---|---|---|
Alavi (1994) | Empirical | 79 treatment, 48 control students (MBA) | Format, technology | Computer-mediated collaborative learning increases skill development, perceived and actual learning, and satisfaction. |
Arbaugh (2000a) | Empirical | 97 MBA students | Learner, instructor, format, technology | Instructor efforts to create an interactive environment (i.e., interaction ease and emphasis, classroom dynamics) predict perceived internet-based learning. Technology features (i.e., ease of use, usefulness), student characteristics (i.e., gender, age, prior experience, time online), and flexibility (i.e., course flexibility, program flexibility) are not significant. |
Arbaugh (2000b) | Empirical | 111 MBA students in 5 courses | Format, technology | Technology flexibility and an interactive environment are more important for learner satisfaction than the ease or frequency with which the medium can be used. |
Arbaugh (2000c) | Empirical | 27 treatment, 33 control students (MBA) | Format, technology | Internet-based courses do not diminish learning and lead to increased female participation in class discussions. |
Arbaugh (2008) | Empirical | 656 students in 55 MBA courses | Learner, instructor, format, technology | The Community of Inquiry framework (i.e., social, teaching, and cognitive presence) predicts perceived online learning and satisfaction with the delivery medium. Gender affects perceived learning. Gender, semester, and number of prior online courses predict satisfaction. |
Arbaugh (2014) | Review | n/a | Format, technology | Learner control and group collaboration enhance learning in blended environments. |
Arbaugh and Benbunan-Fich (2006) | Empirical | 579 MBA students in 40 course sections | Instructor, format | Collaborative online learning results in higher perceived learning and satisfaction than individual online learning, independent of the teaching approach. Group learning is positively moderated by objectivist teaching (i.e., knowledge transmission), while individual learning is positively moderated by constructivist teaching. |
Arbaugh and Duray (2002) | Empirical | 120 MBA students | Format, technology | Perceived web-based learning and satisfaction are positively affected by flexibility and negatively affected by class size. Prior online learning experience influences satisfaction. |
Arbaugh et al. (2009) | Review | n/a | Learner, instructor, format, technology | Online courses are at least comparable to classroom courses with respect to learning outcomes. Antecedents of learning effectiveness differ across business disciplines. |
Arbaugh and Rau (2007) | Empirical | 575 MBA students in 40 course sections | Format, technology | Different management disciplines cease to be significant predictors of perceived online learning when accounting for structural (i.e., class size, media variety, exams, projects) and behavioral (i.e., interaction with peers, instructor, interface) characteristics. However, differences among disciplines remain significant predictors of satisfaction. Some characteristics predict satisfaction and outcomes in opposite directions (e.g., media variety, peer interaction). |
Asarta and Schmidt (2013) | Empirical | 179 students in 3 course sections | Learner, technology | Timing and regularity of online access predict student performance, while number and length of access do not. |
Asarta and Schmidt (2017) | Empirical | 347 students across 4 treatment groups, 257 students across 3 control groups | Learner | Previously weak students perform better in traditional environments, while previously strong students perform better in blended environments. The environment does not matter for average students. |
Beege, Schneider, Nebel, and Rey (2017) | Empirical | 88 mostly undergraduate students across 4 treatment groups | Instructor, technology | Educational videos with a frontal (rather than lateral) instructor orientation positively influence retention, as para-social interaction may trigger beneficial affective states and deeper cognitive processing. Instructor proximity does not affect learning. |
Buttner and Black (2014) | Empirical | 82 treatment, 64 control students | Learner, format, technology | Implementation of an online learning system improves test results. Neither an additional test nor more time invested moderate outcomes. |
Concannon et al. (2005) | Empirical | 600 undergraduate students | Learner, instructor, format | Preferred educational resources, attitudes toward computers, study patterns, and career plans affect e-learner satisfaction. However, the main antecedents are peer interaction and instructor support. |
D’Mello et al. (2014) | Empirical | Study 1: 63 undergraduate students; Study 2: 76 undergraduate students | Learner, format | Deliberate confusion positively affects actual learning. Prior knowledge shows small moderation effects. |
Daspit and D’Souza (2012) | Empirical | 203 undergraduate students | Format, instructor | In a wiki environment, teaching and social presence affect cognitive presence, which confirms that the instructor retains an important role in technology-mediated settings. |
Demetriadis, Papadopoulos, Stamelos, and Fischer (2008) | Empirical | Study 1/Study 2: 8 treatment, 8 control undergraduate students | Learner, instructor, format | Scaffolding (e.g., via appropriate questioning) positively influences knowledge acquisition and transfer in a technology-enhanced environment. Learners with critical thinking skills benefit the most from scaffolding. |
Deschacht and Goeman (2015) | Empirical | 1883 undergraduate students | Format | Blended environments lead to increased dropout rates and better exam performance. |
Dindar and Akbulut (2016) | Empirical | 572 undergraduate students across 7 treatment conditions | Learner | Concurrent multitasking and daily media exposure negatively affect retention. Concurrent multitasking impedes topic interest. Sequential multitasking, digital device experience, and daily multitasking habits are not related to retention. |
Eid and Al-Jabri (2016) | Empirical | 203 undergraduate and 105 graduate students | Format, technology | Chatting, online discussions, and file sharing predict knowledge sharing, which in turn predicts perceived learning. Enjoyment and entertainment also predict learning. |
Eom and Ashill (2018) | Empirical | 305 undergraduate and 67 graduate students | Learner, instructor, format | Six interdependent factors (i.e., course-design quality, instructor, student motivation, student-student dialog, student-instructor dialog, and self-regulated learning) explain perceived e-learning, which predicts satisfaction. |
Eom et al. (2006) | Empirical | 397 graduate and undergraduate students | Learner, instructor, format | While course structure, instructor feedback, self-motivation, learning style, interaction, and instructor facilitation affect satisfaction, only instructor feedback and learning style directly predict perceived e-learning. Satisfaction also predicts perceived e-learning. |
Evans (2008) | Empirical | 196 undergraduate students | Technology | For review, podcasts are superior to textbooks or student notes in terms of time required and perceived learning. |
Fritz (2011) | Empirical | Students in 131 courses | Learner, technology | Students who are more active in the learning management system earn higher grades. |
Fryer and Bovee (2016) | Empirical | 975 undergraduate students | Learner, instructor | Instructor support has direct and indirect effects on learner motivation. Effort beliefs predict task value and ability beliefs, which predict e-learning completion. |
Garrison and Kanuka (2004) | Review | n/a | Format | Communities of Inquiry (i.e., cognitive, social, and teaching presence) are relevant for both face-to-face and online settings. Blended environments can enhance meaningful learning. |
Grabe and Christopherson (2008) | Empirical | 329 undergraduate students | Learner, technology | The use of online resources and class attendance is positively related to exam performance. Online resources may compensate for a lack of class attendance. |
Guo, Kim, and Rubin (2014) | Empirical | 6.9 million sessions across 4 edX courses, interviews with 6 edX staff | Instructor, technology | Shorter videos are more engaging. Videos with a personal feel can be more engaging than high-quality studio recordings. Informal videos in which the speaker is visible are more engaging than slides alone. Instructors who speak faster and with enthusiasm are more engaging. |
Hazari, CO’M, and Rutledge (2013) | Empirical | 102 undergraduate students | Format, technology | Blogs can improve outcomes by fostering deeper learning and engagement in an interactive environment. Peer interaction can be used as part of constructive feedback and self-evaluation. |
Huang (2014) | Empirical | 389 undergraduate students | Learner, technology | Perceived usefulness and playfulness are related to mobile learning satisfaction, which in turn predicts the intent to continue. Resistance to change has a minor influence on satisfaction. Self-management moderates the relationships between perceived usefulness, playfulness, and resistance to change and satisfaction as well as the relationship between satisfaction and the intent to continue. |
Hwang and Arbaugh (2006) | Empirical | 196 undergraduate students | Learner, technology | Discussion board feedback-seeking behaviors are related to actual learning if triggered by a competitive attitude (i.e., preventing others from getting ahead of oneself or personal diligence to get ahead of others). Traditional feedback-seeking measures of asking the instructor or peers do not have a positive effect on learning performance. |
Kember, McNaught, Chong, Lam, and Cheng (2010) | Empirical | 595 students | Technology | Features that promote constructive dialogue and interactive activities encourage deeper learning and enhance understanding of contents. |
Kizilcec, Bailenson, and Gomez (2015) | Empirical | Study 1: 2951 participants; Study 2: 12,468 participants | Instructor, technology | Videos in which the instructor can be seen need to balance the increased extraneous load with gains from social and other nonverbal cues. When the instructor is visible, the cognitive load and perceived social presence increase, but learning outcomes and attrition remain constant. There is no “one-size-fits-all” approach. |
Knoerzer et al. (2016) | Empirical | 75 students across 3 treatment groups | Learner, format | Negative emotions positively affect online learning, perhaps due to more detailed information processing. Positive emotions negatively affect learning, perhaps because they distract from the material. Emotions do not influence motivation. |
Kreijns, Kirschner, and Vermeulen (2013) | Conceptual | n/a | Format | Sociability, social space, and social presence determine social interaction, which predicts learning. |
Krentler and Willis-Flurry (2005) | Empirical | 549 undergraduate students | Learner, technology | Online discussion boards enhance student learning. The relationship between technology usage and learning is moderated by student major (i.e., marketing and computer information systems) and total amount of internet use (i.e., university and private use). |
Lancellotti et al. (2016) | Empirical | 247 treatment, 232 control undergraduate students | Technology | Watching a set of short, concept-focused videos improves exam scores. Gender and ethnicity do not moderate this effect. |
Liu (2012) | Empirical | 11,351 undergraduate and graduate students | Format | Motivation for taking a course, students’ class status, and instructors’ academic rank have significant impacts on distance learning. |
López-Pérez et al. (2011) | Empirical | 985 students in 17 groups | Learner, format | Blended environments reduce dropout rates and improve exam performance. Learning depends on motivation, age, prior experience, and class attendance for both face-to-face and online elements. Gender, perceived utility, and satisfaction do not predict learning. |
Macfadyen and Dawson (2010) | Empirical | 118 undergraduate students in 5 classes | Learner, technology | 15 variables tracked by the learning management system predict actual learning. They correctly predict 81% of failing students. Key variables, such as number of contributions, mails sent, and completed assessments, explain more than 30% of the variance in final grades. |
Markel (1999) | Review | n/a | Format, instructor | The literature offers negative descriptions of teacher-centered lectures, which can scare away potential teachers. The false dichotomy between boring lecturers and exciting distance educators inaccurately suggests that the technology, not the teacher, makes a good course. |
Mayer (2002) | Review | n/a | Format | Nine instructional design principles affect cognitive processing: multimedia, spatial contiguity, temporal contiguity, coherence, modality, redundancy, pretraining, signaling, and personalization. |
Mayer, Dow, and Mayer (2003) | Review | n/a | Format | Four methods foster cognitive processing and, thereby, learning across media. The multimedia effect combines words and pictures, the coherence effect excludes extraneous material, the spatial contiguity effect places text next to corresponding pictures, and the personalization effect applies a less formal presentation style. |
Mayer and Chandler (2001) | Empirical | Study 1: 30 undergraduate students; Study 2: 29 undergraduate students (2 treatment groups in both studies) | Format | Presenting information in separate parts allows learners to build multiple mental representations that can be integrated when watching the parts or the entire presentation again (i.e., partial revision). Learner control over pace leads to skipping of sections (i.e., learners end up with shorter parts), which benefits cognitive processing. |
Mayer et al. (2003) | Empirical | Study 1: 52 students across 2 treatment groups; Study 2: 78 students across 4 treatment groups; Study 3: 54 students across 2 treatment groups; Study 4: 39 students across 2 treatment groups | Format | Students learn better if animations are complemented with spoken language rather than printed text (i.e., modality principle), if they are able to control the pace and order of the presentation (i.e., interactivity principle), and if they answer conceptual questions while learning (i.e., self-explanation principle). Complementing narrated text with the instructor’s image does not enhance actual learning due to the additional extraneous load (i.e., presence principle). |
Mayer and Moreno (2003) | Conceptual | n/a | Format | Cognitive load is central to multimedia design. Strategies such as off-loading, segmenting, pretraining, weeding (i.e., cutting into parts), signaling, aligning, eliminating redundancy, synchronizing, and individualizing diminish extraneous load and free up capacity for germane load. |
McGill and Klobas (2009) | Empirical | 267 students | Technology | Task-technology fit (TTF) directly and indirectly predicts perceived learning through attitude toward technology use and actual technology use. The direct effect on actual learning is marginal. TTF also predicts expected consequences of technology use but these are not related to actual technology use. Instructor norms predict technology use. Perceived learning does not predict actual learning. |
McLaren (2004) | Empirical | 208 undergraduate students in 5 courses, 2 treatment types each | Format | While online delivery increases dropout rates, actual learning is independent of the format of instruction. |
Moreno (2006) | Review | n/a | Format, technology | The modality principle (i.e., combination of visual and audio) moderates learning across media. A method that has learning benefits in a lower technology environment also supports learning with higher technologies. The latter does not have additional learning benefits. A “media-enables-method” hypothesis is derived (as opposed to “method-affects-learning” and “media-affects-learning”). |
Moreno and Mayer (2007) | Review | n/a | Learner, format | Cognitive learning theories should account for learner motivation, metacognition, and prior knowledge. Design principles for interactive multimedia environments include guidance, reflection, feedback, control, and pretraining, as they encourage relevant and/or reduce extraneous cognitive load. |
Nemanich, Banks, and Dusya (2009) | Empirical | 149 undergraduate students across 2 treatment groups | Learner, instructor, format | Perceived instructor expertise, content relevance, and social richness enhance student enjoyment. Perceived confidence in instructor expertise and content relevance also strengthen the understanding of course concepts. Enjoyment is positively associated with learning performance in the classroom, while student ability is positively associated with learning performance online. |
Nihalani, Mayrath, and Robinson (2011) | Empirical | Study 1: 42 students across 2 treatment groups, 24 control students (“novice” undergraduates); Study 2: 42 students across 2 treatment groups, 20 control students (“expert” undergraduates) | Learner, instructor | Learners with little prior knowledge benefit more from individual feedback than from collaboration with other novices. For students with high prior knowledge, individual feedback may inhibit learning and reverse the benefits of expertise. |
O’Flaherty and Phillips (2015) | Review | n/a | Format | There is no “one-size-fits-all” approach to flipped learning, but core features include content in advance, educator awareness of students’ understanding, and higher-order learning during classes. |
O’Neill and Sai (2014) | Empirical | 48 students | Format | Respondents believe they learn better face-to-face. They are aware of the greater risk of failure or dropout in online courses. |
Owston, York, and Murtha (2013) | Empirical | 577 students in 11 courses | Format | High achievers are most satisfied with blended courses, would take one again, and prefer them to fully face-to-face or online courses. They also find blended courses more convenient engaging, and feel they learn key concepts better in blended courses than in traditional face-to-face courses. |
Palocsay and Stevens (2008) | Empirical | 327 undergraduate students across 4 treatment groups | Learner, instructor, technology | Teacher experience and student academic competence predict actual learning. The specific technology used for web-based homework does not affect learning. |
Piccoli et al. (2001) | Empirical | 70 students across 2 treatment groups, 76 students across 2 control groups (all undergraduate) | Learner, format, technology | Actual learning in virtual versus traditional environments is similar. Hence, the increased learner control in the virtual environment does not benefit learning. Satisfaction in the virtual environment is even lower. Only computer self-efficacy is higher. |
Plass, Heidig, Hayward, Homer, and Um (2014) | Empirical | Study 1/Study 2: 121/103 graduate students across 4 treatment conditions | Learner, format | Distinct choices and combinations of instructional design features (e.g., colors, shapes) can induce positive emotions, which predict comprehension and knowledge transfer in multimedia learning. |
Redpath (2012) | Review | n/a | Format, technology | Online delivery provides sufficient interaction, collaboration, and learning outcomes to support a quality business education. |
Scheiter and Gerjets (2007) | Review | n/a | Learner, format, technology | Self-controlled multimedia environments are well suited for improving learning among students with high prior knowledge, better self-regulatory skills, and more positive attitudes. |
Selim (2003) | Empirical | 403 undergraduate students | Technology | Usefulness and ease of use predict acceptance and use of a course website. Ease of use is mainly mediated by usefulness. |
Selim (2007) | Empirical | 538 undergraduate students | Learner, instructor, format, technology | Eight determinants of e-learning satisfaction across four categories: instructor characteristics (attitude toward and control of technology, teaching style), student characteristics (motivation and technical competency, interactive collaboration, course content and design), technology (ease of access, infrastructure), and university support. |
Seufert (2003) | Empirical | 86 students across 2 treatment groups and one control group | Learner, instructor | The effect of help depends on learners’ prior knowledge. In cases of low prior knowledge, help negatively effects comprehension and recall performance. For medium prior knowledge, directive (as opposed to non-directive) help enhances both recall and comprehension due to its summarizing and repeating function. In cases of high prior knowledge, help barely affects learning. |
Sloan and Lewis (2014) | Empirical | 70 undergraduate students in 2 course sections | Learner, technology | Access to lecture-capture videos is associated with higher exam scores, even after controlling for previous exam performance. |
Snowball (2014) | Empirical | 50 undergraduate students | Format, technology | Partially replacing lectures with online activities and resources improves actual learning. More active online resources (e.g., multiple-choice questions) are most beneficial for student performance. Some essentially passive activities (e.g., short online lectures, mini-movies) may be useful for demonstrating how to explain and apply concepts. |
Solimeno, Mebane, Tomai, and Francescato (2008) | Empirical | 82 treatment students, 88 control students (all graduate) | Learner, instructor, technology | Technology improves perceived and actual learning among students with low anxiety, high problem-solving efficacy, and time-management problems. Tutor characteristics do not influence learning. |
Song et al. (2004) | Empirical | 76 graduate students (all participated in a survey, 14 were also interviewed) | Learner, format | Course design, learner motivation, time management, and comfort with the technology affect perceived online learning. Technical problems, a lack of community, time constraints, and difficulty in understanding the course objectives are challenges. |
Sun et al. (2008) | Empirical | 295 students | Learner, instructor, format, technology | E-learner satisfaction is affected by learner computer anxiety, instructor attitude toward e-learning, course flexibility, course quality, perceived usefulness of e-learning, perceived ease of use, and diversity in assessments. Learners’ attitude toward computers, learners’ internet self-efficacy, timeliness of instructors’ responses, technology quality, internet quality, and perceived interaction with others do not predict satisfaction. |
Terpend, Gattiker, and Lowe (2014) | Empirical | 180 undergraduate students in 6 course sections | Technology | Perceived ease-of-use and price predict e-textbook adoption. Perceived usefulness, internet self-efficacy, and environmental concerns are not significant. The grades of e-textbook adopters and hardcopy users do not differ. |
Um, Plass, Hayward, and Homer (2012) | Empirical | 118 undergraduate students across 4 treatment groups | Learner, format | Positive emotional design negatively influences perceived task difficulty and positively affects motivation. It promotes comprehension, transfer, and satisfaction. Emotional design does not increase extraneous cognitive processing. |
Volery and Lord (2000) | Empirical | 47 students | Learner, instructor, format, technology | Technology (ease of access and navigation, interface design, interaction), instructor (attitude toward students, technology control/technical competence, teaching style/interaction), and learners’ prior technology experience predict perceived learning. Internet access at home, study program, country of origin, and gender are not significant. |
Walker, Curren, Kiesler, Lammers, and Goldenson (2013) | Empirical | 516 students | Format, technology | Peer networking via discussion boards leads to better performance. Reading discussions and posting improve final grades. |
Webster and Hackley (1997) | Empirical | 247 mainly graduate distance-learning students | Learner, instructor, format, technology | Medium richness relates to all perceived outcome variables. Other important antecedents include technology reliability, technology quality, instructors’ attitudes, teaching style, instructors’ control over the technology, number of student locations, students’ comfort with their images on screen, and classmates’ attitudes. |
Woo (2014) | Empirical | 63 undergraduate students | Learner | Motivation and cognitive processing predict actual online learning. |
Wu et al. (2010) | Empirical | 212 e-learning participants | Learner, format, technology | Computer self-efficacy, system functionality, content features, and interaction affect performance expectations. Interaction also affects the learning climate. Performance expectations and the learning climate affect satisfaction with blended environments. |
Xu and Jaggars (2014) | Empirical | More than 40,000 students | Learner, technology | All types of students suffer in online courses. Those exhibiting the most decline are males, younger students, black students, and students with lower GPAs. Controlling for individual and peer effects as well as disciplines shows the widest performance gaps. |
Yourstone, Kraye, and Albaum (2008) | Empirical | 190 undergraduate students in 4 course sections across 2 treatment types | Technology | Immediate feedback technologies, such as clickers, have a positive impact on actual learning. |
Zacharis (2015) | Empirical | 134 undergraduate students | Format, technology | Four out of 29 system measures predict 52% of the variance in final grades: reading and posting messages, content creation, quiz efforts, and number of files viewed. |