Self-Regulated Learning
in Online Education:
A Review of the Empirical Literature

Introduction

Distance education is hardly a new phenomenon in the United States. Since the development of the postal service in the 19^th century, correspondence courses have provided distance education to students across the country (Phipps & Merisotis, 1999). This trend continued well into the 20^th century with the advent of television and radio—media technologies that allowed for expanded opportunities to learn at a distance (Simonson, Smaldino, Albright, & Zvacek, 2003). Today, computer-mediated communications and the Internet have resulted in a “rapid and explosive development of interest in and discussion about distance education” (Moore, 2003a, p. xiii). Even prestigious universities who once shunned distance education are now making substantial investments in distance learning technologies (Larreamendy-Joerns & Leinhardt, 2006; Moore, 2003b). Concurrently, business and military organizations are attracted to the potential for computer-mediated distance learning to provide “anytime, anywhere” access to education and training, thereby greatly reducing training costs and increasing accessibility to training materials (Fletcher, Tobias, & Wisher, 2007).

Distance Education Defined

So what is distance education? Although a single unifying definition is difficult to locate in the literature, Moore and Kearsley (2005) have provided a comprehensive description of this unique educational phenomenon. In their view, “distance education is planned learning that normally occurs in a different place from teaching, requiring special course design and instruction techniques, communication through various technologies, and special organizational and administrative arrangements” (Moore & Kearsley, 2005, p. 2). This broad definition encompasses many different learning and teaching formats, including paper-based correspondence courses, audio and video conferencing, and computer-mediated instruction. Although these formats are distinct from one another, geographical separation of teacher and student tends to be the defining characteristic.

Online Distance Education

In the last decade, distance education has changed significantly with the advent of computer-mediated learning, two-way interactive video, online or Web-based learning, and a host of other learning technologies (Simonson et al., 2003). Today there is little doubt that the Internet has become the technology-of-choice for learning and teaching at a distance (Dabbagh & Bannan-Ritland, 2005). Much of this popularity stems from the fact that the Internet is an inherently flexible technology that can be applied in a variety of ways and in a plethora of educational contexts—from simple course administration and student management to teaching entire degree programs online (Wisher & Olson, 2003). Furthermore, the recent expansion of widespread broadband access has brought the Internet into millions of homes, schools, and businesses, thereby providing students and teachers with the opportunity to exploit the Internet’s innate flexibility as a learning and teaching tool (Moore & Kearsley, 2005).

Almost without exception, institutions have recognized the Internet’s value as an educational tool and are developing online distance learning programs. For example, a recent survey of 2,200 U.S. colleges and universities by the Sloan Consortium (2006) found that 96% of large institutions (greater than 15,000 total enrollments) have some online offerings; 62% of Chief Academic Officers rated learning outcomes in online education as the same or superior to traditional, face-to-face instruction; 58% of schools identified online education as a critical long-term strategy; and overall online enrollment increased from 2.4 million in 2004 to 3.2 million in 2005.

Likewise, the U.S. military has recognized the utility of online education. In 1999 the Office of the Under Secretary of Defense created a collaborative effort between the public and private sectors to develop the standards, tools, and learning content necessary to harness the power of information technologies to modernize military training (Advanced Distributed Learning, n.d.). Known as the Advanced Distributed Learning (ADL) initiative, this effort was designed to make education and training available to the military’s more than three million personnel anytime, anywhere (Curda & Curda, 2003). Not surprisingly, online instruction is considered a critical component of the ADL initiative (Fletcher et al., 2007).

Research on Distance Education

Traditionally, research in the area of online education, specifically, and distance education, more generally, has focused on group comparisons; that is, online/distance learners versus traditional classroom students (Berge & Mrozowski, 2001; Bernard et al., 2004b; Phipps & Merisotis, 1999; Russell, 1999; Sitzmann, Kraiger, Stewart, & Wisher, 2006). With few exceptions, results from these studies suggest that, “the learning outcomes of students using technology at a distance are similar to the learning outcomes of students who participate in conventional classroom instruction” (Phipps & Merisotis, 1999, p. 1). Additionally, the attitudes and satisfaction of distance learners have been generally characterized as positive (Dabbagh & Bannan-Ritland, 2005; Hara & Kling, 1999).

Recently, however, several authors (Abrami & Bernard, 2006; Bernard, Abrami, Lou, & Borokhovski, 2004a; Bernard et al., 2004b; Dillon & Greene, 2003; Gibson, 2003; Perraton, 2000; Phipps & Merisotis, 1999; Saba, 2000) have identified major deficiencies in past research on distance learning. Along with a surplus of methodological problems, which have long plagued the empirical literature, two important issues have been identified. First, a large proportion of the distance education research has emphasized comparisons of achievement outcomes between groups of distance and traditional learners, at the expense of any consideration for within group variation in achievement and satisfaction among distance learners. Second, much of the research has lacked a theoretical or conceptual framework. In response to these problems, experts in the field of distance education (Abrami & Bernard, 2006; Bernard et al., 2004a, 2004b; Perraton, 2000; Phipps & Merisotis, 1999; Saba, 2000) have challenged researchers to (1) focus future studies on within group differences among distance learners; specifically, those attributes—motivational, cognitive, and otherwise—that contribute to success in distance learning environments; and (2) conduct research that is grounded in learning theory and which builds on the work of others.

Self-Regulated Learning

As online education has grown, so too has interest in academic self-regulation (Schunk & Zimmerman, 1998). Academic self-regulation, also known as self-regulated learning (SRL), has been defined as, “an active, constructive process whereby learners set goals for their learning and then attempt to monitor, regulate, and control their cognition, motivation, and behavior, guided and constrained by their goals and the contextual features of the environment” (Pintrich, 2000, p. 453). Self-regulated learners are generally characterized as active participants who efficiently control their own learning experiences in many different ways, including establishing a productive work environment and using resources effectively; organizing and rehearsing information to be learned; maintaining positive emotions during academic tasks; and holding positive motivational beliefs about their capabilities, the value of learning, and the factors that influence learning (Schunk & Zimmerman, 1994, 1998). Moreover, self-regulation is not an all-or-nothing phenomenon. Instead, students are self-regulating to the extent that they are cognitively, motivationally, and behaviorally involved in their own learning activities (Zimmerman, 2000b).

Recently, several scholars have suggested that SRL skills may be particularly important for students participating in online education (Dabbagh & Kitsantas, 2004; Garrison, 2003; Hartley & Bendixen, 2001; Schunk & Zimmerman, 1998). For example, Dabbagh and Kitsantas (2004) have argued that, “in a Web-based learning environment, students must exercise a high degree of self-regulatory competence to accomplish their learning goals, whereas in traditional face-to-face classroom settings, the instructor exercises significant control over the learning process and is able to monitor student attention and progress closely” (p. 40.). Likewise, in one of earliest discussions of self-regulation and its applicability to open learning environments, Kinzie (1990) identified self-regulatory skills as one of three critical requirements for student success. She concluded, in part, that the effective use of SRL strategies is essential in flexible learning situations due to the high degree of student autonomy resulting from the instructor’s physical absence.

In general, investigators who study academic self-regulation are attempting to understand how students become masters of their own learning processes (Schunk & Zimmerman, 1994, 1998). Over the last three decades, scholars interested in academic self-regulation within traditional classrooms have consistently found moderate to strong positive relations between students’ motivational engagement, their use of SRL strategies, and, ultimately, their academic achievement (Pintrich, 1999; Pintrich & De Groot, 1990; Pintrich & Garcia, 1991). For example, in one of the earliest studies to employ a SRL perspective, Pintrich and De Groot (1990) surveyed 173 seventh graders and found that higher levels of task value (i.e., the extent to which students find a task interesting, important, and/or valuable; Eccles & Wigfield, 1995) and self-efficacy (students’ confidence in their ability to complete specific learning tasks; Bandura, 1997) were related to students’ use of learning strategies. Furthermore, the researchers found that task value, self-efficacy, and learning strategies use were all correlated with higher levels of achievement, as measured by final course grades, essays and reports, and in-class seatwork.

Although there are various conceptualizations of academic self-regulation (for a review, see Boekaerts, Pintrich, & Zeidner, 2000), several researchers have found social cognitive models of SRL to be particularly useful in analyzing student success in online education (for a review, see Militiadou & Savenye, 2003). Social cognitive models highlight important motivational factors, such as students’ self-efficacy beliefs and goal orientation, as well as learning strategies that appear to benefit students in these highly independent learning situations. Furthermore, a number of investigators have recently emphasized the importance of social and environmental factors on student success in online education (e.g., Gunawardena & Zittle, 1997; Richardson & Swan, 2003). Consequently, a social cognitive perspective on self-regulation, which addresses the interrelationship between the learner, the learner’s behavior, and the social environment (Bandura, 1986, 1997), appears to lend itself well to an understanding of how successful learners function in online situations. 

Review of the Literature

The studies examined in this review were located by searching the publicly available literature from 1995 through 2006. Because the Internet has only recently become the technology-of-choice for learning and teaching at a distance (Moore & Kearsley, 2005), the search was limited to articles that were published after 1994. Electronic searches were performed using various queries, including, for example, self-regulat* AND online, self-regulat* AND Web, and self-regulat* AND distance. The following databases were searched: Academic Search Premier, ERIC, PsychARTICLES, Psychology and Behavioral Sciences Collection, PsycINFO, Web of Science, and Dissertation Abstracts on ProQuest. Once located, abstracts for each study were read and articles that were deemed relevant to the framework described above were retained. Retained articles were printed and read in their entirety.

Understanding the Relationships Between Variables

Much of the research on self-regulation in online education has focused on identifying the motivational, cognitive, and behavioral characteristics of effective self-regulated learners, as well as trying to understand how these components relate to each other and to other adaptive academic outcomes. Using primarily non-experimental, correlational methods, most studies have mirrored the earlier research on self-regulation in traditional classrooms (see, for example, Pintrich & De Groot, 1990; Pintrich & Garcia, 1991). In general, these investigations have attempted to discern if the relationships found in conventional classrooms generalize to online learning environments. Table 1 provides a summary of the non-experimental, correlational studies examined in this review.

Social cognitive theorists assume that effective self-regulation depends, in large part, on students’ confidence in their ability to attain designated types of performances (i.e., their perceived self-efficacy; Bandura, 1997; Zimmerman, 2000a, 2000b). According to Schunk (2005), “self-regulated learners are more self-efficacious for learning than are students with poorer self-regulatory skills; the former believe that they can use their self-regulatory skills to help them learn” (p. 87). As such, researchers interested in using a social cognitive view of self-regulation to understand student performance in online settings have studied, more than any other construct, self-efficacy and its relations to other variables. Overall, results have revealed that when compared to their counterparts with lower perceived self-efficacy, efficacious students report more use of learning strategies (Artino & Stephens, 2006; Joo, Bong, & Choi, 2000); greater satisfaction with their learning experience (Artino, in press; Lim, 2001); increased likelihood of enrolling in future online courses (i.e., choice behaviors; Artino, in press; Lim 2001); and superior academic performance (Bell & Akroyd, 2006; Hsu, 1997; Joo et al., 2000; Lee, 2002; Lynch & Dembo, 2004; Wang & Newlin, 2002). For example, in one of the more comprehensive studies of self-efficacy and its relationship to academic performance, Bell and Akroyd (2006) surveyed 201 undergraduates enrolled in a variety of asynchronous online courses and found that students’ self-efficacy[1] for learning and performance was among the three most powerful predictors of final course grade (β = 1.65, p < .001).

Table 1

Non-Experimental, Correlational Studies of Self-Regulation in Online Education

Moreover, the researchers found that an interaction term (college grade point average [GPA] X self-efficacy) was also a significant individual predictor of final course grade (β = -2.35, p < .001), indicating that self-efficacy beliefs had a greater effect on course grade for students with lower GPAs.

Along with end-of-course grades, several investigators have used student satisfaction with online education as the outcome of interest since satisfaction has been shown to predict course drop-out rates, as well as students’ intentions to enroll in future online courses (for a review, see Dabbagh & Bannan-Ritland, 2005; Moore & Kearsley, 2005; Simonson et al., 2003). For instance, in a study of student satisfaction and choice behaviors, Lim (2001) surveyed 235 adult learners at five American universities and found that computer self-efficacy, along with a linear combination of experiential variables, explained 15% of the variance in students’ overall satisfaction and 12% of the variance in their intentions to enroll in future online courses. And while the effect sizes found in Lim’s (2001) study are considered moderate (Cohen, 1988), Artino (in press) found much larger effects when attempting to predict military students’ satisfaction (model R² = .65) and choice behaviors (model R² = .40) using a linear combination of students’ prior experience, task value, and self-efficacy within the context of self-paced, online training (i.e., computer-based courses accessed through the Internet and completed without an instructor). In this case, however, self-efficacy for learning with self-paced, online training emerged as a significant individual predictor of satisfaction but not choice behaviors. On the other hand, task value—defined as the extent to which students find a task interesting, important, and/or valuable (Eccles & Wigfield, 1995)—was the strongest individual predictor of both satisfaction and choice behaviors. With respect to choice behaviors, these results are consistent with research conducted in traditional classrooms by Eccles and Wigfield (1995), who have shown that value beliefs tend to be better predictors of intentions to take future courses, as well as actual enrollment in those courses, than expectancy beliefs (e.g., self-efficacy).

Other researchers have attempted to use task value as a predictor of adaptive academic outcomes in online settings. In general, results have shown task value to be positively related to students’ metacognition and use of learning strategies (Artino & Stephens, 2006; Hsu, 1997); overall satisfaction (Artino, in press; Lee, 2002); and future enrollment choices (Artino, in press). Unfortunately, only three of the studies reviewed (Artino, in press; Hsu, 1997; Lee, 2002) examined the extent to which task value related to academic performance and learning. In two of these studies (Hsu, 1997; Lee, 2002), the researchers failed to find a significant relationship between task value and course performance when self-efficacy was also included as a predictor. On the other hand, Artino (in press) found that task value was a strong individual predictor of students’ perceived learning (partial r² = .28) when task value was included with self-efficacy and prior experience in a regression model (model R² = .50). In this case, however, the use of a self-report measure of learning was a significant limitation, as this type of subjective measure may bear little resemblance to more direct, performance-oriented outcomes (Mabe & West, 1982; Pace, 1990).

Although investigators have given some attention to the relationships between several motivational components of self-regulation and various academic outcomes (e.g., satisfaction, academic performance, and choice behaviors), very little research has been conducted on how these motivational components relate to students’ academic behaviors, such as their use of cognitive and metacognitive learning strategies. Two exceptions are the studies conducted by Artino and Stephens (2006) and Joo et al. (2000). For example, using path analytic techniques, Joo et al. (2000) found that academic self-efficacy and self-efficacy for SRL both significantly and positively predicted students’ self-reported use of cognitive and metacognitive learning strategies. However, contrary to expectations, neither cognitive nor metacognitive strategies use was related significantly to performance outcomes. Thus, the researchers failed to confirm their hypothesis that learning strategies use mediates the relationship between self-efficacy and student performance. Based on these results, the authors questioned the usefulness of self-reports of strategy use. Furthermore, they recommended that future studies employ more direct, behavioral indicators of learning strategies use to help clarify how students’ motivational characteristics relate to their capacity to apply learning strategies in online environments.

In summary, findings from non-experimental, correlational studies seem to support results from research in traditional classrooms indicating that students’ motivational beliefs about a learning task are related to positive academic outcomes. The existing research in this area, however, suffers from several limitations. First and foremost, results are strictly correlational in nature; therefore, one cannot infer causality from the observed relationships. Although, overall, the results suggest moderate to strong relations between motivational components and adaptive outcomes, the direction of influence between the variables is sometimes ambiguous. For example, although many of the study designs imply that academic performance results, in part, from students’ motivational beliefs, these causal relations could be reversed. Hence, additional research is needed before the exact direction of operation of these social cognitive components can be fully understood.

Second, many of the studies reviewed have suggested that the performance outcomes employed suffered from range restriction, a significant issue in college courses where, often times, the majority of students receive a grade of either A or B. Range restriction has the effect of downwardly biasing the effect size (Cohen, Cohen, West, & Aiken, 2003). Therefore, the failure of several studies to find a significant relationship between motivational components of self-regulation and overall academic performance may have been exacerbated by the restricted range of the criterion measure. Considering this limitation, it is important that future studies utilize other measures of academic success, such as assessments of critical thinking skills and online engagement; outcomes than can be measured through content and discourse analysis of online discussion boards (Hara, Bonk, & Angeli, 2000; Jeong, 2003).

A third limitation of the extant research on self-regulation in online education is a failure to control for prior knowledge when attempting to understand the relations between task value and academic performance. As Tobias (1994) warned in his review of the literature on interest, “research is needed in which both interest and prior knowledge about the same topic are assessed so that the percentages of independent variance attributable to these two constructs may be determined” (p. 50). Because task value includes an interest component, this recommendation is particularly relevant to studies of online education that use task value as an independent variable. 

Finally, the online learning literature is rather limited with respect to the student characteristics investigated. Although self-efficacy and task value have received some emphasis, none of the research reviewed considered the effects of other personal factors, such as different types of affect (mood and emotions); factors that social cognitive theoreticians consider critical to an understanding of individual performance in academic settings (Bandura, 1997; Linnenbrink & Pintrich, 2002; Pekrun, Goetz, Titz, & Perry, 2002).

Self-Regulation and Learner Control

Some researchers have examined how student differences and characteristics of the online environment interact with each other to influence learning. In many ways, these investigations mirror the classic Aptitude-Treatment Interaction (ATI) studies conducted by Cronbach and Snow (1977) that were designed to determine which instructional strategies are more or less effective for particular individuals with specific abilities. As a theoretical framework, ATI posits that optimal learning results when the instruction is closely matched to the aptitudes of the learner.

Using an ATI framework, Eom and Reiser (2000) examined the effects of Self Regulated Learner (SRL) strategies use on achievement and motivation in 37 sixth and seventh graders taking a computer-based course. Essentially, the authors were trying to determine how varying the amount of learner control within the computer-based course might effect the achievement and motivation of students who rated themselves as either high or low in SRL skills. Using a self-report instrument, students were classified as being either high or low self-regulated learners and then were randomly assigned to either a learner-controlled or program-controlled version of a computer-based course. Results revealed that, regardless of how students rated their SRL skills, learners in the program-controlled condition (i.e., learners who had very little control over their progression through the course), “scored significantly higher on a posttest than did learners in the learner-controlled condition” (Eom & Reiser, 2000, p. 247). Additionally, the researchers found that poorer performance in the learner-controlled condition was particularly evident in the students who rated themselves as low self-regulated learners. In fact, students who rated themselves as low in SRL skills scored higher on the posttest (approximately 76.4% higher) when taking the program-controlled condition as compared to the learner-controlled condition. Although this interaction was not statistically significant (perhaps due to inadequate power), the trend supported the researchers’ hypothesis that students with low SRL skills are not as able to learn from computer-based courses that provide high quantities of learner control as students with high self-regulating skills.

In another ATI-type investigation, McManus (2000) attempted to determine what combinations of online course non-linearity (i.e., the extent to which learners were given the opportunity to proceed through the course in a non-linear fashion) and the use of advance organizers (i.e., the presence or absence of short overviews of new material at the beginning of each lesson) would work best for 119 undergraduates reporting different levels of self-regulation. Students’ declarative knowledge was measured by a 12-item, multiple-choice test, and their procedural knowledge was measured by a 20-item, performance assessment. Although the researcher found no significant main effects or interactions, results revealed a near significant interaction between non-linearity and self-regulation (p = .054). According to McManus, these results “suggest that highly self-regulating learners learn poorly in mostly linear Web-based hypermedia learning environments, where they have very few choices, while medium self-regulating learners learn poorly in highly non-linear environments where they are given too many choices” (p. 219). Despite the non-significance of this interaction, the results are promising in that they suggest the ATI framework may be a useful approach that allows researchers to study how individual learner differences and features of the online environment interact with each other to influence learning and performance. 

Taken together, the research on self-regulation and learner control in computer-based environments has failed to find statistically significant results. It is worth noting, however, that these studies, like many others, suffer from serious limitations. For example, in McManus’s (2000) work, scores from the SRL sub-scales possessed marginal internal reliabilities (Cronbach’s alphas ranged from .35 to .67), thereby compromising the ability of the study to uncover noteworthy effects (Thompson, 2003). Additionally, both studies reviewed here attempted to study online instruction by utilizing instruments developed for traditional classrooms. Although some measurement instruments may work equally well in classroom and online settings, considering the differences between the two learning environments, an instrument that works well in the classroom may not be valid in online and/or computer-based learning situations (Tallent-Runnels et al., 2006). It is important, then, that future studies employ appropriate survey instruments that, at the very least, have been pilot tested in the learning environment of interest. Certainly, if the reliability and validity of the measurement instruments used in online education research are not assessed, findings based on those instruments are, at best, questionable (DeVellis, 2003; Tallent-Runnels et al., 2006; Thompson, 2003).

Supporting Self-Regulation in Online Learning Environments

Many experts believe that online learning environments require the learner to assume greater responsibility for the learning process (Dabbagh & Kitsantas, 2004; King et al., 2001; Schunk & Zimmerman, 1998). Furthermore, many of these same experts argue that self-regulatory skills are essential for success in these highly autonomous learning situations and that the development of these skills can be supported by Web-based pedagogical tools (WBPT; Azevedo, 2005; Dabbagh & Kitsantas, 2004; Zimmerman & Tsikalas, 2005). Accordingly, several researchers have attempted to determine the characteristics of effective WBPT, as well as the extent to which various self-regulatory skills might be supported and/or enhanced by these tools (see Table 2 for a summary of these studies). For example, Kramarski and Gutman (2006) randomly assigned 65 ninth graders to one of two online learning environments designed to teach mathematics: one with self-regulatory support (SRS) in the form of metacognitive questioning and the other without explicit support for self-regulation. Results showed that when pre- and post-test scores were compared, students in the SRS group significantly outperformed their counterparts in the non-supported group on all outcome measures, including performance on mathematical explanations, procedural and transfer tasks, and use of SRL strategies. In terms of effect sizes, post-test improvements in the SRS group were moderate on SRL strategies use (d = .45) to large on mathematical explanations (d = 2.24; Cohen, 1988).

Using a similar conceptual framework, Azevedo, Cromley, and Seibert (2004) confirmed the positive benefits of online self-regulatory support. In this case, however, the researchers randomly assigned 51 undergraduates to one of three computer-based scaffolding conditions: adaptive scaffolding (AS; i.e., a teacher or tutor who continuously diagnoses students’ understanding), fixed scaffolding (FS), and no scaffolding (NS). Using a mixed-methods approach, the authors found that AS facilitated positive shifts in students’ mental models (as assessed through the coding of student diagrams) significantly more than FS and NS. Furthermore, the researchers analyzed verbalizations of students’ learning activities and found that a significantly larger number of participants in the AS condition planned their learning, monitored their progress, and used learning strategies. Results from this study were particularly noteworthy because the authors used both qualitative and quantitative methods to analyze both performance and process data.

In summary, results from this line of research have been promising and suggest the following practical and theoretical implications: (1) Web-Based Pedagogical Tools (WBPT) can be an effective way to support and/or enhance students’ self-regulatory skills (Azevedo et al., 2004; Dabbagh & Kitsantas, 2005; Kauffman, 2004; Kramarski & Gutman, 2006; Niemi, Nevgi, & Virtanen, 2003); (2) adaptive scaffolding appears to be more effective in supporting students self-regulatory processes and academic performance than fixed or no scaffolding (Azevedo et al., 2004); (3) different types of WBPTs support different self-regulatory processes (Dabbagh & Kitsantas, 2005; Kauffman, 2004; Kramarski & Gutman, 2006); and (4) WBPTs may be more effective for novice learners with under-developed self-regulatory skills than for veteran learners with more advanced SRL skills (Niemi et al., 2003).

Table 2

Studies of Computer-Based Tools Designed to Support Students’ Self-Regulation

 *Note. A WebQuest is an inquiry-based instructional tool designed to facilitate search and synthesis of information from multiple sources (Dodge, 1997 as cited in Kauffman, 2004).

In terms of research quality, these self-regulatory scaffolding studies tended to use superior research methods when compared to much of the empirical work on self-regulation in online education. For example, in three of five WBPT studies, researchers randomly assigned participants to treatment and control/comparison groups, thereby enhancing the internal validity of their experiments and improving their ability to establish causal relationships (Shadish, Cook, & Campbell, 2002). Additionally, all of the investigators used multiple outcome measures and employed both qualitative and quantitative methods to analyze their data. Taken together, these studies have taken a positive step toward improving the methodological quality of research in online distance education (Abrami & Bernard, 2006; Bernard et al., 2004a).

Conclusions

Clearly, more well-designed research is needed on self-regulation and its influence on student success in online learning environments. To date, most studies in this burgeoning field have been descriptive in nature and have suffered from numerous methodological limitations. Despite these limitations, however, the studies reviewed here seem to support the linkages between students’ motivational beliefs about a learning task, their use of learning strategies, and their performance in online settings. Furthermore, although the empirical support is thin, it appears that highly self-regulated learners may have more success in learner-controlled environments than their peers with poorer self-regulatory skills. Finally, some of the highest quality research in online education seems to indicate that providing students with self-regulatory scaffolding can be an effective instructional method—one that instructional designers might do well to consider including as integral to their online courses.

Ultimately, the existing empirical literature supports the trends established in research with more traditional classrooms; specifically, that self-regulation is an important, if not essential skill for effective learning and performance (Miltiadou & Savenye, 2003; Schunk & Zimmerman, 1994, 1998). Therefore, future research should continue to explore self-regulation in online education, with the intent of determining which instructional elements, as well as which existing personal characteristics, behaviors, and attitudes, contribute to achievement in and satisfaction with this emergent form of instruction (Bernard et al., 2004).

References

About the Author

Anthony R. Artino, Jr. is a doctoral candidate in the Cognition and Instruction Program, Department of Educational Psychology, Neag School of Education, University of Connecticut. Anthony is interested in online learning and its use within both higher education and military training environments. In his research, he uses social cognitive views of academic self-regulation to investigate how motivational beliefs and achievement emotions influence student success within highly autonomous, online learning situations. He can be reached at tony_artino@yahoo.com.
 

Anthony R. Artino Jr.
Doctoral Candidate, Cognition and Instruction Program
Department of Educational Psychology
Neag School of Education, University of Connecticut
373 Squaw Hollow Road
Ashford, CT 06278

Phone: (cell) 860.942.9345,

(home) 860.429.0116

Fax: (work) 860.486.3510 

Email: tony_artino@yahoo.com