Despite the growing body of literature on the role of motivation in self-regulation of learning, little evidence exists whether changes in motivational beliefs are associated with the changes in self-regulatory strategy use over-time. Moreover, little is known on the mediational roles of these learning strategies in accounting for variance in achievement. The present study examined the developmental course of motivational beliefs (competence beliefs and task values) and their contribution to self-regulated learning strategies (shallow and deep cognitive and meta-cognitive strategies)  and math achievement in a sample of 498 students (51% girls) in a secondary school. Latent growth analysis showed that the declines in self-regulated leaning strategies are associated with changes in the motivational beliefs. More specifically, changes in competence beliefs, intrinsic value, attainment value and utility value of mathematics were associated with changes in the three self-regulated learning strategies. These findings are consistent with cross-sectional studies that examined the role of these motivational beliefs in self-regulation of learning (e.g., Wolters & Pintrich, 1998). Bootstrap analyses of the mediational roles of learning strategies showed that the learning strategies jointly as well as uniquely, partially mediated the effect of motivational beliefs on achievement. The findings lend support to the theoretical assumptions in the literature that self-regulatory strategy use plays a mediational role.

Introduction

Currently, a number of self-regulated learning models have been developed (e.g. Boekaerts, 1997; Pintrich, 2000; Rheinberg, et al., 2000, Zimmerman & Schunk, 1994). Although the key constructs that these models propose may vary from one model to another, their core assumption is that motivation is the key to self-regulation. Several studies have confirmed this general premise in the past decades. Studies have shown that students who believe that they are capable of executing a given activity are able to employ self-regulated learning strategies more efficiently than those who doubt their confidence (Pajares, 2008). Similarly, a number of studies have shown that students who adapt a learning goal orientation are more adept at using deep learning strategies than those who adopt a performance goal orientation (Fryer & Elliot, 2008). Studies have also found that students who attribute their success or failure to controllable factors are more likely to use self-regulated learning strategies than those who attribute the cause of their success or failure to uncontrollable factors (Schunk, 2008).   Finally, research has shown that students who value what they learn are more likely to employ self-regulated learning strategies than those who do not value what they learn (Wigfield et al., 2008). Although such studies have broadened our knowledge, several issues remain unclear. First, longitudinal studies on the relationship between motivational beliefs and self-regulated learning strategies are very limited. Secondly, regardless of the increasing focus on the role of motivational beliefs in self-regulation of learning, the roles of task values in self-regulation have been given scant attention (Brophy, in press, Wigfield et al., 2008). More specifically, the role of intrinsic value (interestingness of a task), attainment value (importance of doing well on a task for confirming ones self-schema) and utility value (usefulness of a task for an individual’s future goals) have not been examined fully(but see Lens Vansteenkiste, 2008). Finally, despite the the theoretical presumptions that self-regulated learning would mediate the associations between the motivational beliefs and achievement, only very few cross-sectional studies have tested this important assumption.

Objectives

The objectives of the study were (a) to examine whether competence beliefs and values

(intrinsic, attainment and utility) explain the changes in the use of self-regulated learning strategies (shallow and deep cognitive strategies and meta-cognitive strategies) over time; (b) to investigate the mediational roles of the self-regulated learning strategies in the associations between the motivational beliefs and achievement.

Method

The participants of the study were 498 grade 7 students (52% girls) in a suburban community. They completed questionnaires on competence and value beliefs as well as self-regulated learning strategies both adapted from existing measures (Wigfield &Eccles, 2000; Wolters, etal., 2005  respectively) three times in a one year longitudinal study. To explain changes in self-regulated learning strategies we used latent growth curve modelling. To test whether the self-regulated learning strategies mediated the associations between the motivational beliefs and achievement, we used a bootstrapping analysis.

Results

The first objective of the study was to examine whether competence beliefs and components of task value explain the changes in the use of self-regulated learning strategies. To attain this objective we tested 12 separate linear latent growth curve models (i.e. 4 motivational beliefs X 3 learning strategies). Among the 12 models tested, 9 models fitted the data reasonably well [(RMSEAs (.08 to .01),CFIs(.90, to.96)]. The results showed that in almost all of the models, the initial motivational beliefs explained the initial level (intercept) of the learning strategies and the changes in the motivational beliefs explained the changes (slopes) in the learning strategies. More specifically, the changes in competence beliefs explained the changes in shallow, deep and meta-cognitive strategy use. Similarly, changes in intrinsic and utility values predicted the slope of surface and meta-cognitive learning strategies  but only the intercept of intrinsic value predicted intercepts of the three strategies.  Attainment value on the other hand influenced the slope of all the strategies but its influence on deep strategy intercept was negative. Most of the predictions are supported and are in line with Wigfield et al.’s (2008) argument that task value components may relate to components of self-regulated learning.

            The second objective of the study was to examine the extent to which the three learning strategies mediated (jointly and uniquely) the associations between the motivational beliefs and achievement. To answer this research question we utilized motivational beliefs measured at time 1 and learning strategies measured at time 2 and achievement measured at time 3. We utilized a relatively new technique of testing multiple and unique mediation effects –a bootstrapping strategy. This strategy is helpful in discerning whether the three learning strategies jointly mediate the associations between the motivational beliefs and achievement. It also tells us whether a learning strategy has a unique effect, ones other strategies are controlled. Analyses results showed that after partailling out the effects of prior achievement, the three learning strategies jointly mediated the association between the four motivational variables and achievement(bootstrap confidence intervals were CI.₉₅  (.06, .22) for competence, CI.₉₅  (.09,.22)  for intrinsic value CI.₉₅  (.07,.21)  for attainment value and finally CI.₉₅  (.04,.18)  for utility value. In addition, the analyses for the unique effects[1] of the mediators revealed that whereas shallow learning strategies and meta-cognitive strategies uniquely mediated the association between both competence and intrinsic value and achievement, deep strategy did not. Deep and meta-cognitive strategies uniquely mediated the effect of attainment value. Shallow strategy did not have a unique effect. Finally, while shallow and deep strategies uniquely mediated the effect of utility value on achievement, meta-cognitive strategy did not.

Significance of the study

We believe that the results of the current study add to the existing work on the role of motivation in self-regulated learning. Particularly, the study provides evidence on the role of task value components (intrinsic value, attainment value and utility value) in self-regulated learning. The findings also suggest that attending to students’ competence and value related beliefs has the dual advantage of enhancing and maintaining self-regulated learning as well as improving achievement.