**4. Results**

Table 1 shows the means and standard deviations for all the variables in this study. Table 2 shows the correlation findings among these variables. Mastery-development goals were positively associated with deep and all forms of self-regulatory strategies. They were also associated positively with learning attitudes, efficacy and control beliefs. As expected, these goals were negatively related to surface strategies. Extrinsic work-related goals were associated with the use of deep and effort management strategies. Also, these goals were


Note 1: \* *p*<.05; \*\**p*<.01

Table 1. Descriptive statistics, reliability scores and correlation analyses.

The Role of Self-Efficacy, Control Beliefs and

*Step 1* 

*Step 2* 

*Step 3* 

goals

goals

goals

*Step 4* 

goals

goals

goals

Mastery-development

Performance-approach

Social enhancement

Mastery-development

Performance-approach

Social enhancement

Efficacy x social enhancemen

Note. \*\* p<.0001; \*p<.005

Table 2. Regression analyses 1.

Predictors Deep Surface Self-

Achievement Goals on Learning Among Distance Learners 241

Age 1 -.06 .13\* -.02 -.14 -.07 .06 -.16\* Age 2 .20\* -.08 .11 .09 .07 .15\* .07

Age 1 -.03 .12 .02 -.11 -.05 .08 -.13\* Age 2 .18\* -.07 .10 .08 .06 .14\* .05 Efficacy beliefs .39\*\* -.19\*\* .43\*\* .28\*\* .28\*\* .27\*\* .36\*\*

Age 1 .004 .07 .03 -.12 -.05 .08 -.04 Age 2 .16\*\* -.05 .10 .08 .02 .13\* -.01 Efficacy beliefs .26\*\* -.17\*\* .34\*\* .21 .17\* .22\* .19\*\*

Extrinsic work goals .01 .08 -.11\* -.07 .08 -.01 .13\*

Age 1 .002 .08 .03 -.12 -.06 .07 -.04 Age 2 .16\* -.04 .11 .09 .02 .13\* -.01 Efficacy beliefs .28\*\* -.16\* .35\*\* .22\*\* .17\* .22\* .20\*\*

Extrinsic work goals .01 .07 -.11\* -.08 .09 -.01 .14\*

Efficacy x Mastery .04 .04 -.05 -.003 .04 -.02 .000 Efficacy x Performance -.09 -.03 -.03 .001 -.04 -.02 -.07 Efficacy x Extrinsic work .03 -.10 -.01 -.02 .06 .05 .03

(1) R2 = .05, p<.0001 for Step 1; R2 ∆ =.15, p<.0001 for Step 2; R2 ∆ =.10, p<.0001 for Step 3; R2 ∆ =.01, p =.29 for Step 4 (2) R2 = .04, p<.005 for Step 1; R2 ∆ =.04, p<.001 for Step 2; R2 ∆ =.05, p<.005 for Step 3; R2 ∆ =.01, p=.33 for Step 4 (3) R2 = .02, p=.103 for Step 1; R2 ∆ =.18, p<.0001 for Step 2; R2 ∆ =.05, p<.0001 for Step 3; R2 ∆ =.01, p=.50 for Step 4 (4) R2 = .04, p<.005 for Step 1; R2 ∆ =.08, p<.0001 for Step 2; R2 ∆ =.03, p=.06 for Step 3; R2 ∆ =.005, p=.80 for Step 4 (5) R2 = ..02, p=.10 for Step 1; R2 ∆ =.08, p<.0001 for Step 2; R2 ∆ =.07, p<.0001 for Step 3; R2 ∆ =.01, p=.53 for Step 4 (6) R2 = .02, p=.08 for Step 1; R2 ∆ =.07, p<.0001 for Step 2; R2 ∆ =.01, p=.56 for Step 3; R2 ∆ =.01, p=.67 for Step 4 (7) R2 = .04, p<.001 for Step 1; R2 ∆ =.13, p<.0001 for Step 2; R2 ∆ =.33, p<.0001 for Step 3; R2 ∆ =.01, p=.52 for Step 4

.34\*\* -.23\*\* .20\*\* .11 .23\*\* .09 .62\*\*

.08 .000 .13\* .15\*\* .17\* .07 -.08


.35\*\* -.22\*\* .20\*\* .11 .23\*\* .08 .63\*\*

.07 .000 .13\* .15\* .16\* .07 -.09


.07 -.04 .11 .07 .04 .08 -.03

monitoring

Time Effort Help

seeking

Attitudes

positively related to learning attitudes, efficacy and control beliefs. Performance approach goals were associarted with the use of adaptive strategies including deep, self-monitoring, time management, effort management and help-seeking strategies. Despite that these performance considerations were correlated positively with efficacy and control beliefs, they were not associated with learning attitudes. Finally, the correlation analyses showed that social enhancement goals were associated with self-monitoring strategies only. These goals were positively related to learning attitudes, efficacy and control beliefs. Taken together these analyses showed that there was a close relationship between goals, the use of strategies, learning attitudes and motivational beliefs. Mastery-development goals had the strongest correlation with adpative learning strategies, regulatory strategies and learning attitudes.

Prior research on achievement goals (e.g. Miller *et al*., 1993; Kaplan & Midgley, 1997) showed that students' self-efficacy is an important predictor in the use of learning strategie and will moderate the effects of achievement goals on learning. In this study, both efficacy and control beliefs were assumed to moderate the effects of various goals on strategies and attitudes. Both efficacy and control beliefs were included in the regression models to assess their relative predictive ability as compared to different types of goals. Two sets of hierarchically ordered regressions were conducted: 1. *Regressing efficacy beliefs, mastery development goals, extrinsic work goals, performance-approach goals and social enhancement goals on learning strategies, self-regulatory strategies and learning attitudes; 2. Regressing control beliefs, mastery-development goals, extrinsic work goals, performance-approach goals and social enhancement goals on learning strategies, self-regulatory strategies and learning attitudes.* To control for the effects of age difference, two dummy-coded age variables were entered in the first step, followed by efficacy or control beliefs in step two, the four goals in step three, and the interaction terms were entered in the final step. Following Aiken and West (1991), all the predictor variables were centred and the interaction terms were constructed using these centred variables. Dependent variables were kept in their original metric. The centring procedure reduces multicollinearity among first order variables and the interaction terms. Table 2 shows the standardised coefficients in the regression equations. The predictors including dummy-coded age groups, efficacy or control beliefs, the four goals and goal x belief interaction explained a significant amount of variance (as indicated in R2 ) in the use of learning strategies, self-regulatory strategies, and learning attitudes.

This study expected to find significant interaction terms between efficacy beliefs and the four goals on the use of learning strategies, self-regulatory strategies, and learning attitudes. Contrary to our prediction, the current results did not provide any support to these hypothesised interactions. Regression analyses did not locate any significant interactions.

Table 3 shows the result when control beleifs was taken as a mediator replacing efficacy beliefs. As can be seen, the results showed that this variable moderated the effects of extrinsic work goals and mastery-development goals on learning strategies and attitudes. Following Aiken and Wesr (1991) these significant interactions were interpreted first before examining the main effects. These significant interactions indicated the following:



positively related to learning attitudes, efficacy and control beliefs. Performance approach goals were associarted with the use of adaptive strategies including deep, self-monitoring, time management, effort management and help-seeking strategies. Despite that these performance considerations were correlated positively with efficacy and control beliefs, they were not associated with learning attitudes. Finally, the correlation analyses showed that social enhancement goals were associated with self-monitoring strategies only. These goals were positively related to learning attitudes, efficacy and control beliefs. Taken together these analyses showed that there was a close relationship between goals, the use of strategies, learning attitudes and motivational beliefs. Mastery-development goals had the strongest correlation with adpative learning strategies, regulatory strategies and learning attitudes.

Prior research on achievement goals (e.g. Miller *et al*., 1993; Kaplan & Midgley, 1997) showed that students' self-efficacy is an important predictor in the use of learning strategie and will moderate the effects of achievement goals on learning. In this study, both efficacy and control beliefs were assumed to moderate the effects of various goals on strategies and attitudes. Both efficacy and control beliefs were included in the regression models to assess their relative predictive ability as compared to different types of goals. Two sets of hierarchically ordered regressions were conducted: 1. *Regressing efficacy beliefs, mastery development goals, extrinsic work goals, performance-approach goals and social enhancement goals on learning strategies, self-regulatory strategies and learning attitudes; 2. Regressing control beliefs, mastery-development goals, extrinsic work goals, performance-approach goals and social enhancement goals on learning strategies, self-regulatory strategies and learning attitudes.* To control for the effects of age difference, two dummy-coded age variables were entered in the first step, followed by efficacy or control beliefs in step two, the four goals in step three, and the interaction terms were entered in the final step. Following Aiken and West (1991), all the predictor variables were centred and the interaction terms were constructed using these centred variables. Dependent variables were kept in their original metric. The centring procedure reduces multicollinearity among first order variables and the interaction terms. Table 2 shows the standardised coefficients in the regression equations. The predictors including dummy-coded age groups, efficacy or control beliefs, the four goals and goal x belief interaction explained a significant amount of variance (as indicated in R2 ) in the use of

This study expected to find significant interaction terms between efficacy beliefs and the four goals on the use of learning strategies, self-regulatory strategies, and learning attitudes. Contrary to our prediction, the current results did not provide any support to these hypothesised interactions. Regression analyses did not locate any significant interactions. Table 3 shows the result when control beleifs was taken as a mediator replacing efficacy beliefs. As can be seen, the results showed that this variable moderated the effects of extrinsic work goals and mastery-development goals on learning strategies and attitudes. Following Aiken and Wesr (1991) these significant interactions were interpreted first before

1. Extrinsic work goals predicated negatively the use of deep strategies and such a

2. Extrinsic work goals predicted positively the use of surface strategies and this

3. Mastery-development goals predicted negatively the use of surface strategies and this

examining the main effects. These significant interactions indicated the following:

relationship was less pronounced when control beliefs were strong;

relationship was more pronounced when control beliefs were strong;

relationship was weakened when control beliefs were strong;

learning strategies, self-regulatory strategies, and learning attitudes.


Note. \*\* p<.0001; \*p<.005

(1) R2 = .05, p<.0001 for Step 1; R2 ∆ =.15, p<.0001 for Step 2; R2 ∆ =.10, p<.0001 for Step 3; R2 ∆ =.01, p =.29 for Step 4 (2) R2 = .04, p<.005 for Step 1; R2 ∆ =.04, p<.001 for Step 2; R2 ∆ =.05, p<.005 for Step 3; R2 ∆ =.01, p=.33 for Step 4 (3) R2 = .02, p=.103 for Step 1; R2 ∆ =.18, p<.0001 for Step 2; R2 ∆ =.05, p<.0001 for Step 3; R2 ∆ =.01, p=.50 for Step 4 (4) R2 = .04, p<.005 for Step 1; R2 ∆ =.08, p<.0001 for Step 2; R2 ∆ =.03, p=.06 for Step 3; R2 ∆ =.005, p=.80 for Step 4 (5) R2 = ..02, p=.10 for Step 1; R2 ∆ =.08, p<.0001 for Step 2; R2 ∆ =.07, p<.0001 for Step 3; R2 ∆ =.01, p=.53 for Step 4 (6) R2 = .02, p=.08 for Step 1; R2 ∆ =.07, p<.0001 for Step 2; R2 ∆ =.01, p=.56 for Step 3; R2 ∆ =.01, p=.67 for Step 4 (7) R2 = .04, p<.001 for Step 1; R2 ∆ =.13, p<.0001 for Step 2; R2 ∆ =.33, p<.0001 for Step 3; R2 ∆ =.01, p=.52 for Step 4

Table 2. Regression analyses 1.

The Role of Self-Efficacy, Control Beliefs and

interactions between control beliefs and goals were located.

factors in the learning process for distance learners.

*p<*.001 controlling for both efficacy and control beliefs).

relation to their career concerns.

Achievement Goals on Learning Among Distance Learners 243

no significant interaction effects between efficacy and goals. As for control beliefs, the analyses focused on the regression model in Step 4 in which several cases of significant

These two regression models showed that efficacy and control beliefs were important predictors of learning strategies, regulatory strartegies and learning attitudes. In particular, learners' efficacy beliefs predicted positively the use of deep strategies(=.26, *p<*.001), selfmonitoring strategies (=.34, *p<*.001), time management strategies (=.21, *p<*.001), effort management strategies (=.17, *p<*.001), help-seeking strategies (=.22, *p<*.001), and finally a favourable learning attitude (=.19, *p<*.001). Learners' efficacy beliefs predicted negatively the use of surface strategies (=-.17, *p<*.001). A similar pattern of predicted relationships was found between control beliefs and learners' learning and attitudes. In particular, learners' control beliefs predicted positively the use of deep strategies (=.20, *p<*.001, Step 4), regulatory strategies (=.21, *p<*.001), effort management strategies (=.21, *p<*.001), helpseeking strategies (=.28, *p<*.001), and finally, a favourable learning attitude (=.24, *p<*.001). In short, these findings confirmed that efficacy and control beliefs were significant cognitive

Another major aim of the current study was to explore the relative importance of various goals on learning among distance learners. The relative importance of these goals in predicting the use of learning, self-regulatory strategies and learners' attitudes towards learning was analysed based on the main effects while taking into consideration the effects of efficacy and control beliefs in the regression equations. After controlling for the level of efficacy and control beliefs, mastery-development goals were the most important variable predicting the levels of deep strategies (=.38, *p<*.001 controlling for efficacy levels; =.34, *p<*.001 controlling for control beliefs), effort management strategies (=.25, *p<*.001 after controlling for efficacy beliefs; =.21, *p<*.001 controlling for control beliefs) and learning attitudes (=.61, *p<*.001 controlling for both efficacy and control beliefs). In addition, mastery-development goals predicted positively the use of self-monitoring strategies (=.21, *p<*.001 controlling for efficacy beliefs; =.22, *p<*.001 controlling for control beliefs). As expected these adaptive goals predicted negatively the use of surface strategies (=-.23,

Performance-approach goals predicted positively the use of self-monitoring strategies (=.13, *p<*.001 controlling for efficacy beliefs; =.21, *p<*.001 controlling for control beliefs), time management strategies (=.15, *p<*.001 controlling for efficacy beliefs; =.20, *p<*.001 controlling for control beliefs), and effort management strategies (=.16, *p<*.001 controlling for efficacy beliefs; =.19, *p<*.001 controlling for control beliefs). These results suggest that distance learners focusing on outperforming others will learn in an organised and regulated manner. Extrinsic work goals did not predict the use of learning and self-regulatory strategies. These work goals however predicted positively learning attitudes (=.13, *p<*.001 controlling for both efficacy and control beliefs). This result confirmed the extrinsic nature of these goals to learning. Distance learners holding these goals focused more on the product of learning in

Finally social enhancement goals predicted positively the use of surface strategies (=.16, *p<*.001 controlling for efficacy beliefs; =.17, *p<*.001 controlling for control beliefs) but


Note. \*\* p<.0001; \*p<.005

(1) R2 = .06, p<.0001 for Step 1; R2 ∆ =.13, p<.0001 for Step 2; R2 ∆ =.11, p<.0001 for Step 3; R2 ∆ =.02, p<.05 for Step 4 (2) R2 = .04, p<.005 for Step 1; R2 ∆ =.03, p<.005 for Step 2; R2 ∆ =.05, p<.005 for Step 3; R2 ∆ =.04, p<.005 for Step 4 (3) R2 = .02, p=.10 for Step 1; R2 ∆ =.08, p<.0001 for Step 2; R2 ∆ =.10, p<.0001 for Step 3; R2 ∆ =.01, p=.63 for Step 4 (4) R2 = .04, p<.005 for Step 1; R2 ∆ =.03, p<.005 for Step 2; R2 ∆ =.05, p<.005 for Step 3; R2 ∆ =.02, p=.29 for Step 4 (5) R2 = .01, p=.11 for Step 1; R2 ∆ =.09, p<.0001 for Step 2; R2 ∆ =.07, p<.0001 for Step 3; R2 ∆ =.01, p=.58 for Step 4 (6) R2 = .02, p=.10for Step 1; R2 ∆ =.10, p<.0001 for Step 2; R2 ∆ =.01, p=.37 for Step 3; R2 ∆ =.01, p=.53 for Step 4 (7) R2 = .04, p<.005 for Step 1; R2 ∆ =.18, p<.0001 for Step 2; R2 ∆ =.29, p<.0001 for Step 3; R2 ∆ =.02, p<.05 for Step 4

Table 3. Regression analyses 2.

Now, let us examine the main effects shown in Table 2 and 3. In the case of efficacy beliefs, the analyses below focused on the regression model resulted from Step 3 because there was

Age 1 -.07 .14\* -.02 -.14\* -.07 .05 -.16\* Age 2 .19\* -.08 .11 .09 .07 .14\* .07

Age 1 -.04 .12 .01 -.13 -.05 .07 -.13\* Age 2 .16\* -.06 .09 .07 .04 .11 .02 Control beliefs .36\*\* -.18\* .29\*\* .16\* .30\*\* .32\*\* .43\*\*

Age 1 .02 .09 .01 -.13 -.05 .07 -.05 Age 2 .15\* -.04 .10 .08 .02 .12 -.02 Control beliefs .23\*\* -.17\* .21\*\* .11 .21\*\* .29\*\* .24\*\* Mastery-development goals .33\*\* -.23\*\* .22\*\* .13 .20\* .05 .60\*\*

Extrinsic work goals .00 .09 -.11\* -.07 .06 -.04 .11\* Social enhancement goals -.03 .15\* .02 -.05 -.18\* -.01 -.07

Age 1 -.03 .10 .00 -.14\* -.06 .06 -.06 Age 2 .13\* -.03 .10 .08 .01 .11 -.04 Control beliefs .20\*\* -.10 .19\*\* .10 .21\* .29\*\* .21\*\* Mastery-development goals .34\*\* -.23\*\* .22\*\* .12 .21\* .06 .61\*\*

Extrinsic work goals .02 .07 -.10 -.06 .08 -.03 .13\* Social enhancement goals -.04 .17\* .02 -.03 -.20\* -.04 -.08 Control x Mastery -.08 .17\* -.04 .11 -.03 -.09 -.09 Control x Performance .00 -.04 .02 -.03 .07 -.03 -.06 Control x Extrinsic work .15\* -.17\* .08 -.002 -.06 .05 .08

(1) R2 = .06, p<.0001 for Step 1; R2 ∆ =.13, p<.0001 for Step 2; R2 ∆ =.11, p<.0001 for Step 3; R2 ∆ =.02, p<.05 for Step 4 (2) R2 = .04, p<.005 for Step 1; R2 ∆ =.03, p<.005 for Step 2; R2 ∆ =.05, p<.005 for Step 3; R2 ∆ =.04, p<.005 for Step 4 (3) R2 = .02, p=.10 for Step 1; R2 ∆ =.08, p<.0001 for Step 2; R2 ∆ =.10, p<.0001 for Step 3; R2 ∆ =.01, p=.63 for Step 4 (4) R2 = .04, p<.005 for Step 1; R2 ∆ =.03, p<.005 for Step 2; R2 ∆ =.05, p<.005 for Step 3; R2 ∆ =.02, p=.29 for Step 4 (5) R2 = .01, p=.11 for Step 1; R2 ∆ =.09, p<.0001 for Step 2; R2 ∆ =.07, p<.0001 for Step 3; R2 ∆ =.01, p=.58 for Step 4 (6) R2 = .02, p=.10for Step 1; R2 ∆ =.10, p<.0001 for Step 2; R2 ∆ =.01, p=.37 for Step 3; R2 ∆ =.01, p=.53 for Step 4 (7) R2 = .04, p<.005 for Step 1; R2 ∆ =.18, p<.0001 for Step 2; R2 ∆ =.29, p<.0001 for Step 3; R2 ∆ =.02, p<.05 for Step 4

Now, let us examine the main effects shown in Table 2 and 3. In the case of efficacy beliefs, the analyses below focused on the regression model resulted from Step 3 because there was

.14\* -.04 .22\*\* .21\*\* .21\*\* .11 -.04

.12\* -.03 .21\*\* .20\* .19\* .11 -.05


monitoring

Time Effort Help

seeking

Attitudes

Predictors Deep Surface Self-

*Step 1* 

*Step 2* 

*Step 3* 

goals

*Step 4* 

goals

Performance-approach

Performance-approach

Control x social enhancement

Note. \*\* p<.0001; \*p<.005

Table 3. Regression analyses 2.

no significant interaction effects between efficacy and goals. As for control beliefs, the analyses focused on the regression model in Step 4 in which several cases of significant interactions between control beliefs and goals were located.

These two regression models showed that efficacy and control beliefs were important predictors of learning strategies, regulatory strartegies and learning attitudes. In particular, learners' efficacy beliefs predicted positively the use of deep strategies(=.26, *p<*.001), selfmonitoring strategies (=.34, *p<*.001), time management strategies (=.21, *p<*.001), effort management strategies (=.17, *p<*.001), help-seeking strategies (=.22, *p<*.001), and finally a favourable learning attitude (=.19, *p<*.001). Learners' efficacy beliefs predicted negatively the use of surface strategies (=-.17, *p<*.001). A similar pattern of predicted relationships was found between control beliefs and learners' learning and attitudes. In particular, learners' control beliefs predicted positively the use of deep strategies (=.20, *p<*.001, Step 4), regulatory strategies (=.21, *p<*.001), effort management strategies (=.21, *p<*.001), helpseeking strategies (=.28, *p<*.001), and finally, a favourable learning attitude (=.24, *p<*.001). In short, these findings confirmed that efficacy and control beliefs were significant cognitive factors in the learning process for distance learners.

Another major aim of the current study was to explore the relative importance of various goals on learning among distance learners. The relative importance of these goals in predicting the use of learning, self-regulatory strategies and learners' attitudes towards learning was analysed based on the main effects while taking into consideration the effects of efficacy and control beliefs in the regression equations. After controlling for the level of efficacy and control beliefs, mastery-development goals were the most important variable predicting the levels of deep strategies (=.38, *p<*.001 controlling for efficacy levels; =.34, *p<*.001 controlling for control beliefs), effort management strategies (=.25, *p<*.001 after controlling for efficacy beliefs; =.21, *p<*.001 controlling for control beliefs) and learning attitudes (=.61, *p<*.001 controlling for both efficacy and control beliefs). In addition, mastery-development goals predicted positively the use of self-monitoring strategies (=.21, *p<*.001 controlling for efficacy beliefs; =.22, *p<*.001 controlling for control beliefs). As expected these adaptive goals predicted negatively the use of surface strategies (=-.23, *p<*.001 controlling for both efficacy and control beliefs).

Performance-approach goals predicted positively the use of self-monitoring strategies (=.13, *p<*.001 controlling for efficacy beliefs; =.21, *p<*.001 controlling for control beliefs), time management strategies (=.15, *p<*.001 controlling for efficacy beliefs; =.20, *p<*.001 controlling for control beliefs), and effort management strategies (=.16, *p<*.001 controlling for efficacy beliefs; =.19, *p<*.001 controlling for control beliefs). These results suggest that distance learners focusing on outperforming others will learn in an organised and regulated manner.

Extrinsic work goals did not predict the use of learning and self-regulatory strategies. These work goals however predicted positively learning attitudes (=.13, *p<*.001 controlling for both efficacy and control beliefs). This result confirmed the extrinsic nature of these goals to learning. Distance learners holding these goals focused more on the product of learning in relation to their career concerns.

Finally social enhancement goals predicted positively the use of surface strategies (=.16, *p<*.001 controlling for efficacy beliefs; =.17, *p<*.001 controlling for control beliefs) but

The Role of Self-Efficacy, Control Beliefs and

approach goals, on strategy use and learning attitudes.

regardless their level of efficacy beliefs.

Achievement Goals on Learning Among Distance Learners 245

salience of performance emphasis. Elliot and Dweck (1988) have created a high level of performance salience focusing participants to their perceived abilities to complete a specific experimental task. In this type of highly controlled setting, one's performance concerns and the demand of performance demonstration are heightened. Students' perceived level of efficacy will therefore be a crucial factor in moderating their performance goals on learning and achievement. Similarly, Church, Elliot & Gable (2001) and Braten and colleagues (2004) based on empirical evidence argued that the mediation role of self-efficacy with goals on learning will be more salient in a highly competitive and evaluative learning environment that focuses on relative performance. The participants and their learning context in the current study were radically different from these previous studies that used on-campus undergraduate students learning in a competitive environment or completing experimental tasks according to a set explicit performance criteria. Rather, the distance learning environment in this study promoted a mastery-focused orientation. Distance learners in this study engaged in their courses through distance learning mode in which face-to-face contact with other learners were limited to optional fortnightly tutorials. Distance learners are expected to complete the assigned readings and learning materials on their own, taking into consideration a suggested time-schedule for monitoring the progress themselves. In other words, the chance for distance learners to compete with each other was limited. Therefore, it can be argued that a lack of emphasis on relative performance might have led to the nonsignificant mediation effect of efficacy beliefs with goals, especially performance-

Not only did the distance learning environment promoted mastery, the current sample of distance learners also held strong mastery goals ( *X* = 3.82). A mastery-focused learning system coupled with mastery-focused personal motivation would probably allow learners in this study to engage in adaptive patterns of strategy use and attitudinal development regardless of their level of efficacy beliefs (Dweck, 1986; Kaplan & Midgley, 1997). Kaplan and Midgley (1997, p.431) when concluding their study examining the moderation hypotheses stated that "*in an environment in which learning goals were emphasised more, level of perceived competence might become less influential for students with a predominant learning goal orientation*". In other words, mastery goals and mastery learning environment have predisposed distance learners to an adaptive pattern of strategy use and learning attitudes

In contrast, the present study found several counts of significant interaction results between control beliefs, extrinsic work goals, and mastery goals. These significant results, though limited, confirmed the hypotheses set for the interaction between control beliefs and goals. These significant mediation effects indicate clearly that control beliefs will enhance the positive effects of adaptive goals such as mastery-development goals and dampen the negative effect of less adaptive goals, in this case, extrinsic work goals. These results suggest that control beliefs are critically important for distance learners. Within a distance learning system, distance learners are free to determine as to when they learn, how they learn and for what reasons they learn. A strong sense of control for distance learners means that they will be able to regulate their learning pace, determine the appropriate use of learning and regulatory strategies and adopt different goals for learning. This study showed that a strong sense of control beliefs enabled learners to use fewer surface strategies and more deep strategies even when holding extrinsic work goals that draw them away from engaging in the learning process. As for the case of mastery-development goals, a strong sense of control

negatively the use of effort management strategies (=-.20, *p<*.001 controlling for both efficacy and control beliefs). These results suggest that these social goals will lead distance learners to reduce their effort in learning by using surface strategies. Unexpectedly, these social goals did not show to have positive effects on learning attitudes.

Taken together, regression analyses found that mastery-development goals were the most significant variable predicting an adaptive pattern of strategy use and learning attitudes. As expected, performance-approach goals were associated with an adaptive pattern of strategy use. Also as expected, extrinsic work goals did not predict significantly the use of learning and self-regulatory strategies. As for social enhancement goals, the current results showed clearly that these goals were maladaptive to learning. Overall, these results confirmed the relative importance of adaptive goals such as mastery-development goals and performanceapproach goals on learning and attitudes among distance learners.
