TPC-Journal-V4-Issue5

The Professional Counselor \Volume 4, Issue 5 549 Next, the three SESSS subscales correlated with the SESRL, which measures students’ beliefs in their capability to engage in common effective self-regulation learning strategies. Again, all three SESSS subscales followed the same pattern, and had a considerable drop in magnitude of correlation with the MSLQ Self- Efficacy subscale, which measures students’ reports of general academic self-efficacy. While smaller, the correlations with the Self-Efficacy subscale (which ranged between .12 and .28) were practically different than 0. The Self-Direction of Learning subscale had the strongest correlations with the other measures, followed by the Support of Classmates’ Learning subscale and finally by the Self-Regulation of Arousal subscale. There is evidence that the three SESSS subscales, the Cognitive Strategy Use subscale and the Self-Regulation subscale of the MSLQ, and the SESRL all measure some common dimension of an underlying construct, probably relating to students’ intentional use of strategies to promote effective learning. The Self-Regulation of Arousal subscale was the only one of the three SESSS subscales to be significantly correlated with the MSLQ Test Anxiety subscale. Conceptually, students’ abilities to self-regulate arousal should be related to their reported levels of test anxiety. However, given the weak relationship ( r = 0.08) and the fact that this observed correlation was actually opposite in direction to the relationship that would be expected, this finding should not be given undue weight. The results of this study offer little to no evidence that the three SESSS subscales measure the same construct as the MSLQ Test Anxiety subscale. This weak or nonexistent relationship is discriminant evidence in that the other SESSS subscales did not correlate strongly or at all with the MSLQ Test Anxiety subscale. However, the three other MSLQ subscales and the SESRL all showed moderate, negative correlations with the MSLQ Test Anxiety subscale. These differing patterns of correlation with MSLQ Test Anxiety suggest that the SESSS subscales capture a different dimension than the common dimension of the underlying construct measured by the MSLQ subscales (including the Test Anxiety subscale) and the SESRL . Discussion The observed pattern of results is very useful in determining the validity of inferences that currently can be made from SESSS scores. Based on prior factor analytic studies (Brigman et al., 2014; Carey et al., 2013), the present authors made an attempt to create three SESSS subscales based on the items that load most strongly on each of three underlying factors. These three SESSS subscales showed good internal consistency and moderate intercorrelations suggesting that the three subscales most probably measure related but discriminable dimensions of students’ success skill use. However, the three SESSS subscales showed essentially the same pattern of correlation with comparison scales . Similarly, the pattern of results suggests that there is little or no overlap between the construct measured by the three SESSS subscales and the construct measured by the Test Anxiety subscale of the MSLQ. Unfortunately, these results do not shed much light on any differences among the SESSS subscales. Each subscale showed essentially the same pattern of correlations with the comparison scales even where differences would have been expected. For example, the authors would have expected the SESSS Self-Regulation of Arousal subscale to correlate significantly with the MSLQ Test Anxiety subscale, since individuals who are better able to regulate their levels of emotional arousal would be expected to experience less specific anxiety. These results suggest that the SESSS as a whole represents a valid measure of students’ intentional use of strategies to promote academic success. While prior factor analytic studies (Brigman et al., 2014; Carey et al., 2013) have suggested that the SESSS has three related dimensions, making inferences based upon the three SESSS subscales related to these dimensions is not warranted. Instead, until evidence can be found that these three subscales measure discriminable dimensions of success skill use, the SESSS should be used as a unitary measure in research and practice. Future research in this area is necessary.