Adaptation of Perceived Self-Regulatory Efficacy Scale for Writing into Turkish Language

Abstract

This research aimed to adapt perceived self-regulatory efficacy scale for writing to Turkish language. The data for factor analysis and reliability was obtained from a total of 450 undergraduate students (308 male, 142 female, age mean=20.82, SD=1.43) studying at a public university in Turkey's Mugla province. Additionally, for the criterion validity of the scale, another data was gathered from a total of 250 undergraduate students (154 male, 76 female age mean=20.89, SD=1.55). The scale, which is consisted of 25 items, revealed one-factor model and this factor explained 38% variance. Factor loading of the scale ranged from .18 to .74. Item-total correlations were from .16 to .69. Beside this, the internal consistency (Cronbach's Alpha=α) of the scale was .93 from the first sample and it was .90 from the second sample. The item analysis of 27% upper and lower group means calculation yielded a statistically significant "t value". In regard to the criterion validity of the scale, there was positive and significant relation between the adapted scale and academic self-efficacy scale.© 2015 IOJES. All rights reserved Keywords:1 Self-regulation, validity, reliability, efficacy Extended Summary PurposeThe aim of this study is to adapt the scale named perceived self-regulatory efficacy for writing which was developed by Zimmerman and Bandura (1994) to Turkish culture. MethodThe data for factor analysis and reliability was obtained from a total of 450 undergraduate students (308 male, 142 female, age M=20.82, SD=1.43) studying at a public university in Turkey's Mugla province. For the test-retest reliability of the scale, the data was collected from 50 undergraduate students and for the language validity of the scale, the data was obtained from 36 undergraduate students. On the other hand, for the criterion validity of the scale, another set of data was collected from a total of 250 undergraduate students (154 male, 76 female age M=20. 89, SD=1.55). When the sample groups were formed, convenience sampling method was utilized. Perceived self-regulatory efficacy for writing and academic self-efficacy scales were used to collect the data.To test the construct validity of the scale, exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were employed. While internal consistency coefficient (Cronbach's Alpha α), test-retest correlation coefficient, Spearmen-Brown Split-Half correlation coefficient were used for the reliability of the scale, independent samples t-test was used to test the mean differences in the 27% upper and lower groups. In addition, sampling adequacy and Barlett Sphericity tests were used to discover whether the sampling was