Design and Validation of ELGOS, ESES and LLSS
This section discusses the content and sources of ELGOS, ESES, and LLSS, as well as the validity and reliability of these scales. To guarantee the reliability and validity of these scales, a preliminary study was conducted to develop and explore the structures of these scales. Items were selected according to the principle of comprehensiveness and economy with acceptable reliability and validity. In the preliminary study, 252 college students from three universities were involved. The number of valid questionnaire was 230. Among the 230 subjects, 92 were females and 138 males. The data obtained were then examined for the reliability and validity of the instruments. Cronbach’s alpha was used to indicate the reliability of the measures, and exploratory-factor analysis was employed to analyze the construct validity.
English-Learning Goal Orientations Scale
A scale measuring English-learning goal orientations was designed especially for the present study. Based on the relevant literature and guided by the theories of academic goals, the items concerning English-learning goal orientations were mainly developed based on The Patterns of Adaptive Learning Survey (Midgley et al., 1998, 2000) and Motivated Strategies for Learning Questionnaire (MSLQ) (Pintrich, Smith, Garcia, & McKeachie, 1991) according to the characteristics of English course.
ELGOS includes 15 items. In the preliminary study, an exploratory factor analysis was performed to explore the structure of ELGOS. Three factors were identified. Items in Factor 1(items 2, 7, 10, 14, 18) were related to the goals of acquiring knowledge and increasing one’s competence. Factor 1 was labeled mastery goals. Factor 2 (items 3, 4, 8, 11, 15) represented the goals of demonstrating one’s competence and was named performance-approach goals. Factor 3 received loadings from five items (items 5, 12, 16, 19, 21) concerning the goals of avoiding demonstrating incompetence and negative judgments and was labeled performance- avoidance goals.
The reliabilities of ELGOS were also examined. The Cronbach’s alpha for the total scale was .86, for the first factor (i.e., mastery goals) .84, for the second factor (i.e., performance-approach goals) .84, and for the third factor (i.e., performance-avoidance goals) .83. The results indicate that the internal consistency of ELGOS in the pilot study was highly acceptable.