National Taiwan Normal University
教育心理學報  回首頁
Apycom jQuery Menus
534 publication date:JUN, 2022
Confirmatory Research on the Chinese Version of the Ohio State Teacher Efficacy Scale for Elementary School Teachers
    Author:Chun-Wei Chen
Research Article

The theoretical backgrounds for the development of quantitative scales for measuring teacher self-efficacy include the locus of control theory (Rotter, 1966) and self-efficacy theory (Bandura, 1977). Although scales based on the locus of control theory were the first to emerge, these scales are now rarely used. By contrast, the scales developed on the basis of self-efficacy theory, including the teacher efficacy scale (TES; Gibson & Dembo, 1984) and the Ohio State Teacher Efficacy Scale (OSTES; Tschannen-Moran & Woolfolk Hoy, 2001), are commonly used (Chen & Wu, 2011). The short form OSTES comprises 12 items and exhibits excellent reliability and validity for evaluating elementary and middle school teachers. At the elementary and middle school level, increasing student engagement through instructional strategies and classroom management is a key goal for teachers (Chen & Wu, 2011). The research value of using the OSTES to evaluate teacher self-efficacy stems from its relevance to that goal. Related studies have employed the OSTES to measure the self-efficacy of elementary and middle school teachers in the United States (Harper et al., 2018; Woolfolk Hoy et al., 2008), Canada (Ross & Bruce, 2007), the United Kingdom (Woodcock et al., 2019), Scotland (Brady & Woolfson, 2008), Italy (Lent et al., 2011), Iran (Mohamadi & Asadzadeh, 2012; Rastegar & Memarpour, 2009), South Korea (Choi et al., 2019; Klassen et al., 2009), and Singapore (Chong et al., 2010; Klassen et al., 2009; Klassen et al., 2008). By using factor analysis, studies have mostly supported the viewpoint of Tschannen- Moran and Woolfolk Hoy, who have suggested dividing the OSTES into the three factors of instructional strategies, classroom management, and student engagement.

    Chen and Wu (2011) collaborated on the Chinese translation of the OSTES (C-OSTES) and recruited middle school teachers to be research participants to explore its psychometric properties. Their findings revealed satisfactory fit for the individual items in the C-OSTES. The factor structure of the C-OSTES is identical to that of the original OSTES, and the C-OSTES demonstrated excellent construct validity, discriminant validity, and reliability when used to measure the self-efficacy of middle school teachers. However, Cheung (2008) performed an exploratory factor analysis of the Hong Kong and Shanghai versions of the OSTES and revealed that the scales had a single-factor structure and a two-factor structure, respectively. Confirmatory factor analysis (CFA) indicated that the OSTES had poor fit in measuring the self-efficacy of teachers in The Republic of Cyprus and South Korea (Klassen et al., 2009). The C-OSTES exhibited adequate validity and reliability for measuring the self-efficacy of Taiwanese middle school teachers (Chen & Wu, 2011). The current study explored the psychometric properties of the C-OSTES for measuring the self-efficacy of Taiwanese elementary school teachers, the appropriateness of each C-OSTES item, and whether the C-OSTES has adequate validity and reliability.

    A total of 1,021 Taiwanese public elementary school teachers were recruited as research participants for this study. First, item analysis and assessment were performed on each C-OSTES item to determine its fit. CFA, Pearson product-movement correlation analysis, multivariate analysis of variance (MANOVA), and internal consistency analysis were subsequently performed to evaluate construct validity, criterion-related validity, discriminant validity, and reliability. In terms of construct validity, the CFA results indicated that the overall model fit, preliminary fit, and fit of the internal structure model of the three-factor correlated factor model were satisfactory, suggesting that the C-OSTES had adequate construct validity. In terms of criterion-related validity, the C-OSTES subscales and the overall scale exhibited positive correlations with three criteria, namely affective organizational commitment, job satisfaction, and organizational citizenship behaviors. This finding indicated that the C-OSTES had adequate criterion-related validity. This study applied the Fornell-Larcker criterion (Fornell & Larcker, 1981) to perform an evaluation of discriminant validity and determined that the C-OSTES exhibited satisfactory discriminant validity. We further employed MANOVA to inspect the differences in C-OSTES scores across elementary school teachers of varying education levels and teaching positions. Our results revealed significant differences in C-OSTES scores across education levels and teaching positions. In particular, teachers with a master’s degree or above scored higher in the instructional strategies and student engagement subscales than did those with a college education level or lower. Teachers holding a concurrent position as supervisor, director, or homeroom teacher scored higher in classroom management than did subject-area teachers. Teachers holding a concurrent position as supervisor or director scored higher in student engagement than did subject-area teachers and teachers holding a concurrent position as homeroom teacher. These results indicated that the C-OSTES exhibited adequate discriminant validity. Finally, in terms of reliability, Tschannen-Moran and Woolfolk Hoy (2001) recruited 255 US teachers as participants and performed reliability analysis of the OSTES. Their results revealed that the Cronbach’s α of the instructional strategy, classroom management, and student engagement subscales was .81– .86. Chen and Wu (2011) recruited 747 Taiwanese middle school teachers as research participants and analyzed the reliability of the C-OSTES; they determined that the Cronbach’s α of the three subscales was .86– .93 and that of the overall scale was .92. Our reliability analysis results indicated that the Cronbach’s α of the three subscales was .90– .93 and that of the overall scale was .94. Our obtained reliability coefficient values were higher than those of Tschannen-Moran and Woolfolk Hoy as well as Chen and Wu. In summary, the C-OSTES has adequate validity and reliability and is suitable for measuring the self-efficacy of elementary school students.

    Klassen et al. (2011) reviewed 218 empirical studies on teacher self-efficacy published between 1998 and 2009 and suggested using the OSTES (Tschannen-Moran & Woolfolk Hoy, 2001) to measure teacher self-efficacy. As also determined by Chen and Wu (2011), the present study concludes that the C-OSTES has suitable validity and reliability. Accordingly, future relevant studies should be conducted to verify the use of the C-OSTES for measuring self-efficacy in teachers. However, our CFA results on the use of the C-OSTES to evaluate Taiwanese elementary school teachers revealed that the three-factor correlated factor model exhibited favorable fit. This result differs from that of Cheung (2008), who revealed that the Hong Kong and Shanghai versions of the OSTES had a single-factor and two-factor structure, respectively. Taiwan, Hong Kong, and Shanghai use Chinese as their official language. However, Taiwan’s education system and policies are completely different from those of Hong Kong and Shanghai, which have similar education systems and policies but some distinct cultural traits. Future studies should conduct cross-national comparisons to explore the psychometric properties of the C-OSTES for measuring self-efficacy in Hong Kong and Shanghai teachers. These studies should examine whether the validity of the C-OSTES holds under different scenarios, and they should determine whether the validity of the C-OSTES can be generalized. In Taiwan, many tools have been developed to measure teacher (teaching) self-efficacy (Pan, 2005, 2007; Sun, 1995, 2001, 2003, 2005). Future studies should perform measurements and analyses using both the C-OSTES and other teacher (teaching) self-efficacy scales developed in Taiwan to verify their convergent validity. Additionally, studies on the long form OSTES, which comprises 24 items, may include localized items developed to measure teacher self-efficacy. In the process of creating a teacher self-efficacy scale, researchers may develop a new theoretical structure.

    With respect to the results of the validity analysis, the American Psychological Association (APA), American Educational Research Association (AERA), and the National Council on Measurement in Education (NCME) jointly published the Standards for Educational and Psychological Tests in 1974. According to these standards, validity was divided into criterion-related validity, content validity, and construct validity (American Psychological Association et al., 1974). In the subsequent three editions, the word “tests” in the title was changed to “testing.” In the 1985 edition of the Standards for Educational and Psychological Testing jointly published by the APA, AERA, and NCME, validity was divided into construct-related evidence, content-related evidence, and criterion-related evidence. However, in the 1999 edition, the validity classification system described in the 1985 edition was discontinued; instead, the 1999 edition discussed five types of validity evidence: evidence based on test content, evidence based on response processes, evidence based on internal structure, evidence based on relations to other variables, and evidence based on consequences of testing. The 2014 edition retained the four types of validity evidence and included an additional fifth type, namely evidence for validity and consequences of testing (American Educational Research Association et al., 1985, 1999, 2014). Most Taiwanese empirical studies that have developed, established, and verified scales have referenced the validity classification method proposed in the 1974 edition of the Standards for Educational and Psychological Tests when discussing their validity analysis results (Chen et al., 2020; Guo, 2021; Hsieh, 2020; Lee, 2020; Lee & Hong, 2021; Liu & Chen, 2020; Wang et al., 2019; Wang & Lee, 2020; Wu, 2020). However, future empirical studies aiming to develop, establish, and verify scales relevant to the field should reference the five types of validity evidence proposed in the 2014 edition of the Standards for Educational and Psychological Testing (jointly published by the APA, AERA, and NCME) to explain their validity analysis results.


下載


關鍵詞: Chinese version of the Ohio State Teacher Efficacy Scale (C-OSTES), teacher self-efficacy, confirmatory factor analysis


Master’s Study Duration: The Effects of Active Learning Based on the Belief-Action-Outcome Model

Copyright © 2022 Bulletin of Educational Psychology
Address: No. 162 Hoping E. Rd. Sec. 1, Taipei 10610, Taiwan, R.O.C.
Tel: (02) 77493757, Fax: (02) 23413865, Email: t05002@ntnu.edu.tw
All rights reserved. BEISU Design