Measuring teacher job satisfaction: Assessing invariance in the Teacher Job Satisfaction Scale (TJSS) across six countries

Abstract

Work and organizational psychology has long been concerned with measuring job satisfaction in organizational contexts, and this has carried across to the field of education, leading to a research focus on the work-related satisfaction of teachers. Today, a myriad of organizations continue to assess employees’ job satisfaction on a routine basis (Liu, Borg, & Spector, 2004). Unfortunately, a sort of balkanization of the field has resulted in the production of dozens of specific measurement tools, making it difficult to cross-compare samples and contexts. The present paper tested the measurement invariance of the Teacher Job Satisfaction Scale (TJSS) in six international cohorts (Netherlands, United States, Russia China, Italy and Palestine) of in-service teachers (N = 2,819). Confirmatory factor analysis and multi-group invariance tests were applied. The TJSS-9 displayed robust psychometric proprieties and no substantial departures from measurement invariance (configural and metric). Future research is required to further test equivalence across additional countries, with view to developing a truly international tool for measuring job satisfaction in teaching.

Agency via Life Satisfaction as a Protective Factor From Cumulative Trauma and Emotional Distress Among Bedouin Children in Palestine

“…With regard to internal consistency, the estimated values were generally satisfactory, except for two dimensions of life satisfaction (satisfaction with friends and satisfaction with the environment). This discrepancy may have been due to the fact that in measuring different domains of life satisfaction, the assumption of tau-equivalence (i.e., all factor loadings are the same; Graham, 2006; Pepe et al, 2017) was violated. In addition, large standardized beta weights may be an artifact of common method variance (CMV; i.e., variance that is attributable to the measurement method rather than to the constructs themselves, see Podsakoff et al, 2003).…”

Emotion Knowledge, Theory of Mind, and Language in Young Children: Testing a Comprehensive Conceptual Model

“…As is standard when conducting SEM (Kline, 2011; Pepe et al, 2017; Veronese and Pepe, 2018), the model’s goodness of fit with the data was assessed by calculating a set of indexes. These measured the overlap between the observed (S) and reproduced (Σ) matrices of covariance.…”

The contribution of emotion knowledge, language ability, and maternal emotion socialization style to explaining toddlers’ emotion regulation

“…Although this value was below the recommended threshold (although rejection of the dataset is not mandatory even in this case; see Schmitt, ), analysis of the composite reliability coefficient (CR; Raykov, ) prompts a different conclusion: The actual CR value (0.76) supported acceptance of the AKT scores. Plausibly, the discrepancy between the two indicators may have been due to violation of the assumption of tau‐equivalence (i.e., the assumption that all factor loadings are the same; Graham, ; Pepe, Addimando, & Veronese, ). Furthermore, a congeneric model with no error co‐variances can underestimate internal reliability (Novick & Lewis, ), even at the population level.…”