Study of the motivation of teachers in Hong Kong

No other intervention can make the difference a skilful teacher can make in learning. A teacher may account for a variation of up to 30% in student achievement. However, in most western countries, teaching is in crisis, with many studies indicating high attrition among teachers. Therefore, there is an urgent need for further research to identify the attributes that make happy, committed teachers and enable informed interventions that can reverse this trend.

Unlike western countries, the Hong Kong school context is characterised by strong inducements with high salaries, high social obligations and high academic achievement, as evidenced by its top five ranking in the Programme for International Student Assessment (PISA). Consequently, an analysis of the attributes of Hong Kong teachers as predictors of desirable workplace outcomes offers a valuable opportunity to gain new insights for use in other countries.

This thesis aimed to identify and develop a set of motivational attributes for teachers that can predict positive outcomes for job satisfaction, job self-concept and psychological well-being. These predictive attributes can guide further research, facilitate professional development, inform policymaking and guide interventions for the ongoing motivation of teachers.

The thesis consists of three interrelated studies. Study 1 established a psychometrically sound set of attributes (Education motivation attribute profile [EdMAP]). Study 2 analysed the predictive relationship between EdMAP and the outcome measurements of job satisfaction, job self-concept and personal well-being at work, thus establishing the practical value of EdMAP attributes for predicting the desired outcomes. Study 3 used latent profile analysis to unearth any identifiable, practical and theoretically relevant latent profiles or profile-based subgroups within the teacher population that can guide professional development and policymaking.

A sample of teachers from Hong Kong was administered an adapted version (for teachers) of the employee motivation attribute profile (EMAP) survey developed by Marsh et al. (1991). The instrument measured a broad range of motivation-related attributes using five items per attribute. The total useable sample was 896, with 34% males and 66% females.

Study 1 found that each of the attributes used in the investigation was a valid and robust measure of the construct being measured. The data provided an excellent fit to the hypothesised a-priori factor structure (configuration of factor loadings, variances and covariance). Confirmatory factor analysis yielded an acceptable fit. An exploratory structural equation model with all 23 factors concurrently evaluated and loaded into the non-target factors yielded a comparative fit index/Tucker-Lewis index of 0.981/0.969. The results supported a previously hypothesised set of second-order factors consisting of global-leadership, global-goal orientation, global-variety, global-abstract thinking and global-interpersonal formulated by Marsh and McInerney (1991) with varying degrees of model fit.

Study 2 found a predictive relationship between EdMAP attributes and the measured outcomes of job satisfaction, job self-concept and personal well-being at work. Considering the multicollinearity among the factors, the least absolute shrinkage and selection operator (LASSO) regression technique identified a subset of factors that best predicted the outcomes. With β coefficients ranging up to 0.45, decisiveness, leadership, application and emotional control were found to predict most outcomes.

Study 3 found that while most of the teachers were profiled based on the level of their attributes, there was a distinct group with differences in the profile. The four-class model unearthed a group that had very low values for routine and consultative behaviour and high values for variety, emotional control and career orientation. While the covariates of gender and grades taught did not have any effect on the probability of being in a specific class, the study found age-related changes in class probability. Further investigation is required to establish the significance of these changes.

In conclusion, this thesis found that the EdMAP instrument is a psychometrically sound tool, confirming that the results from the Hong Kong sample are suitable to be baselined and used for future investigations into the motivation of teachers and other professions. The results of Study 2 indicated that predictive attributes such as leadership, emotional control and sociability should be targeted in professional development. Policies and interventions to increase leadership opportunities and create environments that encourage behavioural flexibility and group sociability are predicted to lead to higher job self-concept and psychological well-being. Study 3 indicated that there were subgroups within the teacher population that need to be targeted with specific interventions such as being change agents and early adopters.