Prejudice against older teachers: theoretical and methodological inquiries
Knight, Alec (2009) Prejudice against older teachers: theoretical and methodological inquiries. PhD thesis, University of Nottingham.
Research on prejudice against older workers often tacitly subscribes to outdated attitude theory and rarely attempts to examine methodological questions about the effect of measurement strategy on results. This thesis compared empirically the validity of the tripartite and unitary models of attitudes in relation to prejudice towards (a) older teachers, and (b) the employment of older teachers. In addition, two alternative measurement strategies for assessing stereotypes of older teachers were compared (a 15-item rating scale versus a free-response measure). A postal questionnaire survey was conducted on a random sample of members of the National Union of Teachers (n=285) in May 2008. Two equivalent questionnaire versions were constructed that differed only with respect to the measurement strategy for assessing stereotypical beliefs of older teachers (aged 50+). The validity of the tripartite theory of attitudes was tested using hierarchical regression analyses that examined the ability to predict reported attitudes towards older teachers and the employment of older teacher on the basis of (a) stereotypical beliefs of the target group alone, and (b) a combination of stereotypical beliefs, and affective and behavioural associates of the target group. Prediction of attitudes was significantly improved above the level afforded by stereotypical beliefs alone by adding both affective and behavioural information to the regression model, supporting the validity of the tripartite model of attitudes. Measurement strategy was found to have a significant effect on the positivity of stereotypical beliefs elicited (r = -0.515), with the average response valence on the stereotypical beliefs rating scale being significantly less positive than the average response valence on the free-response measure. The content of the rating-scale measure was also found not to be representative of naturally elicited stereotype categories. The theoretical and methodological implications for attitude research in organisations were discussed.
Archive Staff Only: item control page