The components, construction and correlates of quality of school life in secondary education

Abstract

The present thesis is an investigation of the components, constmction and correlates of Quality of School Life (Q.S.L.). Firstly, a new Q.S.L. scale was constmcted and its psychometric properties tested. Secondly, the association between demographics (gender, age / school grade, school), school stress, well - being and personality (self - esteem, affectivity, locus of control) and Q.S.L. was studied. Cross - cultural comparison between Scotland and Greece regarding levels of Q.S.L., as well as correlates and predictors of Q.S.L. across cultures, were also investigated. Finally, the association between Q.S.L. and school performance, school misbehaviour, bullying / victimisation and substance use, in comparison with school, well - being and personality factors, was studied. Participants consisted of a sample from two secondary schools in Stirling area, in Scotland (n = 425) and a sample of Greek secondary school pupils from Agrinio region (n = 173). Data were collected by means of questionnaires. Eleven scales were administered to the Scottish sample (e.g. demographics, affectivity) and six scales (e.g. demographics, school stress) were administered to the Greek sample. Univariate parametric (e.g. t - test, ANOVA) and non - parametric tests (e.g. chi - square) were used in data analysis. Mutli - variate tests (e.g. Logistic regression) were also used. Seven cross - sectional studies are reported. Results indicated that the new Q.S.L. scale has good psychometric qualities both in the Scottish and the Greek sample, although such qualities need to be investigated further. Personality factors were found to be the best predictors of Q.S.L. in two studies. In the study regarding the correlates of Q.S.L., it was found that school self - esteem was the best predictor of Q.S.L., whereas in the cross - cultural study between Scotland and Greece it was found that, for both Scottish and Greek samples, positive affectivity was the best predictor of Q.S.L. However, both Greek and Scottish samples in the cross - cultural study consisted of 4'*' to 6“* graders, whereas the sample in the study regarding the correlates of Q.S.L. (Scottish only) consisted of L* to 6*** grade secondary school pupils. Greek and Scottish pupils were found to differ in relation to Q.S.L. levels total and across domains with a privilege of Scottish pupils regarding Q.S.L. Q.S.L. was not found to be the best predictor of self - rated performance neither across subjects nor overall. School self - esteem was found the best predictor of self- rated performance overall. However, Q.S.L. was found to be associated with school misbehaviour, but again it was not its best predictor. The best predictor of school misbehaviour was found to be gender, with males being more likely to misbehave than females. Nevertheless, Q.S.L. was found the best predictor of overall involvement in bullying and / or victimisation, alongside with school stress, implying its significant association with the phenomenon as a whole. It was also found that peer self - esteem and demographics, such as gender, differentiate bullies and victims. Finally, Q.S.L. was found to predict at best smoking maintenance, whereas other factors (e.g. school stress) where found to predict at best experimentation with smoking, alcohol and illicit drugs and maintenance of alcohol use. The role of Q.S.L. as well as of demographic, school stress, well - being and personality factors in relation to school performance, school misbehaviour, bullying and substance use are discussed. The results of each study are discussed in relation to previous relevant literature, practical implications for each area, limitations of the research, and some suggestions for future research are also provided.