Date of Award


Document Type


Degree Name

Doctor of Philosophy


School of Education

First Advisor

Lionel Matthews

Second Advisor

Elvin Gabriel

Third Advisor

Jimmy Kijai



There is an enormous deficit in the literature on the factors associated with test anxiety generally, and test anxiety and music listening interest in particular, in the Caribbean region. This deficit is made even more distinct by the numerous studies that exist on test anxiety elsewhere and the effect it has on students’ behavior and performance in the classroom.


The purpose of this study was to examine the relationship between musical listening interest and test anxiety, while controlling for variables that are deemed plausible predictors of test anxiety. Predictor variables utilized include sociodemographic factors such as gender, age, marital status, parental status; socio-economic factors as income, and education, and socio-psychological factors which include parental nurturance, self-efficacy and religiosity.


This study employed ex post facto, correlational cross-sectional, research design in which two strategies were utilized to collect the data. An electronic format using Survey Monkey, an online research tool, as well as a hard copy format in which participants were asked to complete the questionnaire in their classes. The convenience sample, (N = 329) was drawn from both undergraduate and graduate students at a Christian university in Jamaica. Bivariate and Multiple regression techniques were used to gauge the relationships between the identified predictor and outcome variables.


Five null hypotheses were assessed by means of bivariate (BS) and multivariate (MS) statistical procedures in order to determine the relationship between test anxiety (examined as overall test anxiety (OTA) and the subscales ii psychological impairment (PI) and iii emotional impairment (EI) and the various measures of the predictor variables. Several relationships were thus identified between the predictor variables and the three measures of test anxiety/test anxiousness. At the bivariate level, marital status, parental status, father nurturance, rhythm and blues and religious music were all found to be significantly related to test anxiety. When t-tests were used to assess the relationship between overall test anxiety (OTA), and psychological impairment (PI), and emotional impairment (EI) and marital status, respondents who were married were found to be less test anxious than those who were not married: OTA (t = 2.20, p = .030); PI (t = 1.94, p = .055); EI (t = 2.48, p = .015). Similarly, when the three measures of test anxiousness were examined against parental status, participants with children were found to be less test anxious than those without children: OTA (t = -3.11, p = .002); PI (t = -3.25, p = .001); EI (t = -2.92, p = .004). The same three measures were assessed against father nurturance using Pearson r correlation. In each instance, statistically significant inverse relationships were found between each of the three measures when examined against father nurturance: OTA (r = - .186, p = .01); PI (r = -.175, p = .01); EI (r = -.183, p = .01). This suggests that the test anxiety levels of participants were lowered with increased levels of father nurturance. When the three measures were examined against rhythm and blues, only emotional impairment showed significance. Participants showed less test anxiousness with increased listening to rhythm and blues in one measure of test anxiousness namely emotional impairment (r = -.145, p = .012). However, religious music demonstrated positive relationships with two measures of test anxiousness PI (r = .148, p = .009); and EI (r = .129, p = .022). This indicates that the more participants listened to religious music the more test anxious they were. At the multivariate level, the three measures of test anxiety were each significantly related to only two predictor variables, religious music and father nurturance. In the case of psychological impairment religious music explained 19% of the variance (β = .191, p < 0.10). However, for emotional impairment 20% of the variance (β =.203, p < 0.10) was explained by religious music listening. On the other hand, when the three measures of test anxiety were regressed on father nurturance inverse relationships resulted. In the instance of overall test anxiety 26 % of the variance was explained by father nurturance (β = -.256, p < 0.01); for psychological impairment, 25% of the variance was explained (β = .254, p < 0.01), and for emotional impairment 24% was explained (β = .237, p < 0.01) by the same variable. It is useful to reiterate here that only father nurturance and religious music maintained a relationship with test anxiousness in the multivariate analysis.


Of the set of musical predictor variables employed in this study, 5 variables reached statistical significance with the three measures of test anxiousness at the bivariate level. These were marital status, parental status, father nurturance, rhythm and blues and religious music. However, at the multivariate level only 2 variables retained significance with the three measures of test anxiety. These were father nurturance and religious music. In terms of policy implication, schools, and education administrators may want to appropriate these findings selectively in addressing the issue of test anxiety.

Subject Area

Test anxiety--Jamaica, Students--Effect of music on, Students--Jamaica--Research