Use this URL to cite or link to this record in EThOS:
Title: Non-mathematics majors studying statistics at universities in Cyprus : factors behind performance
Author: Michailidou, Theognosia
ISNI:       0000 0004 7655 0981
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Statistics is increasingly taught as a part of the curriculum programmes of a wide spectrum of disciplines at higher-level institutions. The main goal of this doctoral study is to get an understanding of non-mathematician students' perceptions, challenges and experiences when undertaking a university level introductory statistics course. This research also seeks to investigate affective, motivational and cognitive factors, which may be associated with students' academic performance in statistics. A mixed-methods research design (that is a combination of quantitative and qualitative data collection methods) was employed. A self-reported questionnaire was administered to a larger sample of students (over five hundred) near the beginning and towards the end of the instruction of a statistics course. A sub-sample of the students was interviewed and thirty of them were selected to act as a source of qualitative evidence to complement the quantitative results. This study reports on data from a sample of undergraduate majors who attended statistics courses from recognised (both public and private) universities in Cyprus across two academic semesters. Students with a variety of mathematics backgrounds and experiences and from diverse academic departments and degree programmes participated in the study. Quantitative data analyses (including multivariate analyses such as structural equation modelling techniques) and interview data analyses (using thematic analysis) were performed. The qualitative findings highlight, amongst other things, the importance of the role of the instructor in the statistics education learning process. The key finding of the structural equation modelling techniques, when modelling performance in statistics, is that self-efficacy and resilience in both questionnaire administrations explain and predict statistics performance over and above the other variables (such as liking, interest, value, difficulty and anxiety) included in the model. More specifically, self-efficacy and resilience are found to be directly and positively related to the performance. Self-efficacy has a prominent position in the model since it is also found to be associated with all the variables incorporated into the model. It is suggested that self-efficacy and resilience, particular in the context of statistics education, are constructs worthy of further investigation from researchers and educators. The potential contribution of the study is to benefit the development of statistics education and offer implications and recommendations for teaching and learning statistics based on the findings.
Supervisor: Homer, Matt ; Monaghan, John Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available