This paper presents explanatory and predictive models of student failure and subsequent attrition within
a first year university Computing Science course, based on data collated from a number of sources.
Previous research from the field of student attrition and failure is reviewed. Such research highlights
the need for data analysis at the departmental level, where the university experience for a student is
primarily shaped. Data was gathered through the use of an online questionnaire, weekly interviews and
a post-exam paper questionnaire. Additional data from the Department of Computing Science
regarding student attendance, assessment and entry scores is also analysed. A large number of variables
were gained, most of which were not significantly related to performance in exams sat mid-way
through the academic year. Notably, attendance and amount of revision did not elicit high correlations
with exam score, and it is suggested that understanding of the course material is a key factor. It is
concluded that the factors examined within this study are not sufficient to predict student performance,
although several areas are recommended for future research, including student self-assessment of
understanding and commitment, previous computing experience and temporal patterns in student
behaviour.