The under-representation and poor retention of women in computing courses at Victoria University is a concern that has continued to defy all attempts to resolve it. Despite a range of initiatives created to encourage participation and improve retention of females in the courses, the percentage of female enrolments has declined significantly in recent years, from 32% in 1994 to 18% in 2004, while attrition rates soared to 40% in 2003. A recent research study investigated these negative trends with respect to gender equity in computing courses: of interest was the possibility of gender bias in the learning environment and its impact on female attrition rates. Focus groups and surveys involving computing students of both genders were used as data collection tools in the study. The overall findings from the focus groups were rather surprising, as they yielded no strong indication of gender bias in the learning environment of the computing course; this applied to the logistical arrangements, academic staff, pedagogical methods, and course content. The thesis that the existence of gender bias in the learning environment contributes to high attrition rates of females in computing courses was not sufficiently supported. While the fact that students, both male and female, found their learning environment gender neutral was comforting, the realization that reasons other than gender bias drove females away from the computing course was not. High attrition rate of females remains the reality. Possible explanations of this phenomenon were suggested by the focus groups, and the search for confirmation of these indications and discovery of other contributing factors continued. The results of a subsequent survey confirmed several of the focus groups findings: the lack of gender balance in the course was considered immaterial, as was the gender of the lecturers and tutors; the course curriculum was deemed gender neutral; and, overall difficulty of the course and overall satisfaction with the course were rated similarly by both male and female students. Some of the survey results contradicted findings of the focus groups, notably the reported equal access to computers in laboratories and groupwork unaffected by the gender of group members. Moreover, although female students did not regard the underrepresentation of females in the course as a problem, they singled out their fellow female students as the vital source of both academic and personal help, thereby reiterating the need for female peers in the course. The first year, particularly the first semester, of the course emerged as the ` make or break' period especially for female students. Not only was it an important period with respect to adjustment to the course but also it was a period most likely to influence most female students' decisions about quitting the course. It appears that staff encouragement mattered little in dissuading female students from quitting, but the support of fellow students was important. Transition into the course, rather than gender bias, was identified as a possible factor contributing to high attrition rates of females in computing courses. Not only did the female students take considerably longer to settle into the course -up to six months, but they also were most likely to drop out of the course after the first year of study. This is of concern for two reasons: firstly, because of the large number of students contemplating dropping out; and secondly, because the role of the academic staff in persuading those students to stay was regarded as largely insignificant by the students. This study provides some explanation of the high attrition of females in computing courses, but the challenge still remains to develop effective strategies to halt this negative trend.
