Computer science: through the eyes of potential students

Em+ student dropouts ia jirst year computer science courses are ~pically high Many reasom are proposed such as the inherent demands of the disa”pline, or that many students believe that computer science enhances general career prospects (leading many to enrol without personal iMerest in the subject matter). There are numerous possibilitiar, all possessing cm intuitively appeakg basis to them kl’oweve~ ajler some years of hearing such possibilities, it occurs that they hinge on a more j%damen!al question“ISit the case that many students who enrol for a jirst computer science course h so with some ve~ limiting nzisconcepn”ons of what the discipline entails?” This paper summarises a pilot study that was motivated by Ws questiorL L The Research Methodology On the sptitic issues of interes~ the literature appears to have very Iitde to offer on which to develop a research response. The research effort was therefore cast as a pilot study, seeking in part to highlight future research directions. Initially motivated by the possibfity that students enrolling for computer science courses may not realise the very nature of the disciplin% the research domain widened to consider the convers% namely that students might also not be enrolling in computer science courses due to similar misconceptions. The nature of the study, then, became an exploration of student preconceptions of what it means to study computer science. 1.1 Sampling Procedures It was assumed that the best place to test the preconceptions of potential students was before their enrolment in a C5mputer science course. The actual point of enrolmenL even before exposure to any of the course contenL would be too late to detect the presence of attitudes which may lead to students not considering participating in such a course. The sampling was therefore directed at senior secondmy students. Additionally, the research was focussed on schools in New South Wales as a matter of convenience. It was further assumed that the sample populations of greatest interest would be located in secondary-level computing courses, in that such students were already indicating a level of interest in technology which would help to identi& them as potential tertiary students in computing. In New South Wales, this would mean that students would be doing one of the Computing Studies subjects. Permission to m:ke digital]hard copy of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed far profit or commercial advantage, the copyright notice, the title of the publication :nd its date appear, and notice is given that copying is by permission of the ACM, Inc. To copy otherwise, to republish, to post on servers or to redistribute to lists, requir= prior specific permission and/or a fee. ACSE9S, Brisbane QLD AUSTRALIA. @lW, ACM l-5 SllS-018-X/9S/0007 $5.oo 145 Eleven NSW high schools offering Computing Studies courses were identified by a simple World-Wide Web search. The schools were contacted by email using the addresses provided on their home pages. Three schools accepted the invitation to participate in the study, and the requested number of survey forms were sent to each. Two of those schools St Pauls Grammar at Penrith and Loreto at Normanhurst returned completed surveys for a combined total of S4 students from years 11 and 12. The purposive nature of the sampling led to a sample which could not be claimed to represent the general high school population. However, this could be considered favorable to the nature of the research. For example, 92.9 percent of students reported home access to a computer, a very high ratio that one would not expect to be generally the case. Yet to the extent that factors such as these serve to more strongly identi& potential students of computing at tertiary levels, it is assumed that misconceptions about such courses in the sample have a high probability of being more widely distributed in the population. One factor that is significantly different in the sample compared to university computing courses is the gender ratio. Females form 71.4 percent of the sample, compared to around 28 percent for our current first year course (a figure that is comparable with that of other institutions). The effect of gender is excluded from the current study, as it was felt that the sample of 24 males would not provide reliability. However, lack of gender representation in computer science is an important issue, and if the study produces an undetected bias towards misconceptions specific to females then the results will still produce important considerations for overall student dropouts and enrolment patterns in the subject.