chapter2

Most social scientists seem to agree that in principle there is no one data collecting procedure that is inherently right or best for collecting data about social or educational issues. The writings of social scientists and educators usually present a list of basic criteria which can be used to weigh the relative merits of the different research methods.

Research experience suggests that the methodology should be treated as a process by which a researcher endeavours to generate an abstract view of the situation. Methodology involves the processes of observation, selection, interpretation and abstraction. A final report should specify the processes by which the evaluative data was gathered, so that the reader will not have to base their interpretations of descriptions on a series of 'taken for granted' assumptions which are supposed to imply meanings and assumptions. This is particularly important in the area of cross-cultural comparisons.

It is recognised that conventional research methodology treats subjects' linguistic typifications, some of which are needed in handling questionnaires, as if they are both obvious and unequivocal responses to stimuli which are themselves unambiguous. It is acknowledged that research instruments designed for use in a particular study always seem to reflect a balance between the ideals of the researcher, the statistical design restrictions and the problems of doing research in a particular setting. It is within this context and these caveats that this research project is set.

Despite the misgivings of the researchers and members of the Advisory Group it was decided that, given the constraints of personnel, time and the need to establish some definitive statements on the performance of indigenous students at the University of South Australia, a questionnaire would be used to focus and structure the research. Open ended questions were to be used wherever possible. These were to be supplemented by in-depth open-ended interviews, group discussions and telephone interviews, with a small representative cohort drawn from the original sample. The project aimed to obtain qualitative information through interviews of successful and incomplete students.

The principal researcher prepared statements about the research study. The statements identified the researchers, the sponsor and the study objectives. These statements stressed confidentiality and respect for the integrity of the students and the staff involved in the programs. A pilot study involving a small group of students was carried out to ascertain the validity of the approach.

The project was submitted to the University of South Australia Research Ethics Committee for clearance (see Appendix 2).

For this study it was possible to identify all present and past Aboriginal and Torres Strait Islander students at the University of South Australia, and also non-indigenous students who had undertaken studies in the Faculty of Aboriginal and Islander Studies. Despite this, it was recognised that it was not possible to contact all the students so identified. The student sample was stratified according to variables such as ethnicity, age, level of education upon entry, completion/non-completion, mode of delivery, marital status and gender to build a matrix so that obvious biases by interviewee classification in the sample could be countered. The sample of non-Aboriginal students of the Faculty of Aboriginal and Islander Studies served as a control group.

Field Work Techniques

Interviews with students were conducted between May and December 1995. While face to face interviews were preferred, telephone interviews were also conducted when there was no other option. Individual interviews were supplemented, in some instances, by small group discussions.

Both subjective and non-subjective factors were identified and operationalised in the form of questions included in the interview schedule. The nature and rationale for these questions are examined in Appendix 2 of this report.

The contact phone lists for participants consisted of three lists that were generated from the Registry office of the University of South Australia. The lists contained the permanent and local phone numbers and last known addresses of all Aboriginal and Torres Strait Islander students who were listed as drop-outs from 1991 to 1994; all indigenous students currently enrolled (1995) in the second and third year of their courses; and all non-indigenous Faculty of Aboriginal and Islander Studies students including those who dropped out from 1991 to 1994. The students were listed alphabetically with gender, age and ethnicity provided, as well as the reason for leaving for those who had withdrawn. In addition students who had completed their courses within the specified time frame were identified. The researchers decided to include indigenous and non-indigenous graduates in order to ascertain whether the problems identified by the indigenous students who had withdrawn from their courses at university were unique to this group or were common to all students.

The first difficulty encountered in the field work was gaining access to the students on the lists provided by Registry. The majority of students were no longer at the phone numbers provided and numerous calls were made in an endeavour to locate the current phone numbers and addresses. Many potential participants could not be traced.

As mentioned earlier, there were also problems using university statistics, one of the problems being the way the data was generated in response to specific questions. Students who gained credit for previous studies and finished their

degrees within a year, for example, were listed as drop-outs because they did not re-enrol the following year. This was made clear when the first student contacted had not in fact dropped out, but completed the degree in 1993.

Potential participants were phoned from Adelaide in order to ascertain current whereabouts and to gauge the number and type of participants available. The research project was discussed over the phone, with an explanation that the phone numbers had been provided by the University Registry. The potential participants were asked if they would be interested in completing an interview and questionnaire in the next month to indicate what they liked and disliked about the university and their course of study. The students were informed that they could stop the interview at any time without penalty, especially for those still attending. They were assured that confidentiality would be respected and that no names or identity would be used in any publication. It was explained that the goals of the study were to ascertain the views of past and present students with respect to the strengths and weaknesses of the university program in which they had been involved-what they liked and disliked about it-and to use the results to improve university policy and practice. In addition, recommendations based on the results would be made with the aim of reducing the Aboriginal and Torres Strait Islander attrition rates in the future. The researcher stressed the importance of getting information directly from the students' perspective. Some past students declined to be involved because of present work/study commitments or other reasons.

The intending participants were contacted by phone a second time to schedule the interview. The interviews were conducted wherever was most convenient for the participants. Mostly they occurred at the place of work, home, in a coffee shop or at the University of South Australia. Informants were again told about the project when they were met in person. They were also given the Research Information Sheet to read. When they had read the sheet they were asked to sign the consent form. The Research Information Sheet was left with the informants with the contact numbers provided.

The interviewer read through the multiple choice questions and made sure the Likert scale was understood as it applied to each question (1 = most important or strongly agree down to 4/5 = not a consideration or strongly disagree). This clarification together with the interactive nature of the interview helped to ensure that all questions were answered. It also provided a way to discuss each question and ask probe questions where appropriate. The informants were asked their opinions about the open ended questions and where the participant was agreeable the researcher wrote down the responses while repeating what was written to the informant. In other instances, the participants chose to complete the open ended questions without the assistance of the researcher.

The interviews averaged an hour in length and after the questionnaire was completed, the informants were urged to contact the researcher or any other phone numbers on the Research Information Sheet if they had any questions in the future.

In a few instances, face to face interview was not possible due to the participant living out of town or interstate, or because of the participants' current work, social or family commitments. At their request the questionnaire and associated papers were mailed to them for their perusal. For these participants, the initial phone contact was as described above.

In other instances, where problems were encountered in contacting 'out of town' students, faculty staff members in the regional study centres acted as facilitators to ascertain whether the students attending the study centre were willing to be involved in the study.

With these prospective participants and in the other instances where face to face interviews were not possible, the purpose of the study was explained at the first contact and the questionnaire discussed to ensure that the respondent understood how the Likert scale was used.

The questionnaire, information sheet, consent form and return stamped addressed envelope were then sent to the participant with a request that they should phone the research assistant (reverse charges) when they were ready to complete the questionnaire. Many participants availed themselves of this offer and the questionnaires were completed by phone. For those who did not and completed the questionnaire themselves, follow up phone calls were made by the researcher in order to clarify any ambiguities in the responses or to rectify incomplete information. In a few instances, participants were reluctant to divulge date or place of birth; the researcher respected the individual's right not to disclose such personal details.

In general participants were eager to provide information, but were cautious in the beginning of the interview because the questionnaires were not anonymous in that names were attached for administration purposes, even though anonymity was guaranteed. The interviewer attempted to make the participants feel comfortable and stressed that the information was confidential and it was important to be truthful so that accurate recommendations could be provided from the results. The participants were interested in the study and expressed a clear desire to help with providing information.