Friday, October 21, 2005

Determining a sample of PhD candidates

While Kevin and I have conducted a national online survey of doctoral candidates recently (see earlier posts), I have also been gathering qualitative data from a series of in-depth interviews. During the early part of my candidature I thought long and hard about how to establish a structured sample of PhD candidates for possible interview. First, I decided to limit the population to full-time candidates enrolled at my university (Kevin is concerned primarily with part-time candidates). Second, I decided to reflect the diversity of candidates (rather than endeavour to establish a representative sample). Third, I decided to call for volunteers who would be interested in talking to me about their current doctoral practices and experiences.

A major reason for reflecting the diversity of candidates was the considerable variation that exists within the doctoral population in terms of gender, age, field of study, year of candidature, scholarship, ethnicity, citizenship, country of permanent residency, academic staff status, location, employment, dependants, disability ... to mention a few. The field of study in which candidates are officially enrolled seemed an obvious place to start. However, disciplines/research have been classified in many different ways, and typically involve complex coding systems (e.g. RFCD, DEST, SEP, ISI-ESI codes), which can include many hundreds of sub-categories. Interestingly, the identification of ‘multidisciplinary’ as an authentic, discrete category is relatively uncommon, with ISI-ESI and PhD Weblogs constituting notable exceptions.

I toyed with the idea of using the ‘hard/soft, pure/applied’ classification developed initially by Becher (1989) and extended by others, but decided to employ the DEST classification that comprises eleven 'broad fields of study'. One reason was a perceived degree of subjectivity associated with classifying fields of study within Becher's framework. Another was that Kevin and I had used the DEST classification for an item on the national survey, and I thought there might be potential for connecting aspects of qualitative and quantitative data sets.

The content and distribution of an email invitation to candidates willing to participate in a semi-structured interview of up to one hour’s duration was negotiated with the Graduate School in May 2005. The email generated 63 volunteers, one of whom withdrew subsequently. Additional demographic information from each respondent was requested in order to develop summary profiles. I then developed a ‘diversity grid’—containing around a dozen variables—that I used to systematically and rigorously select 10-15 candidates for possible interview, who could be seen to reflect diversity at my university in the context of the national PhD population. A major objective of the grid was to avoid any potential skewing of interviewees (e.g. in terms of field of study, gender, age etc).

Friday, October 07, 2005

Meeting with CIs, IPs and SPs

Face-to-face meetings are a regular feature of my life as a doctoral candidate. A couple of weeks ago, the Chief Investigators (CIs) and candidates (Kevin my PhD counterpart at Deakin University, Victoria and me) involved in the ARC project met to discuss the preliminary analysis of the ‘national online survey of the Australian doctoral experience’. Given the tyranny of distance (Canberra—Geelong) and budget constraints, we tend to meet once or twice a year in a lounge at Melbourne airport, with a view to maximising time and other resources.

Our meetings have been concerned primarily with aspects of the online survey—especially planning, implementation and follow-up. Our most recent meeting explored strategies to be employed in relation to the preliminary data analysis. We discussed sets of descriptive statistics, together with individual plans for further analysis that Kevin and I had circulated prior to the meeting. Outcomes included agreement on processes for establishing a common and agreed data file and issues around authorship for the publication of material.

I also meet with my Principal Supervisor for a couple of hours at least once a month. We usually meet in a campus café that serves a good brew and friand, where after a briefing on work I have completed to date, we discuss future directions, tasks and timelines. Although Kevin and I interact mainly by email, we also talk on the phone periodically if extended email exchanges are wearing our fingers to the bone.

I have two meetings scheduled for next week. One is a meeting of the Industry Partners (IPs) in the ARC project—three postgraduate student associations (CAPA, PARSA and DUSA). The logistics of this arrangement mean that one group will be in Melbourne (CAPA and DUSA representatives, Deakin CIs and Kevin), and linked by teleconference to another group in Canberra (PARSA representative, ANU CI and me). The purpose of the meeting is to brief IPs on the preliminary outcomes of the survey and to gain feedback and any additional input.

The other is a meeting of my Supervisory Panel (SP) that comprises two supervisors and three advisers. One of my advisers has a background in Sociology, while another would regard herself as multi-disciplinary, while two CIs and my Centre Director make up the group. Supervisory Panels are a requirement of doctoral study at the ANU, along with annual reports and annual plans that need to be formal approval. I have circulated an agenda and papers for the meeting which will run for a couple of hours. My major objectives include gaining feedback on a set of doctoral narratives that contain a synthesis of extensive interview data, as well as plans for further analysis.