Microsoft Word doctoral thesis final[1]. doc

4.4.3.5 Data collection 
 
The VConf-FGIs were conducted during March, April and May 2006.   
 
All the contributors in the VConf-FGIs received a personalized invitation to participate.  
In the case of South African universities, the researcher made a preliminary phone call to 
each selected institution. Faculty officers or managers from each faculty were requested 
to furnish the researcher with information regarding the presence of female HoDs within 
the faculty, together with the contact details of these HoDs. At a later stage, the 
researcher telephoned each HoD and briefly interviewed them to establish their suitability 
and willingness to participate in the study. Newly appointed HoDs were excluded as they 
would not have had sufficient experience as HoDs to provide rich data. Only those who 
had at least two years’ experience were eventually invited by email to take part in the 
VConf-FGI. A total of nine women participated in the VConf-FGIs. 
 
The first video conference focus group session involved three centres: University of 
Pretoria, University of the Western Cape and Bristol University. Four participants took 
part in the session.  The second session involved two centres : London Imperial and 
University of Cape Town. Two participants took part in the session. The third session 

---

 
125
also involved two campuses of the University of Fort Hare: Alice campus and East 
London campus.  Two participants were involved. The fourth and final session involved 
one centre: London King’s College. There was one participant at this session because her 
schedule made it impossible for her to participate in any of the other sessions. The 
process of setting the date involved first finding a week day when the researcher would 
also be free. Permission was then sought from each centre’s video conference facility 
manager for the use of the facilities. Once availability was established, this was 
communicated to the participants and an email reminder was sent two weeks prior to the 
session. One week before the session, the interview schedule was emailed to each 
participant so that she would familiarise herself with the topics to be discussed. This 
ensured that everybody was fully prepared for the interview. 
 
Prior to each interview session, a test call was scheduled to test the link between the 
centres. This is a vital exercise because it is at this time that any problems can be detected 
and corrected. The test dates were set and tentatively confirmed approximately  a week in 
advance  to allow for any necessary adjustments before final confirmation. 
 
All the interview sessions were conducted in the morning and lasted two hours. Each 
session began at the same time. So this necessitated synchronising South African time 
with UK time, and communicating the exact time of the link with everyone involved.  
The initial plan had been to conduct three VConf-FGI’s, but, as already indicated, it was 
not possible to accommodate one of the UK participants on any of the planned dates. A 
separate session had to be scheduled for her alone. Thus there were four sessions. 
 
Normally, a multi-conference unit requires participating centres to call through to the 
originating centre using the IP (Internet Protocol) number provided. However, as this 
focus group interview was for study purposes, the originating centre made the link to 
each participating centre’s ISDN (Integrated Services Digital Network) number. Once the 
link to each centre had been achieved, and the participants were all on the screen, the 
researcher began each  interview session by greeting  and welcoming everyone to the 
session. Brief introductions were made and the participants were given a chance to get to 
know each other a little before the purpose of the VConf-FGI was reiterated to the 

---

 
126
participants. The researcher also ensured that everyone had brought the interview guide 
along  for ease of reference. Because the time was limited to two hours, and there was 
more than one participant, it was more efficient to follow the interview guide so that 
everyone had a chance to say something on each topic. For the first few topics, 
participants shared their experiences in a round the table sort of way with the researcher 
facilitating the process. 
 
Later the procedure changed and participants were allowed to participate in any order, but 
still following the order of the topics in the interview schedule and ensuring everyone had 
a chance before the discussion moved to the next topic. The researcher’s role during the 
interviews was two-fold: to facilitate the session and ensure that minimum digression 
from the topic occurred, and to remark on issues raised and invite further comments and 
elaborations. The interviews were video-recorded from start to finish. In addition copious 
notes were taken. At the end of each interview, the videotape was played back to see how 
the interview had progressed, what data had been generated and to modify subsequent 
sessions. 
 
4.4.3.6 Data analysis 
 
Analysis began as soon as the first VConf- FGI was completed. To borrow Lee’s and 
Fielding’s (2004:533) words, the interview was mostly ‘topic oriented’ with the focus on 
identifying themes emerging from the data. An adaptation of Krueger’s and Casey’s 
(2000) transcript based data capturing and analysis procedure was followed  
(see  4.2.3.2).  At the end of each interview session, notes that had been taken during the 
interview were examined for key themes and these were written out according to topic, 
and filed. The videotape was played back to listen for, and to note, any patterns in the 
interview procedure and discussion which could be incorporated into subsequent 
sessions. At the end of all four interview sessions there were two data sets. A set of key 
points and a videotape of each session. The videotapes were manually transcribed 
verbatim, handwritten and filed according to session date. The transcription process took 
12 to 16 hours per two hours of videotape. 
 

---

 
127
When the manual transcription was complete, the handwritten transcripts were typed and 
merged into one document arranged according to topic, with the participants’ responses 
listed in order of their turns during the interview. 
 
This permitted easy access to all the participants’ contributions to a topic and facilitated 
identification of similarities and differences in their experiences of each phenomenon of 
interest. At this stage, gaps in some participants’ responses were noted. In cases where 
there were more than three gaps, especially of crucial information, the participant was 
immediately followed up with an email request to complete the information. All the