Microsoft Word 2007 ichrie conference Proceedings Final-Final 06-06-07. doc

Procedure 
Focus group questions were standardized and open-ended, designed to investigate the participant’s point of 
view “without predetermining those points of view through prior selection of questionnaire categories” (Patton, 
1990, p.24). The focus group format was semi-structured, allowing for the use of follow-up probes (Rynes et al., 
1991) when necessary. 
The focus groups were conducted at the participants’ retirement community, with the organization of the 
meetings facilitated by the village manager. Participants were recruited via recruitment fliers posted on each 
community’s bulletin board. Participation was voluntary and all responses were confidential. Focus groups lasted 
between 60 and 90 minutes.
The research team included the first author and two research assistants, with the first author conducting the 
focus groups. After a briefing about the purpose of the focus groups, residents were asked to sign informed consent 
forms attesting to their consent to be interviewed, as well as granting permission for the focus groups to be audio-
recorded. The two research assistants were available to take separate notes of the discussions, in the event that the 
respondents agreed to participate, but did not want their responses to be audio-recorded. As each group gave 
permission to be recorded, the separate note-taking was not necessary. Upon completion of the focus groups, the 
audio recordings were transcribed by a professional transcription service.
In an effort to put the respondents at ease, and to build rapport (Patton, 1990), each focus group participant 
was encouraged to first “tell their story” as to how they chose to live in their given retirement community. After this 
ice-breaker, each group was asked to identify things that made them happy or unhappy, satisfied or dissatisfied, with 
living in their retirement community. Once a reasonably comprehensive list was developed, then each individual 
topic was canvassed in more detail. 
Preliminary to data analysis, the two authors read all of the interview transcripts. In order to insure that all 
focus group responses were coded using the same coding scheme, a consensus coding scheme was developed (Miles 
& Huberman, 1994). In a consensus coding scheme, two or more analysts agree, in advance, what term or phrase 
will be used to label each paragraph, sentence, or phrase that is pertinent to the qualitative inquiry. In this case, it 
was any statement which described a type of satisfaction or dissatisfaction with residing in the retirement 
community. After the coding scheme was agreed upon, each focus group transcript was coded independently by the 
second author and a research assistant. The first author resolved the inter-rater coding disagreements. Initial inter-
rater agreement was very high (91%). 
After the transcripts were coded, and the inter-rater coding disagreements resolved, a list of all of the 
unique satisfiers/dissatisfiers was compiled. The goal was to generate a comprehensive list of satisfiers/dissatisfiers, 
rather than generate a ranking of the importance of different factors, which is more suited to quantitative methods. 
Finally, conceptually related variables were clustered
8
(Miles & Huberman, 1994) into groupings or categories. 
After the clustering was complete, names were assigned to each cluster.