Interviews

К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 
136 137 138 139 140 141 142 143 144 145 146 

The goal of the interviews is to guide the respondents to a discussion of the issues

without leading them to specific predetermined constructs (Rossi, Wright & Anderson,

1983). To achieve this, the interviews use open-ended questions and do not specifically

mention the phenomenon under study. Open-ended questions are augmented with

probes (Rossi et al., 1983), where the interviewer is trained to explore a respondent’s

answer to discover new concepts not originally expected in the interview guide. A sample

interview guide is included in the Appendix. Data elicited in the interviews is iteratively

validated by going back to the respondents to clarify and confirm their responses. For

example, in my study of IT personnel transition, we confirmed the responses by returning

to the participants and verifying what was revealed in the interview transcripts and

asking for clarification on issues such as time frame, and intended meaning.

Identifying Causal Statements

The interview transcripts are analyzed by identifying causal statements embedded in the

answers of the respondents, where a causal statement is defined as a statement that a

respondent makes revealing their belief that one thing causes another (Ford & Hegarty,

1984). The analysis of causal statements is an iterative process and does require informed

decision making on the part of the research team. By using this inductive approach and

allowing constructs to arise out of the interview data, we intentionally refrain from

imposing a conceptual or predefined structure on the data (Mitroff & Mason, 1982) in

a discovery or evocative setting.

The first stage of data analysis in revealed causal mapping is analyzing the interview

transcripts for causal statements, both implicit and explicit. I recommend that the

statements be first pulled out of the interview transcripts into a spreadsheet so they can

be easily traced back to the original interview document page and paragraph from which

they were elicited. The statements need not be put into the spreadsheet verbatim, but

rather they can be moved to the spreadsheet either in fragments or paraphrased.