Page 234 -

222 CHAPTER 8 Interviews and focus groups

INTERVIEW ANALYSIS FOR NOVICES
Interview analysis can be somewhat intimidating. If you're feeling that you're in
a bit over your head, don't panic. Although some analyses might best be done
by an experienced collaborator, you don't need an advanced degree in the social
sciences to get a basic understanding of interview data.
In school, many people have been taught to write notes on index cards—one
idea per card—which can then be sorted and arranged as necessary. You can
break responses to interview questions into individual thoughts or ideas, one per
index card or one per line in a text document. Group lines with common ideas
but don't restrict yourself to putting any idea into only one category. Feel free to
place thoughts in multiple groups, as appropriate.
You might consider assigning categories to comments as they appear in the
transcript. This can be done by annotating each line with a colored piece of text
that names the category. Once you've done this, you can quickly search to find
out all of the instances of a particular category. As the categories begin to grow,
you may see connections between them. You can then put these categories into
broader categories, forming a hierarchy of ideas.
How do you categorize each comment or concept? One approach would be
to group things by the content words—nouns or verbs. You can use these words
to understand the objects with which people work and the actions that they
use with those objects. Organizing comments along these lines can help you
understand the outlines of the problem domain.
As you dig through the interviews, you may begin to find relationships,
information flows, sequences, or other patterns that repeatedly arise out of
the comments. Pictures, sketches, outlines, or other representations of these
interactions can help clarify your understanding.
Focus groups introduce the additional challenge of differing viewpoints. You
might consider grouping comments by individual or by the individual's role. This
might help you understand potentially important differences in perspectives.
In any case, if you are concerned about validity, enlist a colleague to
work with you. You might each independently analyze the data and then
compare your results, in the hopes of working towards a consensus analysis.
Alternatively, you might work together, building agreement as you go along.
This informal analysis shares many characteristics with more rigorous established
practices such as content analysis or discourse analysis. These approaches may differ
in their level of attention to detail and their conformance to established practices
but the goal is always the same: to help researchers move from an unordered and
undifferentiated mess of interview data to a clear, structured understanding.
Informal techniques are often sufficient. If you are trying to build an initial
understanding of a problem, gauge reaction to design proposals, or examine a
problem without aspiring for generality and validity, this approach can be very
productive. If you find that you need to add some rigor, you can always return
to the data for a second, more rigorous analysis, perhaps with the help of a
colleague with relevant experience.