| The chapter, Analyzing
qualitative data, in an evaluation handbook produced
by the Division of Research, Evaluation, and Communication,
with The National Science Foundation, offers the following
practical advice about conducting analyses.1 |
|
|
- Start the analysis right away and keep a
running account of it in your notes: It cannot
be overstressed that analysis should begin almost
in tandem with data collection, and that it is an
iterative set of processes that continues over the
course of the field work and beyond. It is generally
helpful for field notes or focus group or interview
summaries to include a section containing comments,
tentative interpretations, or emerging hypotheses.
These may eventually be overturned or rejected, and
will almost certainly be refined as more data are
collected. But they provide an important account of
the unfolding analysis and the internal dialogue that
accompanied the process.
- Involve more than one person: Two
heads are better than one, and three may be better
still. Analysis, especially qualitative analysis,
need not and in many cases should not, be a solitary
process. It is wise to bring more than one person
into the analytic process to serve as a cross-check,
sounding board, and source of new ideas and cross-fertilization.
It is best if all analysts know something about the
type of analysis being used, as well as the substantive
issues involved. If it is impossible or impractical
for a second or third person to play a central role,
his or her skills may still be tapped in a more limited
way. For instance, someone might review only certain
portions of a set of transcripts or analyses.
- Make sure enough time and money are available
for analysis and writing: Analyzing and writing
up data almost always takes more time, thought, and
effort than anticipated. A budget that assumes a week
of analysis time and a week of writing for a project
that takes a year’s worth of data collection
is highly unrealistic. Along with revealing a lack
of understanding of the nature of analysis, failing
to build in enough time and money to complete this
process adequately is probably the major reason that
evaluation reports come up short.
- Be selective when using computer software
packages, especially in qualitative analysis:
When selecting a quantitative software package, make
sure it is user friendly, and accessible by all of
those who will need access to the data. In addition,
make sure the package can perform any statistical
tests that are needed. With respect to qualitative
software, in recent years, there has been a great
proliferation of qualitative analysis software packages.
Most of these packages were reviewed by Weitzman and
Miles (1995), who grouped them into six types: word
processors, word retrievers, textbase managers, code-and-retrieve
programs, code-based theory builders, and conceptual
network builders. All have strengths and weaknesses.
Weitzman and Miles suggested that when selecting a
given package, researchers should think about the
amount, types, and sources of data to be analyzed
and the types of analyses that will be performed.2
Two caveats are in order. First, computer software
packages for qualitative data analysis essentially
aid in the manipulation of relevant segments of
text. While helpful in marking, coding, and moving
data segments more quickly and efficiently than
can be done manually, the software cannot determine
meaningful categories for coding and analysis or
define salient themes or factors. Regardless of
the software, the analytic underpinnings of the
procedures must still be supplied by the analyst.
Software packages cannot eliminate the thinking
required for a sound qualitative analysis. Second,
since it takes time and resources to become good
at using a given software package and learning its
quirks, researchers may want to consider whether
the scope of their project, or their ongoing needs,
truly warrant the investment.
---------------
1. Source: Frechtling, J., Sharp,
L., & Westat, eds. (1997). Analyzing qualitative
data. In: User-friendly handbook for mixed method
evaluations. Arlington, VA: National Science Foundation,
Directorate for Education and Human Resource.
2. Source: Weitzman, E.A., & Miles,
M.B. (1995). A software sourcebook: Computer programs
for qualitative data analysis. Thousand Oaks, CA:
Sage. |
