Hatch Chapter 4

Analyzing Qualitative Data
A systematic search for meaning
A way to process qualitative data so that
what has been learned can be communicated
to others
Organizing and interrogating data in ways
that allow researchers to see patterns,
identify themes, discover relationships,
develop explanations, make interpretations,
mount critiques, or generate theories
Asking questions of the data
Start soon after data collection has begun
 Allows researchers to shape the direction of
future data collection based on what they are
actually finding or not finding
Keep analyzing until you have answered your
research questions
Part of a continuum
 Typological
 Inductive
 Interpretive
 Political
 Polyvocal
Dividing the overall data set into groups or categories
based on predetermined typologies
 Generated from theory, common sense, and/or research
Good for interview studies and processing artifact
 Not recommended for observational studies
 Efficiency
▪ Because categories are predetermined
 Potentially blinds the researcher to other important
dimensions in the data
Identify typologies to be analyzed
 Selection should be fairly obvious
▪ Predetermined
Read the data, marking entries related to
your typologies
 Read through the data completely with one
typology in mind
 Does this information relate to my typology?
▪ Mark that portion of the data so that you can go back to
it later for closer examination
Read entries by typology, recording the main
ideas in each entry on a summary sheet
 This time only the data within the typology of
interest will be read
 A summary sheet should be created for each
▪ Write a brief statement of the main idea of the excerpt
on the summary sheet
 Not the step to be interpret for significance
Look for patterns, relationships, themes within typologies
 What broad statements can be made that meaningfully bring all of these data
 Patterns are regularities
Similarity (things happen the same way)
Difference (they happen in predictably different ways)
Frequency (they happen often or seldom)
Sequence (they happen in a certain order)
Correspondence (they happen in relation to other activities or events)
Causation (one appears to cause another)
 Relationships are links
Strict inclusion (X is a kind of Y)
Rationale (X is a reason for doing Y)
Cause-effect (X is a result of Y)
Means-end (X is a way to do Y)
 Themes are integrating concepts
▪ What broad statements can be made that meaningfully bring all of the data together?
Read data, coding entries according to
patterns identified and keeping a record of
what entries go with what elements of your
 Make a simultaneous record of where elements
related to the category are found in the data
Decide if patterns are supported by the data,
and search data for nonexamples of your
 Decide if the evidence is strong enough to support
your case, or
 Ask if there is evidence upon which other cases,
even competing cases, can be made
Look for relationships among the patterns
 Step back from individual analyses that have been
completed and look for connections across what
has been found
 Making visual representations of categories can
Write your patterns as one-sentence
 Generalization: expresses a relationship between
two or more concepts
 Making yourself construct sentences forces you to
organize your thinking into a form that can be
understood by yourself and others
 Gives closure to your analyses
Select data excerpts that support your
 Go back to the data to select powerful examples
that can be used to make your generalizations
come alive for your readers
Similar to Typological Analysis, except categories are not
 Begins with particular pieces of evidence, then pulls them
together into a meaningful whole
 Works well with studies that emphasize the discovery of cultural
meaning from large data sets that include observational data
(postpositivist and constructivist)
 Works less well for studies that focus on answering narrowly defined
questions or that rely on interview data almost exclusively
 Its power to get meaning from complex data that have been gathered
with a broad focus in mind
 Provides a systematic approach for processing large amounts of data
Read the data and identify frames of analysis
 What will be my frames of analysis?
▪ Frames of analysis: levels of specificity within which
data will be examined
▪ No analysis yet; put rough parameters on how you will
start looking closely at the data
 Must begin with a solid sense of what is included
in the data set
▪ The data will be read over and over
Create domains based on semantic relationships
discovered within frames of analysis
 Domains are categories organized around relationships that can
be expressed semantically
 Develop a set of categories of meaning or domains that reflect
relationships represented in the data
Strict inclusion (X is a kind of Y)
Spatial (X is a place in Y)
Cause-effect (X is a result of Y)
Rationale (X is a reason for doing Y)
Location for action (X is a place for doing Y)
Function (X is used for Y)
Means-end (X is a way to do Y)
Sequence (X is a step in Y)
Attribution (X is a characteristic of Y)
Identify salient domains, assign them a code,
and put others aside
 Narrow the focus of your analysis
▪ “Data reduction”
 Assign a Roman numeral to each domain and a
capital letter to each included term
 Could this relationship be linked to other domains
discovered in the data?
 More questions to ask yourself on page 168
Reread data, refining salient domains and
keeping a record of where relationships are
found in the data
 Read the data with specific domains in mind
▪ Make a record of where they are located
Decide if your domains are supported by the
data and search data for examples that do
not fit with or run counter to the relationships
in your domain
 Up until now, domains have been hypothetical
and tentative
 Deductive reasoning is fully employed to decide if
the hypothetical categories identified hold up
 Search for counterevidence
▪ Questions to ask yourself on page 170
Complete an analysis within domains
 Looking within the domains identified for
complexity, richness, and depth
▪ Study the data that have been organized into domains
in ways that allow the discovery of new links, new
relationships, and new domains
▪ In search for other possible ways to organize what’s there
▪ Going much deeper into the data by looking beneath the surface
of included terms for richer representations
Search for themes across domains
 Look for connections or themes among them
 Systematic comparison
▪ How does this all fit together?
▪ What’s the same or different about these domains?
 Make a “data display”
▪ Visual formats that present information graphically or
 Write a summary statement
 More analytic questions on page 173
Create a master outline expressing
relationships within and among domains
 Provides an opportunity to refine the analysis
done to this point
Select data excerpts to support elements in
your outline
 Powerful or prescient quotes should be starred in
the data and on the domain sheets
Giving meaning to data
 Generating explanations for what’s going on within
▪ Making inferences, developing insights, attaching
significance, refining understandings, drawing conclusions,
and extrapolating lessons
 Situates the researcher as an active player in the
 Researchers will usually do typological or inductive
analysis prior to this model
 Fits most comfortably within the constructivist
Read the data for a sense of the whole
Review impressions previously recorded in
research journals and/or bracketed protocols,
and record these in memos
 The object is to get a handle on which impressions
might lead to more careful examination
▪ Will lead to the identification of relationships among
impressions and the formation of new impressions
 Memos can take many forms
▪ At this point they should be written in tentative,
hypothetical language with complete sentences and
Read the data, identify impressions, and
record impressions in memos
 Systematically make and record your
interpretations of what is happening within the
social contexts captured in your data
▪ Discover new impressions that may develop into
interpretations that bring meaning to your data
▪ Analytic questions on page 184
 The product of steps 2 & 3 are sets of memos that
form the raw material on which more formal
interpretations can be based
Study memos and salient interpretations
 Read through the entire set of memos
▪ Organize the memos according to how they relate to
one another and how they connect to the issues you
want to address in your research
▪ Begin to get a sense of the big picture you will be
drawing for your reader
Reread data, coding places where
interpretations are supported or challenged
 Search for places that relate directly to the
interpretations in your memos
▪ A deductive activity
▪ What are all the places in the data where my
interpretations are addressed?
Write a draft summary
 It will not include an extensive data display or
context description but will be focused on
communicating the explanations, insights,
conclusions, lessons, or understandings you have
down from your analysis
▪ A “story” that others can understand
▪ Provides a test for logical consistency of your thinking
and expose any gaps in your argument that might exist
▪ Don’t write in shorthand
Review interpretations with participants
 “Member check”
▪ Invite them to a working session
 Participants should have the chance to consider
and give their reactions to the interpretations
included in the summary just written
▪ Can also show copies of memos and even research
Write a revised summary and identify
excerpts that support interpretations
 Communicate the understandings you have
constructed, clarify what they mean in the
contexts of your study, and represent what is
captured in your data
 Identify a collection of possible quotes that will
help convince your readers that your
interpretations are well founded
Provides a framework that builds in analytic
integrity so that findings are grounded in
data while acknowledging the political nature
of the real world and the research act
Designed to accommodate the
critical/feminist paradigm
 It can be modified for analyzing virtually any type
of observation, interview, or unobtrusive data
collected in these kinds of studies
Read the data for a sense of the whole and
review entries previously recorded in research
journals and/or bracketed in protocols
 The object of the reading is to see the forest—the
trees will not go away
Write a self-reflexive statement explicating your
ideological positionings and identifying
ideological issues you see in the context under
 Gives the researcher a chance to spell out what you
believe and where you stand on issues related to your
 Write out your best guesses about the ideological
issues that are salient to the context you are studying
▪ Important to do both these in writing, in paragraph form
Read the data, marking places where the
issues related to your ideological concerns
are evident
 Where are all the places in the data that include
information related to the ideological issued
 Deductive thinking—finding examples that fit
your issues
Study places marked in the data, then write
generalizations that represent potential
relationships between your ideological
concerns and the data
 Sets of generalizations related to each of your
 Discover the connections between what you
thought you might find and what is there
▪ Then develop written generalizations that express the
relationships discovered within each issues
Reread the entire data set and code the data
based on your generalizations
 Going back to the original complete data set
Decide if your generalizations are supported
by the data and write a draft summary
 If they hold up against all the data you have so far
 Product of this step will be a draft summary that
reports the final versions of your generalizations
organized as a narrative
▪ Take this back to the participants of your study
▪ Written for them, the primary audience
Negotiate meanings with participants,
addressing the issues of consciousness
raising, emancipation, and resistance
 Summaries will be designed to expose the
dimensions of oppression experienced by the
individuals being studied
▪ Raise their consciousness about what is going on around
them, and benefits, and why
Write a revised summary and identify
excerpts that support generalizations
 Revise your summary to include what you learned
from the negotiations in the previous steps
One kind of analysis that fits within the
assumptions of the poststructuralist
Read the data for a sense of the whole
Identify all of the voices contributing to the
data, including your own
 You will have structured your data collection
around your objective to capture particular voices
▪ The objective is to identify all possible voices
▪ Later you will decide which voices to include your final report
▪ Essential that you count your own voice
▪ Already should have decided who to talk to, what to ask, what
will be recorded, what will be analyzed, and what will be included
Read the data, marking places where
particular voices are heard
 Assign some sort of identifier to each voice, read
the data, making decisions about whose voice is
represented in each data excerpt and mark the
 Product: separate sets of data divided by voices
Study the data related to each voice, decide which
voices will be included in your report, and write a
narrative telling the story of each selected voice
 Ask the data to tell you what each voice you have
identified has to say about your research focus
▪ Entries related to particular voices should be processed at this time
▪ Make a decision about which voices should be included in the final
▪ Important criteria for inclusion: the contribution of each voice’s story to
revealing different perspectives on the topic of study
 Must be sufficient support in your data to construct a story for each voice
you select
▪ Draft an initial version of the story you plan to tell for each voice
▪ Develop and discover a plot that links the data together
Read the entire data set, searching for data
that refine or alter your stories
 Do not expect everything to fit together in a tidy
Whenever possible, take the stories back to
those who contributed them so that they can
clarify, refine, or change their stories
 This step builds on ethical and methodological
 Improves the balance of power in the construction
and ownership of stories
▪ Improves the quality of the stories that have been
Write revised stories that represent each
voice to be included
 Revise your drafts, taking into account the
comments and concerns of your participants
Assists in the sorting and organization of data
 Can be an efficient alternative to doing the same
work by hand
▪ But cannot perform the “mind-work” that humans can
List of advantages and disadvantages on
page 208
Hatch, J. A. (2002). Doing Qualitative Research
in Education Settings. Albany, New York: State
University of New York Press

similar documents