Qualitative Research

Data Analysis & Interpretation

Dr. Fadzilah Amzah

[email protected]
Learning Outcomes;
 i. Purpose of Qualitative DA & I
 ii.
The Qualitative Process of Data
Analysis
 iii. Strategies for QL Data Analysis
 iv. Interpretation of QL Data
 V. Credibility Issues in QL Data
 VI. Validating of QL Data
A. Purpose & Perspective of QL Data Analysis

 The purpose of data analysis is to bring

order to the data
 Characteristics of qualitative data
 Thick,rich descriptions
 Voluminous
 Unorganized

 Perspectives on analysis and

interpretation
 No single way to gain understanding of
phenomena
 Numerous ways to report data
Perspectives
 Researcher’s messages are not neutral
 Researcher’s language creates reality
 Researcher is related to what and who is
being studied
 Affect and cognition are inextricably linked
 What is understood is not neat, linear, or
fixed
B. Qualitative Process

(Creswell, 2012: 237)

 (i) Data Analysis During Data Collection

 Data analysis is an ongoing process

throughout the entire research project
 Analysis begins with the very first interaction
between the researcher and the participants
 This is a very important perspective given the
interpretive nature of the analysis and the
emergent nature of qualitative research
designs
 Informal steps involve gathering data,
examining data, comparing prior data to
newer data, and developing new data to
gain perspective
(ii) Data Analysis After Data Collection
 General guidelines and strategies but few specific
rules
 Common problems
 Premature conclusions
 Inexperience of the researcher
 Self-reinforcement of the researcher’s own ideas
without support from the data
 Impulsive actions
 Desire to finish quickly
 Most problems are resolved by spending time
“living” with the data
Data Analysis After Data Collection
 Inductive nature of data analysis
 Large amount of data to analyze
 Progressively narrowing data into small groups of
key data
 Multi-staged process of organizing, categorizing,
synthesizing, interpreting, and writing

Transcription is the process of converting audiotape recordings

or field notes into text data
Data Analysis After Data Collection

 Iterative process focused on

 Becoming familiar with the data and identifying
potential themes
 Examining the data in-depth to provide detailed
descriptions of the setting, participants, and
activities
 Coding and categorizing data into themes
 Interpreting and synthesizing data into general
written conclusions
Data Analysis After Data Collection
 Data management
 Creating and organizing data collected during the
study
 Purposes

Organize and check data for completeness

Start the analytical and interpretive process
 No meaningful analysis can be done without
effective data management
 Data Analysis After Data Collection
 Data management
 Suggestions
 Write dates on all notes
 Sequence all notes with labels
 Label notes according to type
 Make photocopies of all notes
 Organize computer files into folders according to
data types and stages of analysis
 Make backup copies of files
 Read through data to make sure it is legible and
complete
 Begin to note potential themes and patterns that
emerge
Data Analysis After Data Collection

 Three formal steps to analyze data

 Reading and memoing
 Describing the context and participants
 Classifying and interpreting
Data Analysis After Data Collection
 Reading and memoing
 Reading field notes, transcripts, memos, and the
observer’s comments
 The purpose is to get an initial sense of the data
 Suggestions
 Read for several hours at a time
 Make marginal notes of your impressions,
thoughts, ideas, etc.
Data Analysis After Data Collection

 Description
 What is going on in the setting and among participants
Purposes

Provide a true picture of the setting and events

to understand and appreciate the context
Separate and group pieces of data related to
different aspects of the setting, events, and
participants
Issues

The influence of context on participants’

actions and understanding
 Data Analysis After Data Collection
 Classifying and interpreting
 The process of breaking down data into small
units, determining the importance of these units,
and putting pertinent units together in a general
interpretive form
 Use of coding and classifying schemes
Topic – A basic unit of information
Category – a classification of ideas or concepts
Pattern – a relationship across categories
 3. Strategies for QL Data Analysis
Eight strategies for starting data analysis
1) Identifying themes
A good place to start analyzing data
Listing themes or patterns you have seen
emerge from the data
2) Coding data
Reducing the data to a manageable form
Guidelines
Read through all the data and attach
working labels to blocks of text
Cut and paste these blocks of text to
index cards to make it easier to organize
the data in various ways
Group the index cards together based
on similar labels
Re-visit each group of cards to be sure
each card still fits
 Data Analysis Strategies
Eight strategies
3) Asking key questions
Working through a series of questions such as
those proposed by Stringer (e.g., who is
centrally involved, who has resources, how do
things happen, etc.)
4) Doing an organizational review
Focus on the organization’s vision and
mission, goals and objectives, structures,
operations, problems, issues, and concerns
5) Concept mapping
Create a visual representation of the major
influences that have affected the study
 Data Analysis Strategies

 Eight strategies
6) Analyzing antecedents and consequences
Mapping causes and effects
7) Displaying findings
Represent findings in effective visual
displays (e.g., graphs, charts, concept
maps, etc.)
8) Stating what is missing
Identify what “pieces of the puzzle” are still
missing
 Computerized Data Analysis

 Softwareis readily available to assist with

data analysis
 Researchers must code the data
 Manipulation of the data is enhanced
 The effectiveness of this manipulation is
dependent on the researcher’s ideas, thoughts,
hunches, etc.
 Thereis considerable debate as to whether
data should be analyzed by hand or
computer
Specific Programs Available

 Atlas.ti ( www.atlasti.com)
 HyperRESEARCH ( www.researchware.com)
 MAXQDA ( www.maxqda.com).
 NVivo ( www.qsrinternational.com).
4. Interpretation of QL Data
 The purpose of the interpretation of qualitative
analyses of data
 Attempts to understand the meaning of the findings
 Larger conceptual ideas
 Consistent themes
 Relationships to theory
 Differentiating analysis and interpretation
 Analysis involves making sense of what is in
the data
 Interpretation involves making sense of what
the data mean
 Interpretation
 Insights into interpretation
 Interpretation is reflective, integrative,
and explanatory
Need to understand one’s own data to
describe it
Integrated into report writing
 Based heavily on connection, common
aspects, and linkages among data,
categories, and patterns
 Interpretation makes explicit the
conceptual basis of the categories and
patterns
 Interpretation
 Four guiding questions
 What is important in the data?
 Why is it important?
 What can be learned from it?
 So what?
 Interpretation
 Six strategies
1) Extend the analysis
 Note implications that might be drawn
2) Connect findings with personal experiences
 The researcher knows the situation better
than anyone else and can justify using his or
her experiences and perspective
3) Seek advice from a “critical” friend
 Seek the insights from a trusted colleague
4) Contextualize findings in the literature
 Uncover external sources that support the
findings
 Interpretation
 Six strategies
5) Turn to theory
Provides a way to link the findings to
broader issues
Allows the researcher to search for
increasing levels of abstraction
Provides a rationale for the work
6) Know when to say, “When!”
Don’t offer an interpretation with which
you are not comfortable
Suggest what needs to be done
5. Credibility Issues - accuracy/authenticity/ trustworthiness

Six questions to help researchers check the quality of their data

 Are the data based on your own observations or hearsay?
 Is there corroboration by others of your observations?
 Inwhat circumstances was an observation made or
reported?
 How reliable are those providing data?
 What motivations might have influenced a participant’s
report?
 What biases might have influenced how an observation was
made or reported?
6. VALIDATING THE ACCURACY OF FINDINGS
 Triangulation
 process of corroborating evidence from different
individuals (e.g., a principal and a student), types of
data (e.g., observational field notes and interviews), or
methods of data collection (e.g., documents and
interviews) in descriptions and themes in qualitative
research
 Member checking
 a process in which the researcher asks one or more
participants in the study to check the accuracy of the
account
 External audit
 a researcher hires or obtains the services of an individual
outside the study to review different aspects of the
research
Thank you…