QUALITATIVE DATA

ANALYSIS 1

AUTHORS

Jyotikumarie Juggernath: African Vision Research Institute (AVRI), Durban, South Africa
Kovin S Naidoo: Brien Holden Vision Institute (BHVI), African Vision Research Institute (AVRI), Durban, South Africa
Urmilla Bob: University of KwaZulu Natal (UKZN), Durban, South Africa
Vadivelu Moodley: African Vision Research Institute (AVRI), Durban, South Africa

PEER REVIEWER

James Loughman: Dublin Institute of Technology (DIT), Dublin, Ireland

WHAT IS QUALITATIVE DATA ANALYSIS?

Quantitative data analysis is based on meanings derived from numbers through the use of statistics that is
represented in the form of charts, diagrams, graphs, etc. Qualitative data analysis, on the other hand, concentrates on
the meanings expressed through words and the analysis is conducted through the use of conceptualisation.
Researchers are presented with a different set of procedures for data analysis that reflect on the philosophical
assumptions that underpin the qualitative research.

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-1
 Qualitative Data Analysis

WHAT IS QUALITATIVE DATA ANALYSIS? (cont.)

There are three related components in the analysis of qualitative data (Hussey and Hussey,
1997: 248)
• The systematic reduction / summarising or condensing of the material
• Thematic structuring of the data, classifying into patterns and interrelationships
• Summarising the data text into more manageable forms such as charts, graphs, tables
and illustrations
In essence, the challenge is to make sense of massive amounts of data, reduce the volume
of information, identify significant patterns and construct a framework for communicating the
essence of what the data reveals.
COMPONENTS OF Analysis examines the meaningful and symbolic content of the qualitative data. For example,
QUALITATIVE DATA by analysing data collected from interview surveys, the researcher may be attempting to
ANALYSIS understand the following:
• Peoples views and interpretation of the situation
• Why do people view the problem in the way expressed?
• What were the events that lead to that view?
• What have they been doing to contribute to or solve the problem?
The analysis of qualitative data involves two components: writing and coding into themes.
Writing: Written recordings of the data collected and the findings
Coding into themes: Identifies text or other meaningful pieces of information and then label
them into themes and sub-themes
Deductive: uses existing theory to shape the analysis process. The researcher must
develop a theoretical or descriptive framework that includes the main components of the
QUALITATIVE DATA study and the presumed relationship among them. This process guides the data analysis.
ANALYSIS Inductive: Develops a new theory emerging out of the data analysis. The researcher
STRATEGIES collects the data and then explores all the variables and components to find themes that he /
she may want to concentrate on. The data is analysed as it is collected, to develop the
conceptual framework to be used during the comprehensive data analysis phase.
In order to efficiently analyse qualitative data, a systematic and well-planned sequence of
collecting and analysing the data must be done by the researcher. Firstly, the researcher
identifies the problem and broadly based research questions. The researcher then uses the
methods of observation, asks a range of closed and open-ended questions and records the
information. Themes and categories are formed by grouping the trends and patterns found in
the examination of the data. The patterns and linkages identified in the examination are
interpreted and the researcher uses the findings to define interrelationships among themes
and categories, and thereafter attaches meaning to them. A tentative theory is developed as
the research unfolds. The inductive reasoning process is then used by the researcher to
RESEARCH STAGES compare the tentative theory to other theories.
The procedure is based on inductive reasoning and therefore has implications for qualitative
data analysis. During the qualitative analysis procedure, 3 approaches can result:
• Interpretive Approach: Some effort in identifying patterns, creating interrelationships
and interpretation is made by the researcher.
• Grounded Theory: Building a theory is the essential aim of this approach, which can
be built from abstraction and interpretation.
• Journalistic Approach: Minimal researcher intervention. Participants speak for
themselves and give explaining and solutions they see as being important.

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-2
 Qualitative Data Analysis

WHAT IS QUALITATIVE DATA ANALYSIS? (cont.)

• Theory guides the approach to analysis:

o Established conceptual framework – predetermined categories according to
research questions
o Grounded theory – theme abstraction from the data and interpretation
o Attention to abnormal cases
PRINCIPLES OF
o Gives a voice to minorities
QUALITATIVE DATA
ANALYSIS o Give way to new insights
o Lead to further investigation
o Principles of qualitative data analysis
• Data analysis is an iterative process
o Range of questions, interviews, reflecting, rephrasing, analysing, theorising,
verifying the observations and discussions
Interrelated Rather Than Sequential
• Data collection involves:
o Reading and observing: become familiar with the data
o Coding the data: examine the data for emerging themes and attach labels or codes
to the texts that represent the themes
• Following data collection:
o Display the information into themes
o Developing hypotheses, questioning and verification
STAGES IN
o Summarise the data – from the displayed data, identify the main points
QUALITATIVE DATA
ANALYSIS
Interpretation
• At each stage: data needs to be observed during all stages to search for core meanings
of thoughts, feelings, and behavioural patterns
• Overall interpretation
o Identify how themes interrelate
o Explain how the data answers the research questions
o What were the findings beyond the context of your research and what other
research should follow from your research?
• Integrative Approach: Aims to aggregate or summarise data.
o Concepts need to be clearly defined in advance
o Phenomena need to be comparable, to allow pooling
o Example: quantitative meta-analysis, meta-aggregation
APPROACHES TO
SYNTHESISING • Interpretive Approach: Aims to develop concepts or theories that integrate themes
QUALITATIVE described in the primary studies
RESEARCH o Involve induction and interpretation
o Themes from the primary studies may be synthesised into a broader explanatory
framework or new theory
o Concepts emerge through the synthesis process
o Example: meta-ethnography

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-3
 Qualitative Data Analysis

DISCOURSE ANALYSIS

Schwandt (2001)
Discourse analysis is an interdisciplinary approach drawing on insights from sociolinguistics,
cognitive psychology, communication studies and other language philosophy.
Hardy and Philips (2002)
Discourses are inter-related sets of texts (including the practices of production,
dissemination and reception) that bring an object into being.
DEFINITIONS Fairclough (2003)
Discourse analysis is concerned with the ideological effects of these constructions- that is
how texts contribute to establishing, maintaining and changing social relations of power,
domination and exploitation.
Dick (2004)
Discourse analysis works on the assumption that individuals construct the world in order to
make sense of it whilst also reproducing or challenging ideological systems of belief that
exist in society at large.

(Philips and Hardy, 2002)

Discourse analysis involves a “strong” social constructivist view of the social world. It is not
METHOD AND simply a set of techniques for conducting research because it also involves a set of
METHODOLOGY assumptions concerning the constructive effects of language. Discourse analysis explores
how the socially produced ideas and objects that populate the world are created at first and
how they are maintained and held in place.

• Media: newspaper articles, radio coverage, internet sites, etc.

• Verbal analysis: meetings, workshops, conferences, etc.
DISCOURSE
SOURCES • Academic: journals, books, etc.
• Documentation: minutes of meetings, reports (Social Impact Assessments, etc.),
policies, etc.

• Hidden relations of power in texts and discussions

• What is not written or said? – issues that are neglected or ignored
• Who is exercising power? – What voices / perspectives are present?
• Who are consulted and who are not?
• Who are the target groups / audience?
• What events / information / aspects are presented and discussed? How?
• What are the contradictions / differences and how are they handled?
ASPECTS WHILE
UNDERTAKING A • What is the context of the text being produced or discussed? (see Figure 8.1 below)
DISCOURSE
ANALYSIS Context

Interpretive Critical Discourse

Structuralism Analysis
Constructivist Critical
Social Linguistic Critical Linguistic
Analysis Analysis

Text

Figure 8.1: Different Approaches to Discourse Analysis (Philips and Hardy, 2002)

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-4
 Qualitative Data Analysis

DISCOURSE ANALYSIS

Investigates the following:

• What terms / concepts / level of communication in terms of language is used?
LANGUAGE ISSUES
• What are the language differences between and among the group being analysed and
who is doing the analysis?

• What texts / meetings / documentation to include and why

• How to analyse the collected data
CHALLENGES • How to be reflexive
• How to write-up the results
• Is information needed accessible / available?

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-5
 Qualitative Data Analysis

PROCESSES IN QUALITATIVE DATA ANALYSIS

• Read for content

o To see if what you set out to collect is being received
o To identify themes and develop possible explanations
o To identify new themes / topics that need to be explored in further fieldwork
• Note the quality of the data
o Are the responses superficial, or rich and deep?
o Are the descriptions of observations coming across clearly?
o Is there sufficient contextual detail?
o Problems in the quality of the data may mean the following:
1. READING AND
- The questions are not asked properly
OBSERVING THE
DATA - The venue may not be appropriate
- Group composition is inappropriate in size
- The interviewer style of asking questions may not be appropriate
- Information is not being immediately recorded
• Observing patterns
o Following the identification of themes, examine the patterns
- Are themes evident in all or some of the data?
- Are there evident relationships between the themes?
- Conflicting responses
- Is more information needed to increase understanding?
• Code the identified themes
o No standard procedure to code themes
o Deriving codes and the level of detail needed varies among different researchers
o Codes / labels are generally inserted into the margins
o Flag ideas you find in the transcript
o Create sub-themes for in-depth investigation
o Insert explanatory notes / questions during the coding process
o Coding – Identifying emerging themes
2. CODING • Code continuously as the process unfolds
o Use a systematic approach
o Helps to identify gaps or questions: Can go back to collect more data immediately
o Biases are detected early
o Concepts can be re-defined
• Building theme related files
o Conduct a coding sort
- Cut and paste together into one file similarly coded blocks of text
- Use identifiers that help you to identify the original source

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-6
 Qualitative Data Analysis

PROCESSES IN QUALITATIVE DATA ANALYSIS (cont.)

o Conduct quantitative and qualitative analysis

o Capture the variation or richness of each theme
3. DISPLAYING
o Note differences between individuals and sub-groups
DATA
o Organise into sub-themes
o Revisit the data and examine evidence that supports each sub-theme
o Extract meaning from the data
4. DEVELOPING o Do the created categories make sense?
HYPOTHESES, o What information contradicts the researcher objectives?
QUESTIONING
AND o Is there missing information?
VERIFICATION o What other opinions are there to consider?
o Is there any biased influence from the researcher?
o Selects the key points in the data
5. DATA
o Sorts out the main theme and identifies sub-themes
REDUCTION /
SUMMARY o Sifts for essential information
o Use visual techniques to represent e.g. diagrams
• Reliability
o Is the meaning constant with data collected?
o Have they been verified with the respondents?
o Have multiple perspectives been considered?
o Are there findings that go beyond your intended objectives?
6. MEANING OF • Dependability
INTERPRETATION o Can findings be replicated?
• Confirmability
o Audit trail
- Permits external review of analysis decisions
• Transferability
- Can lessons be transferred to other contexts?

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-7
 Qualitative Data Analysis

CASE STUDY DATA ANALYSIS

Pattern Matching and Time Series analysis are the most commonly used methods to analyse case study data:

PATTERN MATCHING

Attempt to link several pieces of information to an idea or theory.

For example: If the researcher was examining Vision Centres, he / she would have made an assumption that Vision
Centres with broader community support from eye care groups in the community would be more successful in
developing vision centres with other vision institutions / organisations and authorities.

TIME SERIES ANALYSIS

Data point trends are compared to a significant theoretical trend or other exploration.

For Example: If the researcher was testing the theory that social entrepreneurial ventures in spectacle provision helps
alleviate poverty, he / she would need to gather the data that is relevant, compare it to the theoretical unfolding of
events in spectacle provision by social entrepreneurs and see if it fits the time order suggested to alleviate poverty.

Feb 2013, Version 1-1 Research Methodology and Project Design

Chapter 8-8