qualitative data analysis - viva university · · 2013-11-18this session seeks to discuss...

Research Methods – Dr Richard Boateng [[email protected]] Photo Illustrations from Getty Images – www.gettyimages.com 1

Qualitative Data Analysis

Lecturer/Convenor:

Richard Boateng, PhD.

Email:

[email protected]

Office: UGBS RT18 (rooftop)

Qualitative Data Analysis

Part 1


Class Website

• www.vivaafrica.info

Use the class website WEEKLY, ask/comment on the articles,

and JOIN the FACEBOOK

Please Add your name and the course code


Learning Objectives

This session seeks to discuss qualitative data analysis

methods for case study research.

• By the end of the session, students will learn about

Miles and Huberman Data Analysis Approach.

• Students will also be able to understand and explain

within case analysis and cross-case analysis.


Readings

• Miles and Huberman Data Analysis

– http://tinyurl.com/qbewmxr

• Analysis of Qualitative Data (Punch)

– http://tinyurl.com/o3ceabq


Readings

• Creswell, J. W. (2007). Qualitative Enquiry

and Research Design: Choosing Among

Five Approaches. SAGE Publications.

London

• Data Analysis and Representation

• Chapter 8

www.tinyurl.com/creswell2007a


Relevant Chapters

• Mason, Jennifer. Qualitative Researching,

2nd edition, Sage, 2002.

• Miles, Matthew B., and Huberman, A.

Michael. Qualitative Data Analysis, 2nd

ed., Sage, 1994. (or recent edition)


Other Readings

• Eisenhardt, K. M. (1989) Building Theories from Case study

Research, Academy of Management Review, Vol. 14, Issue 4,

pp 532-550.

– www.tinyurl.com/8m27nd2

• Yin, R. K. (2003). Case Study Research: Design and Methods

(3rd edition, Volume 5). SAGE Publications: London.

– www.tinyurl.com/9co7xfa

Other books – www.goo.gl/l6RA2

http://www.tinyurl.com/9co7xfa


Components of Case Study Research Design

1. A study’s questions;

2. Its propositions, if any;

3. Its unit(s) of analysis;

4. the logic linking the data to the

propositions;

5. the criteria for interpreting the findings.


Recap: Forms of Qualitative Data

1. Observation (non-participant – participant)

2. Interviews (close – open)

3. Documents and Text (private – public)

4. Audio-visual (audio, pictures, mobile phone

text, social media, video et cetera)


Miles and Huberman’s Data Analysis • Transcendental Realism

– Data reduction, data display and drawing and

verifying conclusion


Miles and Huberman’s Data Analysis • Data Reduction

– Occurs throughout the analysis.

– Early stages: it begins with editing, segmenting and

summarizing data.

– Middle stages: it begins with coding and memoing to

find themes, clusters and patterns.

– Latter stages: conceptualizing and explaining to

develop abstract concepts.

– You need to reduce data without losing key

information which support explanation and give

evidence. Do not strip data from their context.


Miles and Huberman’s Data Analysis • Data Display

– Organize, compress and assemble information

– “YOU KNOW WHAT YOU DISPLAY”

– HOW: Graphs, Charts, Networks, Diagrams of

Different types (Venn Diagrams, Casual Models)

– WHY: organize and summarize the data to

establish/explain themes, and also become the basis

for future analysis

– Good qualitative analysis involves repeated and

iterative displays of data.


Miles and Huberman’s Data Analysis • Drawing and Verifying Conclusions

– Reduction and Display enables the Drawing Conclusions

– “YOU KNOW WHAT YOU DISPLAY”

– First-level/Early Conclusions may be identified

• Vague and ill-formed at this stage –More Tentative

– Conclusions come in the form of Propositions – once

they have been verified

– 13 Tactics for drawing meaning and conclusions

– 13 Tactics for testing and confirming findings


Tactics for Generating Meaning 1. Noting Patterns and themes

2. Seeing Plausibility

3. Seeing Clustering - what goes with what

4. Making Metaphors - integration in diverse pieces of data

5. Counting - to see what is there

6. Making contrasts/comparisons

7. Partitioning Variables

8. Subsuming Particulars into the General

9. Factoring

10. Noting relationships between Variables

11. Finding Intervening Variables

12. Building a Logical Chain of Evidence

13. Making Conceptual/theoretical coherence

Page 245 - 277


Testing and Confirming Meanings 1. Checking for representativeness

2. Checking for researcher effects

3. Triangulating Data Sources

4. Weighting the evidence

5. Checking the meaning of outliers

6. Using Extreme cases

7. Following up surprises

8. Looking Negative evidence

9. Making if-then tests

10. Ruling out spurious relations

11. Replicating a finding

12. checking out rival explanations

13. Getting feedback from informants

Page 245 - 277


Coding to start Analysis

1. Coding the data

– Reducing the data into meaningful segments

and assigning names to the segments

2. Combining the codes into broader

categories or themes

3. Displaying and making comparison in

data graphs, tables, and charts



Codes are tags, names or labels. Coding is the process of

putting tags, names or labels against pieces of data. Data can

individual words, or small or large chunks of the data.

Purpose

1. Index data, providing a basis for storage and retrieval.

2. Become the basis or new data for future analysis

3. Pulling themes together and identifying patterns

Types of Codes

Descriptive Codes/Topic Codes

– No or little inference is made in such cases

Inferential/Pattern Codes

– Pull together material into smaller and more meaningful units



HOW TO CODE;

Two approaches are not an either-or decision

Framework Approach

1. Prespecified codes from literature or research framework

2. Be alert for other codes suggested by data

DATA-DRIVEN APPROACH

1. First set of codes are derived from data

2. Relies on Coding scheme after initial analysis

Punch – Introduction to Research Methods in Education


Types of Codes

Miles and Huberman (1994)

Descriptive Codes

Pattern Codes

Richards (2005)

Topic Codes

Analytic Codes

Grounded Theory Codes

In vivo codes – Focus on what is in the data

Open Codes – raises the conceptual level of data

Axial Codes – focus on Interconnections between open codes

Selective Codes – raises the conceptual level of data again


‘My supervisor barely met with me; I had to do much by myself’

‘Supervisors are usually Worldbank experts and very knowledge but often away

for global assignments’

‘Two of my colleagues got internships at WorldBank during their PhD, however,

it affected their completion time’

‘I spent more time on skype for discussions with my supervisor’

Limited face-to-face interaction

Knowledgeable supervisors

Technoculture

Proactive / Individualistic

Global Orientation

Types of Codes

Descriptive Codes

– Coding and storing information about the cases being studies

Topic Codes

– Labels pieces of text according to its subject

Inferential/Pattern Codes

– Pull together material into smaller and more meaningful units



Types of Codes

Analytical Coding

• Involves the interpretation of data and the conceptualizing and

theorizing of data

Example


Memoing in analysis

Starts alongside coding…

When Data finds links with theory and previous literature discussion


Memoing in analysis


Record all Ideas when they happen and as they happen, as memos. When

it happens during coding, stop and record the idea.


Mobiles and micro-trading

• <


Abstracting and Comparing Continuum of abstraction: Leave the concrete to abstract

Specific to General


Grounded Theory Analysis


Grounded Theory Analysis

• Substantive codes from open coding

– Initial Conceptual categories in the data

• Theoretical Codes from axial coding

– Connecting categories

• Core coding from selective coding

– Higher-order conceptualization of the

theoretical coding to develop theory


Organizing Qualitative Data for Case Study

Step Activity

Data Managing Create and organize files for data

Reading,

Memoing

Read through text, make margin notes

form initial codes

Describing Describe the case and its context

Classifying Use categorical aggregation to establish

themes or patterns

Interpreting Use direct interpretation

Develop naturalistic generalizations

Representing,

visualizing

Present in-depth picture of the case using

narrative, tables and figures


Analysis within Case study

Within Case Analysis:

• Analysis consists of making a detailed

description of the case and its setting.

• In Categorical Aggregation, the researcher

seeks a collection of instances from the data,

hoping that issue-relevant meanings will

emerge.

• Also, the researcher establishes patterns and

looks for a correspondence between two or

more categories.



Case study Research:

• In direct interpretation, on the other hand, the

case study researcher looks at a single

instance and draws meaning from it without

looking for multiple stances.

• It is a process of pulling the data apart and

putting them back together in more meaningful

ways.



Cross-case study

• Analytic technique when the researcher

studies two or more cases.

• A word table can be created to display the

data from individual cases according to

some uniform framework.



Case study Research:- Cross-case

• The implication of this is that the researcher can

look for similarities and differences among the

cases.

• Finally, the researcher develops naturalistic

generalizations from analyzing the data; generalizations that people can learn from the case

either for themselves or to apply to a population of

cases.


Other Qualitative Data Analysis

• Narratives and Meaning

• Conversational Analysis

• Discourse Analysis


Other Qualitative Data Analysis


• Thank YOU


Case Study Research Process