Summary of Qualitative Data Analysis

Transcribing Qualitative Data

Effective transcription is foundational for qualitative data analysis. Key considerations include:
1. Accuracy and Verbatim Reflection:
1. Verbatim transcription is essential, capturing all speech elements such as pauses,
intonation, and nonlinguistic utterances.
2. Common transcription errors include:
1. Deliberate Alterations: Intentional "tidying" like omitting profanities or verbal
fillers.
2. Accidental Alterations: Misinterpretation of words or punctuation that changes
meaning.
3. Unavoidable Alterations: Omissions of nonverbal cues, such as body language
or tone, due to format limitations.
2. Training and Feedback:
Researchers must collaborate with transcriptionists to establish:
1. Conventions for inaudible sections.
2. Guidelines for emotional or complex material.
3. Notation preferences for clarity and consistency.

II) Developing a Category Scheme

This stage involves organizing data into manageable units for analysis:
1. Data Organization
• Objective: To reduce large sets of data into smaller, manageable units.
• Method: Classify and index data in a way that key parts can be retrieved without needing to
reread everything.

2. Developing a Category Scheme

• Preliminary Scheme: Sometimes researchers develop an initial template before data collection,
but more often the scheme is developed after analyzing the data.
• Categories: These are based on concepts identified in the data. The process involves:
• Close reading of the data to identify clusters and underlying concepts.
• Detail and specificity: Categories can range from concrete to abstract, depending on the
goals of the research.

3. Descriptive Categories
• Concrete Focus: In descriptive studies, the categories tend to be specific and concrete, often
focusing on actions, events, or experiences.
• Abstract Focus: In studies aiming to develop a theory, categories are more conceptual and
abstract.

5. Labeling Categories
Importance of Labels: Labels should be clear, sufficiently descriptive, and sometimes provocative to
capture the essence of the category.
Process: Abstract concepts are identified and labeled, forming the foundation for the category scheme.
III) Coding Qualitative Data
A process that involves reading and re-reading the data to ensure it accurately corresponds to the
developed categories.

1. Initial Coding Process

• Reading the Data: Once the category scheme is developed, the researcher reads through the
entire data set and assigns codes corresponding to the categories.
• Challenges:
• Ambiguity: Sometimes it's hard to determine the best code for a piece of data.
• Understanding: The underlying meaning of some segments may not be immediately
clear and could require multiple readings.
• Nuances: It often takes two or three readings to fully capture the nuances of the material.

2. Emerging Categories

• Incomplete Categories: As the researcher codes, new categories may emerge that weren’t part
of the initial template.
• Revisiting Previous Data: If a new category is discovered, it’s important not to assume it was
absent from the material that has already been coded. The researcher must go back and recheck
the previously coded data to ensure it fits into the new category.
• Complexity of Adjustments: While making changes midway can be frustrating, it's crucial to
build a comprehensive and accurate category system.

IV) Manual Methods of Organizing Qualitative Data

• Although computer software for managing qualitative data has become more prevalent, traditional
manual methods are still sometimes used, especially in smaller-scale studies or when researchers
prefer a hands-on approach.

IV) Manual Methods of Organizing Qualitative Data

1. Color-Coding with Paper Clips or Post-It Notes

• Simple Category Systems: This method is useful when the category system is simple and
manageable. Researchers use different colored paper clips or Post-It Notes to code specific
sections of text.

2. File Card System (Phenomenological Research)

• Physical Organization: In phenomenological research, researchers often use a file card system.
Significant statements from interviews or narratives are written on individual file cards.
• Sorting by Themes: The file cards are sorted into piles representing different themes, making it
easy to review material related to each theme. Different colored cards may be used for data
from different participants.

3. Conceptual Files
• File Folder System: In this method, researchers create a separate physical file for each category
or concept. Each file contains excerpts from the data that correspond to that category.
Process: Initial Reading: Researchers read through the entire dataset, marking relevant
sections with codes in the margins.
 • Cutting the Data: After coding, the data is physically cut up into excerpts corresponding
to each category. These are then filed in the appropriate folder.

V) Computer Programs for Managing Qualitative Data (CAQDAS)

• These tools allow researchers to manage and analyze large volumes of text-based data more
efficiently, without the need for physically cutting up narrative materials. While software can aid
in coding, organizing, and retrieving data, it does not perform the actual coding or provide
analysis on its own—these tasks still require critical thinking from the researcher
CAQDAS can be categorized into several types, each with specific functions:
1. Text Retrievers
2. Code-and-Retrieve Software
3. Theory Building Software
4. Concept Mapping Software
5. Data Conversion/Collection Software: This type of software converts audio into text, such as
voice recognition programs that transcribe spoken language into written text. This is particularly
useful for transcribing audio recordings of interviews.

Thematic Analysis:
• Themes: Themes are abstract entities that unify meanings from the data. They are central to
understanding experiences and can be identified through principles like similarity (finding
commonalities) and contrast (highlighting differences).
• Thematic analysis requires both across-case and within-case analyses to ensure a
comprehensive understanding of experiences. It also necessitates recognizing how themes may
vary between participants or contexts.

Tools for Analyzing and Displaying Data:

• Charting and Visualization: Tools like flowcharts, timelines, and two-dimensional matrices help
display the relationships and evolution of themes and data points over time.
• Timelines: Useful for studies involving dynamic experiences, such as the decision-making
process, to illustrate key moments and factors influencing decisions.
• Matrices: These can help organize data by participants or themes, and spreadsheets are often
used for sorting and analysis.

Who will do the analysis? Can be by single researcher, or two or more researchers
• Multiple analysts may sometimes enhance the trustworthiness of qualitative study, but can be
time consuming because teams typically meet regularly to arrive to arrive at a consensus.

Who will do the transcriptions? The major data sources are audio recorded interviews and field nots.
Preferably, the researcher “the process of transcription, while it may seem time consuming, frustrating,
and a time boring, can be an excellent way to start familiarizing yourself with data” Braun and Clarke
2006. Others, urge transcriptions by professionals as a mean of enhancing consistency and accuracy.
 The Qualitative Analysis Process

Developing a Coding Scheme

A code is used to identify a data segment such as a phrase or sentence.
 Most coding schemes are data-driven and developed through induction as analysts read their
data.
 Careful reading of the data is required to identify the concepts.
 Coding can be done using computer software or through a paper and pencil method.
 The formatting generally involves two columns: one for data and the other for the codes.
 Codes can be words, phrases, or questions (e.g., powerless, fearful, losing hope, what went
wrong?).
1. Types of Coding
 Descriptive Coding: Mainly nouns.
 Process Coding: Observable activities.
 Concept Coding: Using words or phrases.
 In Vivo Coding: Using participants' words.
 Holistic Coding: Chunks of data rather than smaller segments.
Key Points
 Once coding has been developed, the data are read in their entirety and coded for
correspondence to the codes.
 It may take several readings and rereadings of data to grasp the nuances.
 It is necessary to read all previously coded material to ensure that the code is applied
comprehensively.
 Coding converts the data into smaller, more manageable units that can be retrieved and
reviewed.
Analysis Process
 The analysis of qualitative materials often begins with the identification of broad categories,
which are clusters of codes that are conceptually connected.
 In many qualitative studies, the next phase involves the identification of themes. A theme is an
abstract entity that brings meaning and identity to a current experience.
 A theme often cuts across several categories and is a recurring regularity that captures
meaningful patterns in the data.
Additional Tools
 A dendrogram is another device that depicts the clustering of codes and categories. It is a tree
diagram that illustrates the arrangement of clusters in a hierarchical ordered system.
 Some qualitative analysts use metaphors and figurative comparisons to evoke usual and
symbolic analogies.

A dendrogram
4. Framework Analysis
 Five Steps:
1. Familiarization: Reading and listening to data.
2. Identification: Labeling data in manageable units.
3. Coding and Indexing: Using CAQDAS.
4. Charting: Organizing data for each theme into a matrix.
5. Mapping and Interpretation: Searching for patterns and relationships.