Data Analysis Methods: Qualitative

Data Analysis Methods: Qualitative

By KHO CHUNG WEI

In this topic, you will:

• Analyse qualitative data

• ​Interpret qualitative data

• Present qualitative data​

TSLB3143

Fundamentals of Research in Education

What forms of data do qualitative researchers collect?

TSLB3143

Fundamentals of Research in Education

1. Fieldnotes = ​the written account of what the researcher hears, sees, experiences and thinks in the course of collecting and reflecting on the data in a qualitative study (best written right after you leave the field)

   • Descriptive = represents the researcher's best effort to objectively record what has occurred in the field:

     • portraits of the subjects

     • reconstruction of dialogue

     • description of physical setting

     • accounts of particular events

     • depiction of activities

     • observer's behaviour

What forms of data do qualitative researchers collect?

TSLB3143

Fundamentals of Research in Education

1. Fieldnotes

   • Reflective = observer's comments and memos:

     • reflections on analysis

     • reflections on method

     • reflections on ethical dilemmas and conflicts

     • reflections on the observer's frame of mind

     • points of clarification

2. Transcripts from taped interviews

What forms of data do qualitative researchers collect?

TSLB3143

Fundamentals of Research in Education

3. Documents

   1. personal documents = any first-person narrative that describes an individual's actions, experiences, and beliefs:

      • intimate diaries

      • personal letters

      • autobiographies

   2. official documents = documents produced by organisations for specific kinds of consumption:

      • internal documents = communications circulated inside an org.

      • external communication = materials produced for public consumption

      • student records and personnel files

What forms of data do qualitative researchers collect?

TSLB3143

Fundamentals of Research in Education

3. Documents

   c. popular culture documents, e.g. videos, educational and feature films, rock

   and roll, magazines, television, romance novels, advertisements

   d. ​photography:

   • ​found photographs = photographs that are available because others have taken them

   • researcher-produced photographs

4. Official statistics and other quantitative data

Multiple Choice

When you collect samples of a student's work (e.g. their exercise books, worksheet), which form of data is it?

• fieldnote

• personal documents

• official documents

• photographs

• official statistics and other quantitative data

Some definitions (in qualitative research)

• ​Data analysis = the process of systematically searching and arranging the interview transcripts, fieldnotes, and other materials that you accumulate to enable you to come up with findings

• ​Data interpretation = developing ideas about your findings and relating them to the literature and to broader concerns and concepts

​

​Findings and ideas about findings emerge together. Therefore, it is difficult to separate data analysis from data interpretation in qualitative research.​

TSLB3143

Fundamentals of Research in Education

Suggestions on making analysis and interpretation an ongoing part of data collection

1. ​Force yourself to make decisions that narrow the study

2. ​Develop analytic questions

3. ​Plan data-collection sessions in light of what you find in previous observations

4. ​Write many "observer's comments" about ideas you generate

5. ​Write memos to yourself about what you are learning

6. Try out ideas ​and themes on informants

7. ​Begin exploring the literature while you are in the field

8. ​Play with metaphors, analogies and concepts

9. ​Use visual devices, e.g. diagrams, tables, matrices, graphs

10. Do not be afraid to speculate​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

• ​Codes

  • = labels that assign symbolic meaning to the descriptive or inferential information compiled during a study

  • attached to data chunks of varying size and can take the form of a straightforward, descriptive label or a more evocative and complex one (e.g. metaphors)

  • primarily used to retrieve and categorize similar data chunks so the researcher can quickly find, pull out, and cluster the segments relating to a particular research question, construct, or theme

  • prompts / triggers for deeper reflection on the data's meanings

  • enables data condensation and discovery

  ​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Elemental methods to coding

1. ​Descriptive coding

   • assigns labels to data to summarize in a word or short phrase—most often a noun—the basic topic of a passage of qualitative data

   • eventually provides an inventory of topics for indexing and categorizing, which is especially helpful for studies with a wide variety of data forms

   • Example from the fieldnotes about a lower-middle-class neighbourhood:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Elemental methods to coding

2. ​In vivo coding

   • uses words or short phrases from the participant’s own language in the data record as codes (phrases used repeatedly are good leads)

   • appropriate for all qualitative studies, but particularly for studies that prioritise ​and honour the participant's voice

   • Example from an interview transcript about a girl's experiences in school:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Elemental methods to coding

3. Process coding

   • uses gerunds (“-ing” words) exclusively to connote observable and conceptual action in the data, implying actions intertwined with the dynamics of time, e.g. things that emerge or change

   • appropriate for all qualitative studies, but particularly for grounded theory

   • Example from an interview transcript about a girl's explanation of how rumours get spread:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Affective methods to coding

4. Emotion coding

   • labels the emotions recalled and/or experienced by the participant or inferred by the researcher about the participant

   • particularly appropriate for studies that explore intrapersonal and interpersonal participant experiences and actions

   • Example from an interview transcript about a man complaining about one of his colleagues:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Affective methods to coding

5. Values coding

   • reflects a participant’s values (V), attitudes (A), and beliefs (B), representing his or her perspectives or worldview

   • appropriate for studies that explore cultural values, identity, intrapersonal and interpersonal participant experiences and actions in case studies

   • Example from an interview transcript about a university student discussing her political beliefs:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Affective methods to coding

6. Evaluation coding

   • assigns judgments about the merit, worth, or significance of programs or policy

   • appropriate for policy, critical, action, organizational, and evaluation studies, particularly across multiple cases and extended periods of time

   • Example from an interview transcript of a teacher assessing a program:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Literary and language method to coding

7. Dramaturgical coding

   • applies the terms and conventions of character, play script, and production analysis onto qualitative data

   • appropriate for exploring intrapersonal and interpersonal participant experiences and actions in case studies, power relationships, and the processes of human motives and agency

   • Example from an instructor's dilemmas with her unit's budget cut:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Exploratory methods to coding

8. Holistic coding

   • applies a single code to a large unit of data, rather than line-by-line coding, to capture a sense of the overall contents and the possible categories that may develop

   • most applicable when the researcher has a general idea as to what to investigate in the data

   • Example from fieldnotes of how a new faculty oriented to academia:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Exploratory methods to coding

9. Provisional coding

   • begins with a “start list” of researcher-generated codes, based on what preparatory investigation suggests might appear in the data before they are collected and analyzed

   • can be revised, modified, deleted, or expanded to include new codes

   • appropriate for qualitative studies that build on or corroborate previous research and investigations

   • Example from a researcher about to

     interview people who quit smoking:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Exploratory methods to coding

10. Hypothesis coding

    • application of a researcher-generated, predetermined list of codes onto qualitative data specifically to assess a researcher-generated hypothesis

    • Statistical applications, if needed, can range from simple frequency counts to more complex multivariate analyses

    • appropriate for hypothesis testing, content analysis, and analytic induction of the qualitative data set, particularly the search for rules, causes, and explanations in the data

    • E.g.: It is hypothesised that the responses to a particular question on language issues will generate one of four answers:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Procedural methods to coding

11. Protocol coding

    • according to a preestablished, recommended, standardized, or prescribed system

    • applied the comprehensive list of codes after data collection is completed

    • appropriate for qualitative studies in disciplines with previously developed and field-tested coding systems

    • E.g.:a selected list of codes from a protocol used to determine the causes of family violence:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Procedural methods to coding

12. Causation coding

    • extracts attributions or causal beliefs from participant data about not just how but why particular outcomes came about

    • searches for combinations of antecedent and mediating variables that lead toward certain pathways and attempts to map a three-part process as a CODE 1 > CODE 2 > CODE 3 sequence

    • appropriate for discerning motives, belief systems, worldviews, processes, recent histories, interrelationships, and the complexity of influences and effects on human actions, evaluating the efficacy of a particular program or as preparatory work before diagramming or modeling a process through visual means such as decision modeling and causation networks

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Procedural methods to coding

12. Causation coding

    • E.g.: a survey respondent describes in writing what challenges she faced when she took speech classes in high school

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Grammatical methods to coding

13. Attribute coding

    • the notation of basic descriptive information such as the fieldwork setting, participant characteristics or demographics, data format, and other variables of interest

    • appropriate for all, but particularly for those with multiple participants and sites, cross-case studies, and studies with a variety of data forms

    • Example from a dataset

      about an educational study:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Grammatical methods to coding

14. Magnitude coding

    • supplemental alphanumeric or symbolic codes or subcodes applied to existing coded data or a category to indicate their intensity, frequency, direction, presence, or evaluative content

    • can be qualitative, quantitative, and/or nominal

      indicators to enhance description​

    • appropriate for mixed methods and qualitative studies

      that also support quantitative measures as evidence of

      outcomes

    • Example in a school improvement project:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Grammatical methods to coding

15. Subcoding

    • a second-order tag assigned after a primary code to detail or enrich the entry

    • can be employed after an initial general coding scheme and the researcher realises that it's too broad, or particular relationships emerge

    • appropriate when general code entries will later require more extensive indexing, categorizing, and subcategorizing into hierarchies or taxonomies, or for nuanced qualitative data analysis

    • Example from a set of fieldnotes describing school facilities:

​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Grammatical methods to coding

16. Simultaneous coding

    • application of two or more different codes to a single qualitative datum, or the overlapped occurrence of two or more codes applied to sequential units of qualitative data

    • appropriate when when the data’s content suggests multiple meanings (e.g., descriptively and inferentially) that necessitate and justify more than one code

    • Example from fieldnotes about a community theatre programme: ​

​

TSLB3143

Fundamentals of Research in Education

Multiple Choice

Are you allowed to use more than one method during coding?

• Yes

• No

Fill in the Blank

Which method is used for the following code?

ANXIETY

Fill in the Blank

Which method is used for the following code?

"HURT SOMEONE BAD"

Fill in the Blank

Which method is used for the following code?

SMELLING NEW THINGS

Fill in the Blank

Which method is used for the following code?

MEMORIES OF SMOKING

Analysis and interpretation

Steps in coding

1. Create codes:

   1. deductive coding = ​developing a provisional "start list" of codes prior to fieldwork - comes from conceptual framework, RQs, key variables

   2. inductive coding = other codes emerge progressively during data collection - better grounded empirically

2. Revise the codes

3. Ensure ​conceptual and structural unity in code lists:

   • ​Not a catalogue of disjointed descriptors, but a conceptual web that includes larger meanings and their constitutive characteristics

4. Operationally define the codes

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

TSLB3143

Fundamentals of Research in Education

​Levels of coding details:

• ​depending on the aim of the study

• ​too much simultaneous coding suggests unclear / incomplete coding system

• ​not every portion must be coded - materials unrelated to RQ, trivial

Analysis and interpretation

• First cycle coding = ​a way to initially summarize segments of data

• ​Second cycle coding = pattern coding = a way of grouping the summaries from first cycle into a smaller number of categories, themes or constructs

  • ​explanatory / inferential codes

  • identify an emergent theme, configuration, or explanation

  • four functions:

    • condenses large amounts of data into smaller number of units

    • gets researcher into analysis during data collection

    • help researcher to elaborate a cognitive map

    • lays the groundwork for cross-case analysis by surfacing common themes and directional processes

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

What pattern codes look like:

1. ​Categories or themes

​

​

2. Causes / explanations

​

​

3. ​Relationships among people

​

4. Theoretical constructs​

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Steps in pattern coding​

1. Emerge from repeatedly observed behaviours, actions, norms, routines, and relationships; local meanings and explanations; commonsense explanations and more conceptual ones; inferential clusters and metaphorical ones; single-case and cross-case observations

2. ​Map the pattern codes visually in a network display to see how the components interconnect

3. ​Qualify the pattern codes by specifying the conditions under which it holds

4. ​Create subcodes for general pattern codes

5. Check out the pattern codes in the next​ wave of data collection, engaging in if-then tactic or rival explanation

TSLB3143

Fundamentals of Research in Education

Open Ended

The following is a selected series of codes related to the first month of withdrawal symptoms described by a participant in a smoking cessation treatment program. What are some possible pattern codes that emerge in the second cycle coding?

Open Ended

Cluster 1: ANXIETY, NERVOUSNESS, “HURT SOMEONE BAD”, ANGRY

Cluster 2: DEEP BREATHING, THROAT BURNING, “EATING A LOT MORE,” SMELLING NEW THINGS

Cluster 3: WANDERING AROUND, HABITUAL MOVEMENTS, RESTLESSNESS

Cluster 4: “FELT LIKE CRYING”, MEMORIES OF SMOKING

What do the codes in each cluster have in common? Name the possible pattern codes for the above clusters of first cycle codes.

Analysis and interpretation

Approaches to analyse the pattern codes

1. Narrative description

   • ​identify and elaborate on the pattern code, weaving its component first cycle codes into the narrative and supporting it with field notes:

   • story-line function of narrative enables researcher to outline the plots of human action and how participants changed throughout the course of the study

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Approaches to analyse the pattern codes

2. Matrix display

   • organise the vast array of condensed material into an "at-a-glance" format for reflection, verification, conclusion drawing, etc.

   • ​pattern codes in one column and the related first cycle codes in another

TSLB3143

Fundamentals of Research in Education

Analysis and interpretation

Approaches to analyse the pattern codes

3. Network display

   • codes become possible labels for bins

   • lines and arrows indicate connections and flows

TSLB3143

Fundamentals of Research in Education

Within-case analysis vs cross-case analysis

• Primary goal of within-case analysis: to describe, understand, and explain what has happened in a single, bounded context

• ​Purposes of cross-case analysis:

  • ​to enhance generalizability or transferability to other contexts

  • to deepen understanding and explanation

• Strategies for cross-case analysis:

  1. ​case-oriented strategies = a theoretical framework is used to study one case in depth, and then successive cases are examined to see whether the pattern found matches that in previous case

  2. variable-oriented strategies = look for themes that cut across cases

  3. mixed strategies = combine / integrate both strategies

TSLB3143

Fundamentals of Research in Education

Content Analysis

• ​to determine the presence of certain words, themes or concepts within some given qualitative data

• ​Types of content analysis:

  1. ​Conceptual analysis

     • to examine the occurrence of selected terms in the data

     • Steps:​

       1. Decide the level of analysis: word, phrase, sentence, themes

       2. Decide the number of concepts to code for

       3. Decide whether to code for existence or frequency of a concept

       4. Decide on how you will distinguish among concepts

       5. Develop rules for coding

       6. Code the text and analyse

TSLB3143

Fundamentals of Research in Education

Content Analysis

• ​Types of content analysis:

  2. ​Relational analysis

  • to explore the relationships between concepts

  • Steps:​

    1. Determine the type of analysis:

       1. affect extraction: emotional evaluation of concepts

       2. proximity analysis: evaluation of co-occurrence of concepts

    2. Code for existence of concepts

    3. Explore the relationship between concepts: strength, sign, direction

    4. Code the relationships

    5. Map out representations

TSLB3143

Fundamentals of Research in Education

Thematic Analysis

• to identify common themes that come up repeatedly

• Approaches:

  1. Inductive = allowing the data to determine the themes

  2. Deductive = applying preconceived themes based on theory to data

  3. Semantic = analysing explicit content of the data

  4. Latent = reading into the subtext and assumptions underlying the data

• Steps:

  1. ​Familiarisation

  2. Coding

  3. Generating themes

  4. Reviewing themes

  5. Defining and naming themes

  6. Writing up

TSLB3143

Fundamentals of Research in Education

Poll

How do you feel about the lectures on this topic?

Open Ended

Q&A / Reflection / Issue

Please ask at least ONE question.

OR Tell us your thoughts on what you have learned these two weeks.

OR State ONE issue/difficulty that you have faced during these two weeks' lectures.