Qualitative Data Analysis Methods

Qualitative Analysis Challenges

• There are no universal rules for qualitative analysis.
• A voluminous amount of narrative data requires lots of intensive work.
• Strong inductive powers and creativity are needed.
• Qualitative analysis involves condensing rich data into concise reports.

Qualitative Data Management and Organization

• Data must be transcribed.
• A coding scheme must be developed.
• Qualitative data must be coded.
• Data can be organized manually using conceptual files
• Data can be organized using Computerized methods of organization using CAQDAS

A General Analytic Overview

• Identification of themes or broad categories.
• Search for patterns among themes, variations in the data.
• Development of charting devices and timelines.
• Use of metaphors to evoke a visual analogy.
• Validation of themes and patterns.
• Weaving thematic pieces into an integrated whole.

Content Analysis Steps

• Analyze narrative data content to identify prominent themes and patterns among them.
• Break down data into smaller units.
• Code and name units in line with their content.
• Group coded material based on shared content.

Ethnographic Analysis: Spradley's 12-Step Method

• Locating an informant.
• Interviewing the informant.
• Making ethnographic records.
• Asking descriptive questions.
• Analyzing ethnographic interviews.
• Making a domain analysis as a first level of analysis.
• Asking structural questions.
• Making a taxonomic analysis (second level).
• Asking contrast questions.
• Making a componential analysis (third level).
• Discovering cultural themes - theme analysis (fourth level)
• Writing the ethnography.

Domain Analysis

• Domains are broad categories representing units of cultural knowledge.
• Ethnographers identify relational patterns among terms in the domains used by culture members.
• Ethnographers focus on the cultural meaning of terms/symbols and their interrelationships.

Taxonomic Analysis

• It is the second level in Spradley's (1979) data analytic method.
• Ethnographers decide how many domains the analysis will include.
• Depending on the number of domains for in-depth or less intensive study, a taxonomy is made.
• Taxonomies are systems of classifying and organizing terms, developed to illustrate a domain's internal organization.

Componential Analysis

• Multiple relationships among terms in the domains are examined.
• Analysis focuses on similarities/differences among cultural terms in a domain.

Theme Analysis

• Involves uncovering cultural themes.
• Domains are connected in cultural themes, providing a holistic view of culture.
• Cultural meaning is the outcome of this analysis.

Van Manen's Phenomenological Method: Six Activities

• Turning to the nature of the lived experience.
• Exploring the experience as we live it.
• Reflecting on essential themes.
• Describing the phenomenon through writing and rewriting.
• Maintaining a strong relation to the phenomenon.
• Balancing the research context by considering parts and a whole.

Three broad schools of Phenomenology

• Duquesne School (descriptive phenomenology)
• Colaizzi
• Giorgi
• Van Kaam
• Utrecht School (descriptive and interpretive phenomenology)
• Van Manen
• Heideggerian hermeneutics (interpretive).
• Gadamer
• Diekelmann, Allen, and Tanner
• Benner

Benner's Hermeneutic Analysis

• Search for paradigm cases.
• Thematic analysis.
• Analysis of exemplars.

Grounded Theory Analysis

• Employs a constant comparative method of analysis.
• Features two competing grounded theory strategies:
• Glaser and Strauss (Glaserian)
• Strauss and Corbin (Straussian)

Coding: Glaserian Approach

• Substantive codes:
• Open codes identifies a core category
• Open codes are used in the first stage of constant comparison
• Breaking down data to examine similarities and differences.
• One type of core category is a basic social process (BSP).
• Level I (in vivo) codes, Level II codes, and Level III codes (theoretical constructs).
• Selective codes: These relate only to the core category.
• Theoretical codes:
• Relate to how substantive codes relate to one another.

Examples of Families of Theoretical Codes (Glaser)

• Process: Includes stages, phases, passages, and transitions.
• Strategy: Encompasses tactics, techniques, and maneuverings.
• Cutting point: Involves boundaries and turning points.
• The six Cs: Includes causes, contexts, conditions, contingencies, consequences, and covariances.
• There are 18 families of codes in this approach.

Strauss and Corbin's Method of Grounded Theory

• Open coding: Data broken into parts and concepts based on interpreted meaning of raw data.
• Axial coding: used for context, locating/linking action-interaction within a framework of sub concepts to explain what, why, and consequences of interactions.
• Selective coding: Involves deciding on the central (or core) category.

Constructivist Grounded Theory Approach

• Theories include researchers' experience and involvements.
• Initial coding: studying data to learn what participants view as problematic.
• Focused coding: identify most significant initial code and then theoretically code.

Debates About Rigor and Validity in Qualitative Research

• Focuses on quality-related goals and terminology in qualitative research.
• A major dispute is about whether "validity" and "rigor" are appropriate terms.
• Some reject validity/rigor, some accept them, others seek parallel goals.

Terminology Proliferation and Confusion

• There is no common vocabulary.
• Terms used: goodness, truth value, integrity, trustworthiness, validity and rigor.

Controversies in Qualitative Research

• Some frameworks/criteria aim to be generic across qualitative traditions.
• Other frameworks are specific to a tradition or analytic approach within it.

Lincoln and Guba's Framework

• A widely cited "gold standard".
• The key goal is trustworthiness where it:
• Concerns the "truth value" of data, analysis, and interpretation.
• A parallel perspective, with analogs to quantitative criteria.
• Encompasses four criteria:
• Credibility
• Dependability
• Confirmability
• Transferability

Lincoln and Guba's: Credibility

• Enhancing the findings' believability and demonstrating credibility to external readers.
• Refers to confidence in the truth of the data and interpretations.
• It is the analog of internal validity in quantitative research.
• It is arguably the most important criterion for assessing a qualitative inquiry's quality and integrity.

Dependability

• Refers to stability of data over time and conditions.
• The analog of reliability in quantitative research.

Confirmability

• Refers to objectivity, with congruence between independent people about data accuracy and meaning.
• The analog of objectivity in quantitative research; data represent participant information and interpretations.

Transferability

• The extent to which qualitative findings can be transferred to other settings or groups.
• Analog of generalizability or external validity in quantitative research.

Authenticity

• Researchers fairly and faithfully show a range of different realities.
• Convey the feeling/tone of participants' lives as they are lived.
• No analog in quantitative research.
• Added to the Lincoln and Guba's framework.

Strategies to Enhance Quality in Qualitative Inquiry

• Researchers can take many steps to enhance the quality of their inquiries.
• Consumers can assess the quality-enhancement efforts by looking for steps and assessing strengthening integrity/validity/trustworthiness.

Strategies During Data Collection

• Prolonged engagement: investing sufficient time to have in-depth understanding.
• Persistent observation: intensive focus on salience of data being gathered.
• Reflexivity strategies: attending to researchers with an affect on the data.
• Comprehensive and vivid recording of information.
• Maintenance of an audit trail.
  • A systematic collection of documentation/materials and a decision trail.
• Member checking:
• Providing feedback to participants about emerging interpretations and obtaining their reactions is controversial.

Data and Method Triangulation: Denzin

• Data triangulation: Use multiple data sources to validate conclusions (time, space, person triangulation).
• Investigator triangulation: Not relevant to data collection.
• Method triangulation: Use multiple data collection methods to study the same phenomenon.
• Theory triangulation: Not relevant to data collection.

Strategies Relating to Coding and Analysis

• Search for disconfirming evidence as the analysis proceeds.
• Includes purposive/theoretical sampling of cases that can challenge interpretations.
• Negative case analysis: a specific search for cases that appear to discredit earlier hypotheses
• Peer review and debriefing:
  • Sessions with peers specifically designed to elicit critical feedback.
• Inquiry audit: a formal scrutiny of data and relevant supporting documents and decisions by an external reviewer.

Strategies Relating to Presentation

• Thick and contextualized description: vivid portrayal of study participants
• Researcher credibility: enhancing confidence by sharing relevant aspects of researcher's experience, credentials, motivation, context, and the phenomenon in the study.

Interpretation of Qualitative Findings

• Interpretation relies on adequate "incubation". --The process of living the data
• Interpretation shares issues with quantitative research:
  • Issues include credibility, meaning, importance, and transferability.