week 11-1: understanding the analysis of qualitative data & week 11-2: appraising trustworthiness and integrity in qualitative research

qualitative analysis

no universal rules for analyzing qualitative data

• no one way to do analysis → follow a general recommendation: need to read transcripts + incorporate field notes and come up with themes, talk to other researchers and use triangulation

very labor intensive + requires creativity

more difficult to do than quantitative but easier to understand

what is data and goal in qualitative research/analysis

often words; audio/video recorded and turned into transcript

distill/reduce the large amount of data into essential themes

(ex. using 2-5 focus groups to collect data and then reduce into few essential themes + use quotation as raw data)

(several passes; 1st pass just read, 2nd pass identify common words, 3rd pass identify themes and differentiating different themes)

qualitative data management + organization

developing a coding scheme → descriptive, process, concept, in vivo, holistic

coding qualitative data

organizing the data

• manual methods or organization (conceptual files)

• computerized methods or organizing using CAQDAS

descriptive coding

uses mainly nouns as codes and is often used by beginning qualitative researchers; does not provide much insight into meaning

ex. "the other day, we ran out of everything and we had to go to a church and get food.”; code: food pantry use

process coding

often using gerunds (-ing words) as codes to connote action and observation activity in the data

ex. "the other day, we ran out of everything and we had to go to a church and get food”; code: dealing with food shortages

concept coding

involves using a word/phrase to represent symbolically a broad meaning beyond observable facts/behaviors

usually nouns or gerunds

ex. "the other day, we ran out of everything and we had to go to a church and get food"; code: coping with the risk of hunger

in vivo coding

involves using participant-generated words and phrases; used as initial coding in many grounded theory studies

ex. "the other day, we ran out of everything and we had to go to a church and get food."; code: ran out of everything; had to go to church for food

holistic coding

involves using codes to grasp broad ideas in large chunks of data rather coding smaller segments

ex. I buy on deals. I learned how to, you know, what to buy and what not to buy. Where to shop, where to look for sales. I’ll go to all the stores. And I clip coupons from the paper and stuff. But sometimes that’s not enough. The other day, we ran out of everything and we had to go to a church and get food.”; code: food management strategies

coding qualitative data

once a coding scheme has been developed, the data are read in entirety and coded for correspondence to the categories

one paragraph → 3-4 different codes

researchers may have to modify the initial coding scheme

• new ideas for new codes

• must reread all previously coded material to see if need to apply new code

qual analytic procedures

puts segments together into meaningful conceptual patterns

1. identify broad categories

2. identify themes

3. some use metaphor as an analytic strategy

4. weave thematic pieces into an integrated whole

identify broad categories - qual analytic procedure

clusters of code that are connected conceptually

ex. I feel like i failed my patient; i let my patient down → can be clustered together to form a category

even though they are not using the same exact words, can connect them to a code

identify theme - qual analytic procedure

an abstract entity that brings meaning and identity to a current experience and its variant manifestation

are never universal (unique to the particular transcript or study that you are looking at)

how are they patterned

repeated over time and connect them → can become meaningful

researchers use metaphor as analytic strategy - qual analytic procedure

a symbolic comparison, using figurative language to evoke a visual analog

final analytic stage - qual analytic procedure

researchers weave the thematic pieces into an integrated whole

provide an overall structure to the data

qual content analysis

analyzing content of narrative data to identify prominent themes and patterns across them

breaking down data into smaller units

meaning units

smallest segment of text that contains a recognizable piece of information

manifest content

what the text actually says

latent content

interpretation of meaning

thematic analysis

Braun and Clarke state that __ is an accessible/theoretically flexible approach to analyzing qualitative data

seen as foundational method for qual analysis

6 phases

6 phases for thematic analysis

1. familiarizing oneself with the data

2. generating initial codes

3. searching for themes

4. reviewing those themes

5. defining and naming the themes

6. producing the report

ethnographic analysis

continually look for patterns in the behavior and thoughts of participants, comparing one pattern against another

use of maps, flowcharts, organizational charts, matrices (two dimensional displays)can help to highlight a comparison graphically and to discover emerging patterns

(how different groups interact with each other; identify implicit relationships between culture; looking for unspoken patterns and relationship)

spradley’s method (4 levels of data analysis)-ethnographic research

domain analysis

taxonomic analysis

componential analysis

theme analysis

(used for ethnographic analysis)

domain analysis (spradley's method)

broad categories that represent units of cultural knowledge

During this first level of analysis, ethnographers identify relational patterns among terms in the domains that are used by members of the culture

The ethnographer focuses on the cultural meaning of terms and symbols (objects and events) used in a culture and their interrelationships.

taxonomic analysis (spradley's method)

the second level

ethnographers decide how many domains the analysis will encompass. After making this decision, a taxonomy—a system of classifying and organizing terms—is developed to illustrate the internal organization of a domain. 

• do you want to include practicing vs retried nurses, outpatient vs inpatient nurses, day shift vs night shift nurses

componential analysis (spradley's method)

multiple relationships among terms in the domains are examined. The ethnographer analyzes data for similarities and differences among cultural terms in a domain. 

• do they have similarities across the different domain and taxonomic groups; are there differences in language and relationships 

theme analysis (spradley's analysis)

 cultural themes are uncovered. Domains are connected in cultural themes, which help to provide a holistic view of the culture being studied. The discovery of cultural meaning is the outcome.

descriptive phenomenology - phenomenological analysis

when people speak, you take them at face value and listen to what they say

descriptive and interpretive phenomenology - phenomenological analysis

have to incorporate own preconception/biases as you are listening to other people talk about their experiences

more common method of phenomenological analysis

benner’s hermeneutic analysis (interpretive) - phenomenological analysis

search for paradigm cases (typical of situation)

thematic analysis

analysis of exemplars

hermeneutics circle

revisiting things and thinking about them, how it makes you feel and then revisiting the things after knowing how it makes you feel

ex. qualitative research on pregnant smokers; did not feel comfortable disclosing that they were still smoking during pregnancy bc ppl would judge them and provide negative feedback; they would not be honest in their response bc they do not want to feel judge; as a researcher you can feel defensive learning about the pregnant women experience when you don't even know the pregnant women

__: thinking about why reading that made you upset, and then reread the section after reflecting on why you were feeling defensive

in qualitative research, researchers include that they felt defensive when reading the research

grounded theory analysis

key word: processes → studying people and how they manage certain process and how they get through/experience different events in their life

• ex. doing interviews and focus groups with people are who serving as caregivers for someone in the family with a terminal diagnosis → how do you cope? how to get through this major event? → trying to figure out commonalities to figure out a theory to help those in similar circumstances

developing theory from ground up

grounded theory analysis exmples

comparison of alternative grounded theory approaches

glaserian approach - grounded theory analysis

when the researcher are looking at transcript/interview and doing initial coding → they are looking at words phrases and themes and at some point they hope to identity basic social process

• basic social process: explains how people come to resolve a problem/concern

once they figure out BSP they go back to start and look at everything through the lens of BSP

BSP is the beginning of the theory and method to help people

coding - glaserian approach (grounded theory analysis)

substantive codes

• open codes — ends when core category is identified (when they do not know what they are looking for)

  • level I (in vivo), level II, level III

• one type of core category is a basic social process (BSP)

• selective codes — codes relating to core category

theoretical codes - how different codings relate to each other; helps researcher take broken pieces of data and lead them back together to look at the bigger picture

(open coding stops when BSP forms and go through selective coding)

glaser example of theoretical codes

glaser: why people behave the way they do

process: stages, phases, passages, transitions

strategy: tactics, techniques, maneuverings

cutting point: boundaries, turning points

6 Cs: causes, contexts, conditions, contingencies, consequences, covariances

glaser 6Cs example

problem being studied: we are studying a lack of student involvement for the big volunteer project that we have on campus

condition: students are not involved as we would like them to be

cause: lack of time, lack of interest

consequences: not enough people to se up this activity

contingencies: also thoughts of a backup/alternate plan → create a remote volunteer project for students who don't have enough time to travel back and forth between campus can actually find some time

covariance: students have a higher degree of workload that affect their time available for extracurriculars

strauss/corbins’ method of grounded theory

three types of coding

• open coding → data is broken down into parts and concepts and are identified for interpreted meaning of the raw data

• axial coding (link action to the concept of the BSP you come with)

• selective coding: deciding on central (core)category

(goal is to organize the data on the BSP)

constructivist grounded theory approach

theories include researchers’ experience and involvements → _{(it is not uncommon for researchers to write about themselves in the article; why they are studying this or why are they an expert)}

initial coding: data are studied to learn what participants view as problematic

focused coding: identify most significant initial code and then theoretically code

qual - rigor/validity debates

argument that qualitative researchers should return to terminology of social sciences

validity is an appropriate quality criterion in both qualitative/quantitative studies, although qualitative researchers use different methods to achieve it

absurdity of validity

qual - generic vs. specific standards

some frameworks and criteria aspire to being generic… that is applicable across qualitative traditions

other frameworks are specific to a tradition or even to a specific analytic approach within a tradition

terminology proliferation and confusion

no common vocab exists: goodness, truth value, integrity, trustworthiness, validity and rigor

lincoln and guba’s quality criteria → gold standard

suggested criteria for enhancing the trustworthiness of a qualitative inquiry: credibility, dependability, confirmability and transferability + authenticity

key goal = trustworthiness

concerns truth value

lincoln and guba - credibility

refers to confidence in truth value of data and interpretations of them

strive to establish confidence in truth of findings

2 aspects:

• carrying out study in a way that enhances the believability of the findings

• taking steps to demonstrate credibility to external readers

crucial criterion in qual research that has been proposed in several quality frameworks

(ex. as the researchers reading through transcript and telling you they have identified 4 themes, the __ is the confidence that the reader has of the 4 themes when compared to the raw data)

lincoln and guba - dependability

stability of data over time and over conditions

would the study findings be repeated if the inquiry were replicated with same participants in the same context

credibility cannot be attained in absence of dependability

lincoln and guba - confirmability

refers to objectivity

• the potential for congruence between 2+ independent people about data’s accuracy, relevance, or meaning

establishing that the data represent the information participants provided and that the interpretations of those data are not imagined by the inquirer

findings must reflect participants’ voice and conditions of inquiry, not researcher’s biases

(has to do with research team; multiple people need to read the transcript; reading them individually and coming up with own codes and then talk tog and discuss common themes → those that come up with the same theme reduce bias and shows objectivity)

lincoln and guba - transferability

analogous to generalizability

extent in which qual findings have applicability in other settings or groups

lincoln and guba - authenticity

conveys feeling tone of participants’ lives as they are lived

sense of mood, experience, language, context of those lives

no analog in quan research

(how well the researcher is writing/describing the results)

qual - quality enhancement strategies

prolonged engagement - investing sufficient time to have in-depth understanding

persistent observation - intensive focus on salience of data being gathered

reflexivity - attending to researcher’s effect on data (make sure the researcher understands their own effect in data)

comprehensive/vivid info recording

maintenance of audit trail and decision trail → actual representation of all the coding and notes during the interpretation; helpful when their is a challenge in the research

member checking - providing feedback to participants about emerging interpretations and obtaining their reactions (controversial)

• if info is not flattering to the participant, the participant can say a lot of untrue info → do not want their interpretation as the research can change

triangulation

the use of multiple referents to draw conclusions about what constitutes the truth

data triangulation

the use of multiple data sources for the purpose of validating conclusions

time and space triangulation

method triangulation

the use of multiple methods of data collection to study the same phenomenon

qual coding/analysis strategies

search for disconfirming evidence as the analysis proceeds, through purposive/theoretical sampling of cases that can challenge interpretations

peer review and debriefing

inquiry audit (formal scrutiny of data and relevant supporting documents and decisions by an external reviewer)

negative case analysis (qual coding/analysis strategies)

a specific search for cases that appear to discredit earlier hypotheses

qual presentation analysis

thick/contextualized description: vivid, portrayal of study participants, their context, phenomenon under study

researcher credibility: enhancing confidence by sharing relevant aspects of researcher’s experience, credentials, motivation

• ex. talk about their own experience → studying smt like caring for someone with terminal illness, and they are doing this researcher bc they are also a caregiver

qual findings interpretation

relies on adequate incubation (the process of living the data)

similar interpretative issues as in research (credibility, meaning, importance, transferability, implications)