1: 06-10-2022 - Analysing qualitative data
Circle of academic life: idea - concepts and theories - research question and study questions -
methodology and method - data gathering - data analysis.
Qualitative data analysis = understanding the meaning of data by making thick descriptions of
participants’ stories and reconstructing patterns of thinking and acting.
→ story telling: rigorous, logical process through which data is given meaning → items: coding
(labeling, providing structure, searching for patterns) -> thematic structuring -> theorising.
→ types of qualitative data: texts, interview transcripts, reports, articles, videos, observations.
Assumptions: (a) reality is socially constructed (by participants). (b) emic = insider’s point of view.
(c) variables and concepts are compex, interwoven and difficult to measure. (d) the researcher is
his/her own instrument. (e) no standardised procedures. (f) personal involvement and partiality.
(g) empathic understanding.
Characteristics qualitative researcher:
- Reflexive awareness: ability to (a) think abstractly, (b) step back and critically analyse
situations, and (c) recognise the tendency towards bias.
→ 2 forms of reflexivity: (1) epistemological (researchers reflect on their assumptions of the world
and the nature of knowledge), and (2) personal (researchers reflect on how personal values,
attitudes, beliefs and aims have shaped the research).
→ achieving reflexivity by: (a) involving multiple investigators, (b) writing a reflexive journal, (c)
reporting research perspectives, values and beliefs in any research report.
- Openness: (a) be flexible and open to criticism, (b) appreciative inquiry (interviewee feels
more empowered after an interview, because of structure and building rapport).
→ achieving building rapport by: check your appearance - remember the basics of good
communication - find common ground - create shared experiences - be empathic - mirror and
match (from perspective of the respondent).
- Sensitivity: sensitive to words, interpretations and actions of respondents.
Qualitative analysis principles: describe, classify, connect → (1) noticing concepts, (2) collecting
examples of these concepts, and (3) analysing these concepts to find commonalities.
Different types of analysis:
- Content analysis = making inferences about data by systematically and objectively
identifying special characteristics (categories) within them (may contain quantitative
elements) → more descriptive: a priori defined codes/categories (structured).
→ deductive research: testing theory and improving it through structured interviews = interview
themes based on a theoretical model → pitfall; too much fitting in existing boxes and looking for
evidence for a relation (closed mind, decontextualisation of data).
- Thematic analysis = (descriptive) method for identifying, analysing and reporting
patterns (themes) within data → middle way between C and G → inductive (themes
emerge from data) vs. theoretical (themes emerge from researcher’s theoretical stance).
- Grounded theory = construction of theory through the open analysis of data → open to
interpretation (unstructured) → not appropriate for 5 months internship.
→ inductive research: designing new theories → data collection and analysis are interrelated,
iterative coding process (continuous improvement) → analytical concepts, connections and
categories emerge → although open to the context, systematic procedures are important (coding
process) → pitfall; too free in accepting all concepts (messy code book) and accepting too vague
relations → there are no clear boundaries between the types of analysis.
→ other types of analysis: narrative analysis (capturing lived experiences), conversational
analysis (everyday conversations), discourse analysis (use of language in social contexts).
, Codes = words/short phrases that represent the essence of narrative/verbal info.
→ properties: (a) categorise data, (b) organise data into segments, (c) develop into a coding guide.
→ coding guide = compilation of emerging codes, brief definitions or properties for each code
(can also include illustrative coded pieces/quotes) → will evolve throughout the analysis →
function: provides guidance for when and how to use the codes.
→ quotes are categorised in subcodes and eventually overarching themes/codes.
Quotes = support analysis and findings, are illustrative, represent a range of issues and opposing
views (between stakeholders) → have to ensure anonymity.
Steps of qualitative analysis:
1. Data curation: transcribe the interviews, field notes, etc → starts right after collection.
2. Familiarisation: collect –> code –> collect –> code → get familiar with the data.
3. Interpretation: read and re-read, suspend your initial interpretation → focussed reading
and open coding (codes arise while reading the data) → use both open and closed coding.
4. Close examination: label the text with keywords → reviewing and axial coding (connect
open codes in categories).
5. Modification: remove duplicates and modify codes, provide an hierarchical order and
integrate theory to generate theory.
6. Patternising: looking for connections or patterns that emerge from the data.
Vertical analysis = understanding the essence of individual interviews or other data (narrative)
→ acquire the unique line of arguing of individuals and their priorities → function: find links and
hierarchies between codes/themes (find the perspective of every individual interviewee).
Horizontal analysis = aggregation and comparison of the content
of data across different interviews or other data → pay attention
to diversity (majority and minority of views count) → function:
find the general perspective, overview of topics and opinions →
typical for content analysis → apply both horizontal and vertical.
→ disadvantages: (a) reduces narratives/individuals to parts, (b)
may lead to overemphasising of elements (comments are
decontextualised), (c) may overlook conflicting remarks within
interviews, (d) neglects the unique line of arguing of individuals.
3 levels of coding:
Open identifying concepts and their properties/ done by; comparing text fragments on similarities
coding dimensions → disaggregation of data into or differences and labeling them with keywords.
units (sensitised) (vertical focus).
Axial relating categories to subcategories (at level of done by; examining a phenomenon, identifying
coding properties/dimensions) → identifying underlying patterns and linking categories.
relationships (horizontal focus).
Selective integrating and refining theory → defining done by; integrating core categories into theory,
coding core concepts/categories and formulating the finding a storyline around core categories and
essence of key concepts. validating relationships (re-iterating).
→ FEX. (1) discovering the vague concept of “hopping”, (2) finding out properties of hopping, (3)
discovering hopping is part of a student subculture.
Validity: whether the researcher is observing and identifying what he/she claims.
Circle of academic life: idea - concepts and theories - research question and study questions -
methodology and method - data gathering - data analysis.
Qualitative data analysis = understanding the meaning of data by making thick descriptions of
participants’ stories and reconstructing patterns of thinking and acting.
→ story telling: rigorous, logical process through which data is given meaning → items: coding
(labeling, providing structure, searching for patterns) -> thematic structuring -> theorising.
→ types of qualitative data: texts, interview transcripts, reports, articles, videos, observations.
Assumptions: (a) reality is socially constructed (by participants). (b) emic = insider’s point of view.
(c) variables and concepts are compex, interwoven and difficult to measure. (d) the researcher is
his/her own instrument. (e) no standardised procedures. (f) personal involvement and partiality.
(g) empathic understanding.
Characteristics qualitative researcher:
- Reflexive awareness: ability to (a) think abstractly, (b) step back and critically analyse
situations, and (c) recognise the tendency towards bias.
→ 2 forms of reflexivity: (1) epistemological (researchers reflect on their assumptions of the world
and the nature of knowledge), and (2) personal (researchers reflect on how personal values,
attitudes, beliefs and aims have shaped the research).
→ achieving reflexivity by: (a) involving multiple investigators, (b) writing a reflexive journal, (c)
reporting research perspectives, values and beliefs in any research report.
- Openness: (a) be flexible and open to criticism, (b) appreciative inquiry (interviewee feels
more empowered after an interview, because of structure and building rapport).
→ achieving building rapport by: check your appearance - remember the basics of good
communication - find common ground - create shared experiences - be empathic - mirror and
match (from perspective of the respondent).
- Sensitivity: sensitive to words, interpretations and actions of respondents.
Qualitative analysis principles: describe, classify, connect → (1) noticing concepts, (2) collecting
examples of these concepts, and (3) analysing these concepts to find commonalities.
Different types of analysis:
- Content analysis = making inferences about data by systematically and objectively
identifying special characteristics (categories) within them (may contain quantitative
elements) → more descriptive: a priori defined codes/categories (structured).
→ deductive research: testing theory and improving it through structured interviews = interview
themes based on a theoretical model → pitfall; too much fitting in existing boxes and looking for
evidence for a relation (closed mind, decontextualisation of data).
- Thematic analysis = (descriptive) method for identifying, analysing and reporting
patterns (themes) within data → middle way between C and G → inductive (themes
emerge from data) vs. theoretical (themes emerge from researcher’s theoretical stance).
- Grounded theory = construction of theory through the open analysis of data → open to
interpretation (unstructured) → not appropriate for 5 months internship.
→ inductive research: designing new theories → data collection and analysis are interrelated,
iterative coding process (continuous improvement) → analytical concepts, connections and
categories emerge → although open to the context, systematic procedures are important (coding
process) → pitfall; too free in accepting all concepts (messy code book) and accepting too vague
relations → there are no clear boundaries between the types of analysis.
→ other types of analysis: narrative analysis (capturing lived experiences), conversational
analysis (everyday conversations), discourse analysis (use of language in social contexts).
, Codes = words/short phrases that represent the essence of narrative/verbal info.
→ properties: (a) categorise data, (b) organise data into segments, (c) develop into a coding guide.
→ coding guide = compilation of emerging codes, brief definitions or properties for each code
(can also include illustrative coded pieces/quotes) → will evolve throughout the analysis →
function: provides guidance for when and how to use the codes.
→ quotes are categorised in subcodes and eventually overarching themes/codes.
Quotes = support analysis and findings, are illustrative, represent a range of issues and opposing
views (between stakeholders) → have to ensure anonymity.
Steps of qualitative analysis:
1. Data curation: transcribe the interviews, field notes, etc → starts right after collection.
2. Familiarisation: collect –> code –> collect –> code → get familiar with the data.
3. Interpretation: read and re-read, suspend your initial interpretation → focussed reading
and open coding (codes arise while reading the data) → use both open and closed coding.
4. Close examination: label the text with keywords → reviewing and axial coding (connect
open codes in categories).
5. Modification: remove duplicates and modify codes, provide an hierarchical order and
integrate theory to generate theory.
6. Patternising: looking for connections or patterns that emerge from the data.
Vertical analysis = understanding the essence of individual interviews or other data (narrative)
→ acquire the unique line of arguing of individuals and their priorities → function: find links and
hierarchies between codes/themes (find the perspective of every individual interviewee).
Horizontal analysis = aggregation and comparison of the content
of data across different interviews or other data → pay attention
to diversity (majority and minority of views count) → function:
find the general perspective, overview of topics and opinions →
typical for content analysis → apply both horizontal and vertical.
→ disadvantages: (a) reduces narratives/individuals to parts, (b)
may lead to overemphasising of elements (comments are
decontextualised), (c) may overlook conflicting remarks within
interviews, (d) neglects the unique line of arguing of individuals.
3 levels of coding:
Open identifying concepts and their properties/ done by; comparing text fragments on similarities
coding dimensions → disaggregation of data into or differences and labeling them with keywords.
units (sensitised) (vertical focus).
Axial relating categories to subcategories (at level of done by; examining a phenomenon, identifying
coding properties/dimensions) → identifying underlying patterns and linking categories.
relationships (horizontal focus).
Selective integrating and refining theory → defining done by; integrating core categories into theory,
coding core concepts/categories and formulating the finding a storyline around core categories and
essence of key concepts. validating relationships (re-iterating).
→ FEX. (1) discovering the vague concept of “hopping”, (2) finding out properties of hopping, (3)
discovering hopping is part of a student subculture.
Validity: whether the researcher is observing and identifying what he/she claims.
