Qualitative Analysis

James Neill

Contents

• Quote
• Overview
• Research questions and hypotheses
• Data familiarisation
• Data coding
• Descriptives
• Motivation notes
• Satisfaction notes
• Writing up
• See also

Quote

"[Qualitative] data analysis is the process of bringing order, structure and meaning to the mass of collected data. It is a messy, ambiguous, time-consuming, creative, and fascinating process. It does not proceed in a linear fashion; it is not neat. Qualitative data analysis is a search for general statements about relationships among categories of data."

Marshall and Rossman, 1990, p. 111

Overview

• Provide a basic analysis of some of the qualitative data (i.e., for at least one of the qualitative questions).
• The analysis you conduct should be relevant to your overall research question; ideally, the qualitative analysis will complement and provide insight into the quantitative results.
• There are no hard and fast rules for the methodological approach to use, however, the approach should be systematic and transparent (clearly explained).
• The basic methodological options are to either treat and analyse the data in a purely qualitative way (e.g., to summarise themes) or to code the responses quantitatively and provide descriptive statistics, graphs, and possibly also inferential testing.
• Also see more information in Lecture 9.

Research questions and hypotheses

• If your approach is mainly exploratory or descriptive, then the qualitative analysis probably should be guided by a research question, e.g.,.
  • "What aspects of the university are students most and least satisfied with?"
• If your approach is mainly explanatory or predictive, then the qualitative analysis probably should be guided by a research question and/or one or more hypotheses, e.g.,.
  • RQ: "What areas of satisfaction and/or dissatisfaction are identified by students?"
  • H₁: "The areas of satisfaction and/or dissatisfaction identified by students will match the three factors in the quantitative student satisfaction measure (Education/Teaching, Campus, Social)."
• See also: Lab report: Research questions and hypotheses

Data familiarisation

• In qualitative analysis, the researcher's subjective judgement becomes part of the research instrument.  Instead of doing upfront work in pre-creating certain categories of response (i.e., as in quantitative research), qualitative research involves creating codings and categories post-hoc.
• The first step is to familiarise yourself with the qualitative data by reading through the responses, then other decisions can be made.
  • Read through the responses on the screen in spreadsheet format.
  • Its easier to read if:
    • sorted [Data - Sort Cases]
    • drag column wider

• Read through the responses - let them soak in - immerse yourself in the data.

Data coding

• Brainstorm a list of the main themes in the data.
• Optional: Copy the responses into a word processing document for sorting and (highlight the data column, then copy and then paste into a word processing document).
• From the word document you can print out or develop codings in the word document (e.g., use highlighter to categorise).
• there's multiple ways the responses could be categorised - choose codes relevant to the research question.
• e.g., the example data on degree enrolled in could be coded by:
  • Level: Pre-undergraduate / Undergraduate / Postgraduate
  • Divisions: BLIS / ComEdu / HDS
• For more subjective open-ended data, such as comments about motivations and areas of satisfaction and dissatisfaction, it will take some time to develop the codes (factors).
• Insert a new numeric variable in your data file, next to the qualitative variable and then enter the relevant numeric codings for each case.
• Add Value Labels, then toggle on Label view.

Descriptives

• Conduct frequency analysis of the numeric codings, and possibly also a bar graph, e.g.,

• Break down the descriptives by any other variables of interest, e.g.,

• Also report percentages and, if you're particularly interested in differences between category codings, then conduct a chi-square test.
• For subjective data, develop a table and /or text which provides a rich / thick description of the meaning of each category, e.g.,
  • Definition
  • Example, typical comments (central tendency)
  • Example, extreme comments (e.g., from best to worst)

Motivation notes

• The question is:
  • What have been the main motivations for you to attend university? Please describe up to 3 key reasons for why you’ve chosen to be a university student?
• The up to three three open-ended motivation responses probably should be considered as one set of answers (i.e., combine the responses into one variable).
• Alternatively, you could conduct a multiple response analysis (described in Lecture 9).
• One way to combine the three variables is to copy all the motivation responses into one column in a new data file.

Satisfaction notes

• The questions are:
  • Please describe what aspect of this university have you been most satisfied with?
  • Please describe what aspect of this university have you been least satisfied with?
• Analysis could focus on responses to one or both of the questions.
• If analysing both questions, this could be done as:
  • Two separate analyses (which could then possibly be compared)
  • A multiple response analysis (described in Lecture 9)

Writing up

There is no strict guideline on how to write up a qualitative analysis.  This is an opportunity to impose your own discipline, structure and focus.  You may consider addressing some of the following:

• Purpose of the analysis, including an underlying research question and/or hypothesis(es) to guide the analysis.
• Approach taken (e.g., qualitative content/thematic analysis, or quantitative coding and statistical analysis) and methodological steps followed
• Descriptive information about the written data, including the extracted themes, the nature of those themes, and possible interrelationships between the themes
• Multiple response analysis
• Break-down of themes by IVs of interest (e.g., Gender)
• Frequencies, percentages, and tests of relationships/differences for coded data
• Tables and possible a Figure to summarise results (note: A figure could be a graph, but it also could be a conceptual diagram showing the main constructs and possible relationships between the constructs; visualisation can be used in many ways to synthesise understandings which emerge from qualitative analysis)
• In describing and summarising the data, it can be helpful to use illustrative quotes to illustrate, for example, common/prototypical responses or a range of responses (rather from "..." to "...").
• If using the analysis to help 'validate' the quantitative instrumentation, provide comparisons and commentary about the extent of overlap and 'validation', as well as gaps and suggest possible factors and measurement items which could be introduced into the quantitative measure to improve its coverage of students' salient areas of motivation and satisfaction.
• Strengths of the analysis - e.g., what unique insight has been contributed?
• Limitations of the analysis - e.g.,  what debatable decisions were made?  How else could the analysis have been tackled?  Is your analysis 'replicable'?  How could the open-ended survey questions be improved?

See also

• A flow chart for the typical process of content analysis research
• A rough guide to analysing qualitative data: Marking up (pdf)
• Content analysis
• Content analysis
• Content analysis: What it is and where it is used
• Presenting qualitative data
• Qualitative content analysis
• Qualitative data analysis
• The content analysis guidebook