What is qualitative data analysis? Different approaches to analysing qualitative data.

What is qualitative data analysis?Different approaches to analysing qualitative data

By the end of the session you will…

• Have an understanding of what qualitative data sets look like

and the common forms in which they are presented

• Have an understanding of the different data analysis methods

that are available

• Be able to prepare, manage and code your data, identify themes

and make recommendations

• Understand the issues surrounding validity and reliability when

analysing qualitative data sets

• Know how to produce a qualitative analysis data report based

on your findings

What is qualitative research?

‘Qualitative Data Analysis (QDA) is the range of processes and procedures whereby we move

from the qualitative data that have been collected, into some form of explanation,

understanding or interpretation of the people and situations we are investigating’.

QDA is usually based on an interpretative philosophy. The idea is to examine the meaningful and symbolic content of qualitative data. By analysing interview data, the researcher may be attempting to identify any or all of:

•Someone's interpretation of the world, •Why they have that point of view, •How they came to that view, •What they have been doing, •How they conveyed their view of their situation, •How they identify or classify themselves and others in what they say,

Types of Qualitative data sets

• Semiotic analysis: Analysing signs and symbols (e.g. A participant captures an image and is filmed whilst talking through what the image means to them– it is important to note that the analysis is of the picture and its meaning in a particular social context)

• Thematic analysis – An account from the participant is captured by an interviewer driving the conversation through asking questions.

• Conversational analysis – capturing a conversation between two people without any prompts.

• Narrative analysis – A deep analysis of the dimensions of social life. This is one person giving an account of an issue, similar to ‘story telling’.

You can apply different analytic methods to a data set depending on how the data has been collected.

Framework analysis

You have a pre-existing research question(s) and you are looking to prove this through analysing the data set.  

VS

Grounded theory

Analyse the data set looking for themes derived from the content and make a decision as to whether 1 theme is worth exploring more than other. Grounded theory is the basis of thematic analysis and state that the theory emerges out of the data.

Different analysis methods

Principles of Grounded theory

‘Naming, defining and understanding what people say’

•Theory emerges out of the data rather than being developed in

advance

•Begin with as few pre-determined ideas as possible

•Coding is shaped by the researchers interpretation of the data

•Characterised by different types and levels of code

•Categories of data are created and refined through constant

comparison

•Analysis and data collection are iterative

•End point is theoretical saturation

Thematic analysis: moving between levels of codes

Things to consider before you start…

• Make field notes throughout the entire data collection and analysis process.

Reflecting on the process of analysis during your fieldwork is key.

• Be aware of researcher effects: For example, if you are analysing interview

data; did you put words in participant’s mouth? ‘Did that make you feel

guilty?’ – Avoid this!

• Important to consider your research design before you collect research as

this will make the analysing part easier (we will come back to this later!)

• Keep memos: They are useful to bridge the gap between the start of the

project and the end of the data analysis. You can find something completely

different at the end but memos allow you to keep evaluating what you are

doing and what you would do different if you were to run research again?

Validity & Reliability

Validity

How can we be sure we are measuring what our we set out to? To what extent are we describing and measuring what our research question set out?

As qualitative researchers we need to be able to account for how inferences were made in our analysis and our conclusions reached. This is the only way our findings are credible.

Create an analysis group – through having different people analysing the data set, your validity will increase.

Reliability

If we repeated the research, would we find the same results?

Objectivity needs to be embedded into the research process. Demonstrating reliability through keeping detailed records of the links between the claims you are making and the evidence on which it is based.

‘Careful Scholarship- Seale 2006’

Be aware of the boundaries of the claims that you make from analysing your qualitative data and remain transparent throughout your analysis about all of the processes that you have used.

Are there any instances that contradict the claim that you are making? Disruptions can exist and not contradict your argument. Be upfront about these anomalies and this will help your reliability.

Preparing and managing your data set

• Leave large margins for annotations

• Do not tidy the language – although it is

tempting to correct language and make it more

your own style…RESIST!

• Keep your field notes to hand when analysing

your data set.

Coding your data set

Formal definition

‘The identification of passages of text (or other meaningful phenomena, such as parts of images) and applying labels to them that indicate they are examples of some thematic idea.

At its simplest, this labelling or coding process enables researchers quickly to retrieve and collect together all the text and other data that they have associated with some thematic idea so that they can be examined together and different cases can be compared in that respect’.

Stage 1 : Initial coding

‘Define, understand and explain what is happening in the data’

1.Number each line in the data

2.Code every new concept – keep this basic and do not attribute a feeling to the code. Simple codes such as ‘bereavement’. To code ‘loss’ would come later because attributing an emotion here.

3.Code everything else, not just content in the data. For example, when the data was collected, where it was collected or if there were any external influences?

Stage 2 : Focussed coding

Synthesise the most significant/frequently used codes across cases to create categories/themes/conceptual similarities.

1.Move from initial descriptive codes, onto groups of codes. This is where ‘loss’ would become relevant.

2.Compare your data to see if the same codes are appearing.

Stage 3 : Selective/theoretical coding

For example, if you had some initial codes of Money, Income, Receipts. The selective code could be Finance.

Ensure you hold coding meetings. Having more than 1 person help with coding adds creativity and validity to your analysis.

Keep a ‘coding book’ throughout the entire coding process. Include this in your appendix when writing your report.

Example…

‘X’ is an overall outstanding lecturer. One, if not the ONLY lecturer on my course that I feel has my best interests at heart and takes an interest in her students. ‘X’ is always networking and engaging with PR practitioners and agencies to ensure we get the best chances of securing Work Experience or future job vacancies. ‘X’ always encourages us to raise our chances for employability and helps us with this through her own personal contacts as well as arranging regular CIPR networking events - dedicating her time outside of class. ‘X’ is fair and organised - which the whole class appreciates - and she is able to communicate her messages well. She explains why we are being taught a certain topic and how this will add to either our employability or personal development. She is always looking for new ways to enhance these two areas, and often drafts in live clients and professionals - which enables us to experience real life simulations, as well as development of our abilities. I could have nominated ‘X’ for 'Open door', 'outstanding feedback' or 'strongest supervision' because she really is a fantastic influence. I think ‘X’ is a credit to Solent and I know I would have dropped out in my second year, if it wasn’t for her support.

Stage 1: Initial coding

1. Outstanding

2. Best interest

3. Takes an interest

4. Networking agencies

5. Securing work experience

6. Raises chances of employability

7. Helps make new contacts

8. Arranges networking events

9. Dedicates time outside class

10. Organised

11. Communicates well

12. Explains why

13. Personal development

14. Develops abilities

15. Fantastic influence

16. Credit to university

17. Would have dropped out if not for lecturer

Stage 2&3…

Stage 2: Focussed coding

Code 1: Organisation/style of teaching: 8, 10, 11, 12, 14, 3, 7, and 2

Code 2: Skills: 1, 5, 7, 10, 11, 12, 14, and 17

Code 3: Outside of classroom behaviour: 4, 5, 6, 7, 8, 9, 17, and 13

Code 4: Employability opportunities: 4, 5, 6, 7, and 8

Code 5: Admiration of lecturer: 15, 16

Stage 3: Selective coding

Improving employability prospects (Code 4)

Adopting a style of teaching that moves beyond standard academic practices/teaching (Code 1, 2, 3)

Admiration/Respect for lecturer (Code 5)

Theoretical saturation

This is where you reach a point where no new information is coming out of the data. Be realistic about how much time and resource you have and stop analysis at a sensible point.

Writing up your findings

What have you found? In many cases what you write may be analytic ideas. In other cases it may be some form of précis or summary of the data,

though this usually contains some analytic ideas.Make it your own!

Ensure you evidence reliability and validity and include your coding book in your appendix

Thank you, any questions?