Qualitative Data Analysis and Interpretation
Many research methods such as naturalistic observations, open ended questions on questionnaires, unstructured interviews, an analysis of participants’ diaries or notes as part of a case study and so on, may generate qualitative data (i.e. data in written form). Although this data can appear very detailed and complicated, one way of simplifying this data to make it easier to analyse and interpret is by converting it into quantitative data.
One way of converting qualitative data into quantitative data is through a content analysis.
Content analysis is a method of qualitative data analysis that involves rating or coding and categorising qualitative data. Rating systems are used to assess a wide range of qualitative data such as written or verbal observation records, questionnaire survey responses (where open-ended questions were employed) and interview content. The idea behind a coding system is to be able categorise occurrences of a particular theme, thereby converting the raw qualitative data into quantitative frequencies (i.e. a count is made of the number of times certain selected types of event happen).
The Procedure of Content Analysis
(1) A sample of materials are gathered (e.g. interview responses may need to be transcribed) and examined (i.e. by reading the text through several times until you know it well).
(2) Key themes are identified and these are used to create categories to help organise and classify the data- these categories will reflect the purpose of the research.
(3) Instances or examples of each theme occurring in the text are gathered and placed in the relevant categories.
(4) Frequencies can then be counted numerically for each of the different categories. The nominal data can then be quantitatively analysed with charts and graphs, measures of central tendency and dispersion produced and statistical analysis can be carried out.
(5) The researcher may then draw conclusions about human behaviour as a whole based on these findings.
A pilot study is often used as part of the analysis to generate and test the coding system to be employed. This helps make sure categories are exclusive and discrete (i.e. there is no overlap) and that all relevant themes are included and clearly operationalised.
Evaluation of using a Content Analysis:
(+) A strength of using a content analysis is that Statistical procedures become possible. When dealing with written/qualitative data it is not possible to use descriptive or inferential statistics. Converting qualitative data into quantitative data makes descriptive and inferential statistics possible which means that researchers are able to test for significance.
(+) Changing qualitative data into quantitative data makes the results more objective.
(-) Turning qualitative data into quantitative data can be seen to be reductionist. The complex and detailed qualitative data gets reduced to numerical figures which in turn reduces detailed/complex human behaviour to numbers and figures.
Although there are ways of converting qualitative data into quantitative data (such as content analysis), most qualitative researchers believe the traditional quantitative methods used by psychologists do not produce results that are applicable to everyday life. If a researcher is trying to produce numbers, then he or she is probably not engaged in qualitative analysis which is less concerned with counting responses or occurrences and more concerned with interpreting the meaning of data (i.e. quality rather than quantity). There are different ways of analysing qualitative data that preserve these principles and therefore do not aim to use numbers at all as part of the data analysis.
This involves taking a body of text (qualitative data) such as a transcript of an interview or an observational record to analyse an existing theory, explanation or hypothesis (from the top-down). This involves organising the qualitative data into specific themes that were pre-identified by the existing theory. This allows the researcher to summarise the data gathered into distinct categories. This kind of analysis is sometimes also referred to as ‘theoretical analysis’ as the theory and themes exist prior to analysis beginning. A similar process can be followed where there is no pre-existing theory and, instead, the themes or categories emerge from the data (from the bottom-up). This is referred to as ‘inductive analysis’ or ‘grounded theory’ and helps provide new insights. Either way, the analysis follows a similar series of stages.
Stages in Thematic Analysis
(1) Transcribe the data if you need to, number each line and read the text through several times until you know it well. As you read the text, make notes of any ideas that occur to you.
(2) Divide the text into ‘meaning units’ using a forward slash (/) between every apparent change in meaning or subject.
(3) Search the entire text for meanings that seem to have a similar theme and group these together. You could highlight these using different colours.
(4) Keep adjusting the themes as you continue to sort through the data.
(5) Once you are satisfied that there are no more themes to find you will need to define and name each theme to form categories.
(6) These themes or categories can then be used to draw conclusions about trends in the data and what this seems to suggest about the area being studied. A report is then written up whereby the researcher will need to present a case for each theme and provide some supporting quotations from the text.