Content analysis

What is content analysis

• An indirect observational method that is used to analyse human behaviour, investigating through studying human artefacts (the things people make)

• Content analysis is often on the written word (qualitative data) or write ups of spoken words (Transcripts)

• This is transformed into quantitative data

How to perform a content analysis

• Decide a research question

• Select a sample (e.g. systematically, randomly etc.) from a larger quantity of all possible data (e.g tweets, diary entries etc.)

• Coding: The researcher decides on categories/coding units to be recorded (e.g. occurrences of particular words), these are based on the researcher decides question

• Work through the data: Read the sample, an tally the number of times the pre-determined categories appear

• Data analysis: Can be performed on the quantitative data to look for patterns

• The coding units should e operationalised, meaning thy are as clearly defined as possible to try to reduce subjective interpretation

Testing for reliability

• Test retest reliability: Run the content analysis again on the same sample and compare the two sets of data

• Inter-rater reliability: A second rater also performs the content analysis, with the same set of data and the same behavioural categories- compare the two sets of data

• How closely the two sets of data match is assessed with a test of correlation such as spearman’s rho

• A correlation of 0.8 or stronger is usually accepted as showing the data is reliable

Content analysis evaluation

• The “artefacts’’ are usually not created for research but are taken form the real world

• This means content analysis has high external validity, and findings should be generalisable to other real world situations

• As the artefacts come from the real world it is often easy to gather a sample

• Other researchers should be able to replicate a content analysis using the same coding units/ behavioural categories and an easy to access sample

• As the researcher will often need to interpret subjective text, this maybe lead to researcher/observer bias, when researchers tend to interpret the text in a way that supports their pre existing views

• Data is created for purposes other than the research as the data was not created under controlled conditions so it may lack validity

Thematic analysis

• When researchers start by attempting to identify the deeper meaning of the text by reading it first, and then allowing themes to emerge

How to performa thematic analysis

• Collect texts/turn recordings into text through transcription

• Read texts/transcripts first to spot patterns that can be coded and colected

• Re-read the transcriptions/codes looking for emergent themes

• It is important when writing about thematic analysis that themes are made clear and are no pre-determined by the researcher but com from/emerge from the text

• As theories come after the discovery of themes, it can be argued that i stops the researcher imposing their own bias on the analysis by only looking for what hey want to se

• High external validity, easy to get a sample, easy to replicate

• Subjective interpretation, data not created in controlled conditions