content and thematic analysis

what is content analysis

is a research technique in which researchers indirectly study behaviour by examining communications that people produce, observe the content they create rather than

further info about what it does

• used to analyse qualitative data by turning it into quantitative data via use of coding

• A content analysis does not collect data directly, rather uses pre-recorded examples of spoken interactions, the written word or media content

• Aim of content analysis is to summarise the main ideas presented in the spoken or written material via structured methods

coding process

• Researcher formulates their research question

• Select sample of preexisting qualitative data sources (some of which they conducted and some was conducted by others) and decide how the material should be sampled (time or event sampling)

• They decide how the data should be recorded for analysis

• Familiarise themselves with the data and decide on the coding of pre-determined categories/coding units

• Researcher works through the data using a tally system/chart to record the number of times the categories appear

• Researcher will then need to test for reliability

how to test for reliability

• Test-retest reliability - run the content analysis again on the same sample (material used first time round) and compare the two data sets

• Inter-rater reliability - second rater conducts the content analysis with the same coding categories and data. Compare the two data sets

• If the results are similar then this shows good inter-rater reliability

• If a correlation of 0.8 or stronger is found that is usually accepted as showing the data is reliable

strengths of content analysis

• Strength

  • Original qualitative data is rich in detail and meaning is it's usually taken from the real world which brings with it external validity

  • Transformation of the qualitative data into quantitative data means that overall trends and patterns can emerge and are easily compared which brings with it reliability

  • Anyone can access the same media and use the same coding system to replicate the content analysis and check results

• Strength

  • Enables researchers to investigate topics which might otherwise be off-limits due to ethical concerns

    • Material they access is in the public domain

    • No need to gain informed consent to access or report on the material

    • Means that researchers have freedom of choice as to what material they access and how much of it they choose to use

    • Usually less ethical issues as no people are directly studied - circumnavigates those

weaknesses of content analysis

• Material that has been produced outside of the research process

  • Means that the research itself is separate from the immediacy of what took place during the production of the material

  • The researcher can only study the data, often without knowing the true context of its origin

  • This could result in the researcher making assumptions about what they are reading which would affect the validity of the findings (researcher bias)

• Limitation

  • Converting qualitative data into quantitative data means that the essence of some of the original data is likely to be lost

    • A numerical value cannot adequately sum up the feelings and emotional content of the original qualitative data

    • This could affect the validity of the findings as the true meaning of the data may not be represented in the findings

what is thematic analysis

• method used to analyse qualitative data using an inductive method

  • Themes emerge from the data, there is no hypothesis-testing involved

further info

• A theme is any feature of the data which recurs throughout

• Similar to content analysis as it is also an indirect method of observation

  • Different as themes only emerge during the analysis and after an analysis has finished 

• Inductive - approach the data with no preconceptions about which themes might emerge

• Deductive - researcher already has ideas about the themes that might emerge from previous research, they examine the data to see whether these themes are present

process for thematic analysis

• Familiarisation - reading and re-reading

  • Researcher reads transcripts over and over again to become familiar with the data and to understand the meanings being communicated

• Identification of themes

  • Researcher identifies recurring themes, meaning ideas (implicit or explicit) that keep 'cropping up'

• Review of themes

  • Researcher reviews these themes and patterns to see if they can explain behaviour and answer research question

• Clear categorisation

  • Researcher categories and defines each theme

    • People in their 20s often refer to 'lack of savings', 'high rents' and 'lack of parental support' as obstacles to house-buying

• Writing up analysis

  • May collect new set of data to test validity of themes and categories

  • Researcher writes up the analysis into their formal report - outcome is qualitative

  • Themes or categories may be listed and supported using example quotes directly from the data

strengths of thematic analysis

• Only qualitative data - researcher can quote directly from the original source material, adding context to their report, bringing real, subjective human experience to the forefront of their findings

  • More descriptive than coding units

• Flexibility - themes are noticed as the emerge as the researcher does not go in with any previous ideas

• Allows researchers own perspective - can bring own personal experience and expertise to the analysis - reflexivity

• Fewer ethical issues - no people are involved so typically less ethical issues are required to be considered

weaknesses of thematic analysis

• Time-consuming - text must be studied, analysed and reviewed repeatedly to identify and categories the themes, meaning some researchers may decide not to use it, limiting its usefulness

• Can't use statistics - as this research method provides qualitative data it means statistical tests cannot be done on the results to investigate significance

• Subjectivity - some themes may be missed that the researchers do not notice or relate to, researcher may be prone to confirmation bias (unconsciously focus on themes that align with their own expectations)