User Tools

Site Tools


evaluative_criteria

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

evaluative_criteria [2015/02/18 17:02] (current)
Line 1: Line 1:
 +**Evaluative criteria for qualitative studies**
 +
 +For a number of reasons, the criteria used to assess the quality of quantitative studies are difficult to apply meaningfully to qualitative research studies. These stem from the small sample sizes and problems of generalising,​ the influence of the researcher (their subjectivity),​ the difficulty of replicating a study due to subject and researcher variability as well as the underlying notion that there exist multiple accounts of social reality rather than a single absolute account (that is, no single ‘right’ answer).
 +
 +**Good practices which support the production of credible outputs from qualitative studies**
 +
 +    - Being reflexive (based on Braun and Clarke, 2013: Chapter 2 and p 279) Qualitative researchers acknowledge that the researcher inevitably influences the research process. For example, what is said in an interview or focus group depends partly on the presence and skill of the researcher. Social science researchers are not robots but rather each brings their own histories, values and assumptions to the process (that is, their subjectivity). Qualitative researchers should avoid inadvertently influencing the results by reflecting on the knowledge they produce and their role in producing that knowledge.
 +    - The coding applied to the text as the first stage in analysis should be thorough, systematic ​ and inclusive, with all the data that was collected being considered, and not just a few vivid examples (an anecdotal approach) (Braun and Clarke, 2013): Chapter 12)
 +    - Being transparent. Researchers should explain how the data collection and analysis were conducted. ​ They should include clear explanations of the process followed when coding, and of the meaning of the categories derived, such that another researcher could conduct the coding and derive similar results.
 +    - There are some fairly commonly used some techniques to assist us in establishing whether we can trust the results. Triangulation uses more than one data source to cross check the findings. Inter-rater reliability involves a second person coding the text; member checks involves going back to the people who were studied and checking whether they think the account produced is believable.
 +    - To assess if the account is credible, there should be a good fit between the evidence contained in the text and the concepts that are developed from the evidence, and any theoretical framework. These should be clearly explained in a narrative account. To enable readers to judge whether the interpretation is adequate, extracts of text are quoted. ​
 +
  
evaluative_criteria.txt · Last modified: 2015/02/18 17:02 (external edit)