This theory and methods question came up in the May 2020 A-level sociology Education with Theory and Methods exam paper.
Outline and explain two problems of using primary qualitative methods in sociological research (10)
Below I include the mark scheme and a possible answer to this question…
Hints for answering this question
The Indicative content the AQA provides in its mark scheme is as follows:
- cannot make generalisations
- cannot isolate cause and effect
- difficulties analysing data
- low in reliability
- subjective
- the imposition problem
- unscientific
- cannot generate social laws.
Personally I see these as falling into three groups which naturally fit together….
- Cannot make generalisations/ cannot generate social laws.
- Cannot isolate cause and effect/ difficulties in analysing data
- Low in reliability/ the imposition problem/ subjective/ unscientific.
So the strategy here is to pick two of these and make these your points.
And it would be good to use examples from some qualitative methods, namely Participant Observation or Unstructured Interviews, ideally both…
Suggested answer
The first problem of using primary qualitative research methods such as Participant Observation is that they are unscientific, and influenced by the subjective biases of the researchers who use them.
A good example of this is Venkatesh’s study Gang Leader for a Day. He seemed to really like the leader of the 200 strong crack-dealing gang he studied, and the fact that he got on so well with him may have influenced the way he interpreted his actions. He may also have been selective in what he wrote, possibly keeping some of the worst crimes out of his publication.
A related problem is that we can never know how bad the bias is, because participant observation also lacks reliability, meaning that they are difficult to repeat in the same way and get the same results.
This is especially true for Participant Observation which is done over a several months of years. It would be impossible for someone else to replicate the context of Venkatesh’s study because the same situations will not occur again as there are so many different actors involved the research on the ground.
If another researcher could go back and repeat the research they may follow different leads, speak to different people, and find an entirely different interpretation of the same gang.
Although this is a moot point because by the time Venkatesh published his in-depth study the projects where he did the research had disappeared, so the length of time it takes to write up findings adds to this problem of reliability.
HOWEVER, one can at least verify the findings by showing them to the respondents who took part.
A second problem of primary qualitative data is that it is not usually possible to isolate cause and effect and uncover ‘social laws’
For example with unstructured interviews which are respondent led you might have a broad topic area and then ask respondents different questions to different respondents.
You would have a nice ‘thick description’ of what respondents think and feel about a topic area, but also just a mess of qualitative data.
The data you would get would be messy, possibly quite long winded and possibly with lots of irrelevant information if the researcher allowed the respondents too much freedom.
Thus if you do 10 unstructured interviews you might end up with 10 very different sets of findings with no common themes.
It would be hard for a researcher to analyse this data and find the common themes, isolate any variables and find out the discrete causes of social action.
To do that you would need to have structured questions and limit the options available to the respondents.
Having said that today it should be possible to use software to find the common themes in qualitative data, but even that might miss out on picking up on respondents talking about the same thing in different ways.
Signposting
For more posts on exam advice for A-level sociology papers please see my exams and essays page.