Week 8 - Coding and Thematic Analysis Flashcards
Coding = Systematic Interpretation
- We are always interpreting language
- Coding just means we do this systematically in a way that people can see how we arrive at our conclusions
- Saying what is represented
'’Hi, you alright?’’
Example of code
Symbolically assigning (creating a code)
- Saying hello
- Asking how you are
- Demonstrating concern
Not ‘one code’ but coding (it’s a process)
1st round = Initial coding
▪ Can be descriptive.
▪ Continues “familiarisation” phase.
2nd (and subsequent) round = Develops meaning.
▪ May become more interpretive.
▪ May be informed by other aspects of the dataset.
▪ May involve changing/expanding/rethinking initial codes.
End of coding = Collate data that is coded in the same/similar way.
▪ The process of coding breaks down the dataset.
▪ So, at the end, you literally break things.
▪ Start making connections to build a pattern.
Why do we code?
1970s - constructionist ideas and concerns about experimental social psychology
- Critique of experimental studies and questionnaire work
(Harré and Secord 1972; Hudson 1972; Tajfel 1972;
Gergen 1978) - What comes to mind when you think of a “Psychological experiment”
- Crisis in Social Psychology (Parker 1989)
if not ‘experimental’ - hpw should social psychology be ‘done’ ?
Critical Social Psychology (from 1980s…)
The ‘Discursive turn’…
- Turn to text…
- Turn to language… (Qualitative approaches)
Traditional Social Psychology
- Maintenance of experimental / quantitative paradigm
Why dont we just measure their attitudes?
An attitude is a …
“mental and neural state of readiness … exerting a
directive or dynamic influence upon the individual’s
response to all objects and situations with which it is
related”
(Allport, 1935, p.810).
Measured through scales
layers of coding - each coding layer becomes more organised
3rd = Overarching patterns to create psychological understanding (themes)
2nd = Interpretative – capturing ‘meaning’ by sorting
1st = Primarily descriptive
Instead of a scale…
- Coding processes allow us to listen to people and see how they talk and write about things to undrstand how they are constructed ( made real )
(For example, we could ask someone
to write an account, story or collect
some data from the internet)
We code because our evidence is talk and text
- we are not ‘reading’ people’s thoughts
- Systematically using evidence from talk and test
How do we code? Bottoms up OR Top Down? Latent or Semantic?
Thematic Analysis can code ‘top down’ – but this is
often theoretically informed and used when you
are seeking specific language use
E.g. identifying sexist language in a media report
An artificial distinction - talk & text always has semantic and latent meanings
Semantic = surface meaning ‘what is said’
Latent = ‘hidden meaning’ (of course the meaning is not
there to be discovered but is a process of interpretation.)
Depends on your RQ but for psychology we usually move
from semantic into latent meaning.
▪ First round – Initial coding = semantic.
▪ Subsequent rounds of coding = semantic and latent.
You can do a thematic analysis on anything and use different ways of coding… but you need to take care of usualy issues.
RQ – what are you interested in finding out?
Sampling – what is the best sample to answer your
research question?
Ethics – can your RQ be answered safely?
You always develop a thematic analysis which will
produce themes to answer your research question.