Data Analysis

Data Analysis

4. Coding interview transcripts and other data

Coding  is a form of data analysis that involves  labelling  and  organizing  data  to  reduce  your  data  to  a  manageable  size.  Coding allows you  to  conceptually  organize your data but  it  does not  provide  an  explanatory  framework to interpret or draw conclusions from your data.

Coding  can  be  used  in  an  exploratory,  inductive  way  such  as in  grounded  theory  (in which  you  generate  theories  from  empirical  data),  or  it  can  be  used to  support  a  theory in  a  more  deductive  manner.

How to code data:

  • Code your  interview  by  extracting
  • Give labels  to  items  that  seems  to share  a  similar  theme  and  that  you  think  may  be  of  theoretical
  • The coding  should  flow  from  the  questions  but  should  also  reflect  what  is  said  which  may  drift  away from the  focus  you
  • Develop master  categories  that  relate  to the  data  (some  data  may  fall  on  more  than  one  category).  Then you  can  have  sub- master categories . For example, ‘class’  can  be  divided  into  working  and  middle
  • Emic codes  /  in  vivo  codes are codes those  used  by  your
  • Etic codes/  analytic  codes  are those  assigned  by  the  researcher  and tend to be more theoretically
  • When you  think  you  have  finished  coding  you  can  develop  a  codebook  – a  list  of  all  the codes.  It can  include  descriptive  codes  and  analytic
  • You may  have  more  than  100  codes  so  you  then  start  to  cluster  codes  into  For example,  common issues  faced  by  interviewees,  characteristics  of  interviewees,  identities,  memories,  representations,  places, etcetera.   You  may  end  up discarding  some  codes.
  • Go through  your  data  again  once  you  have  a  code  book  and  recode  the

Data can be transcribed, annotated and coded manually. Manual coding allows you to become very familiar with your data. There are also computer software programs available for managing and processing data, which can be particularly useful for managing large volumes of data and connecting datasets in different formats.

Reference

Peake, L. 2018. Presentation at the ‘Workshop in Urban Feminist Research: Ethnographic Research Tools’, Ramallah, Palestine, July 2018.