Yesterday @amelia, @Leonie and I started onboarding my replacement: Kate Sim. She will introduce herself when she joins the platform. For now, I just want to share with you the conversation that Leonie and I had with Kate about the nitty-gritty of coding on platform.
We thought this might be relevant for the wider Edgeryder team and in the spirit of transparent research practices.
Always happy to discuss further.
NGI TEAM MEETING + New Ethnographer Training Day! NOTES (16.06.2020)
Using the Codebooks: input codes in alphabetical order with a brief definition (see style guide), mark who is coding which threads
a. Use a dictionary where possible, use invivo description for invivo terms, use own definitions for original codes
b. For repeated codes, mark threads that contain the same code, e.g.
Internet: (see thread on Non-Ubiquitous and Communal Internets).
c. Be collaborative: highlight and comment on your own + other team members’ codes maintaining a dialogue about the coding process is key!
d. Memos: capture your thoughts, comment on each other’s threads, flag issues, things to monitor etc.
Example: Memo #1 (Covid 19 Threads) CCS
Starting to work on the Covid threads, noticing now that many of the initial ones are set up and primarily populated by people of the Edgeryder team. Not a lot of input from other participants on the platform yet. Also side-note, we probably need to make a decision about how we want to deal with the fact that people use the words “corona” “covid” “covid-19” etc interchangeably (or so it seems). Any thoughts on this? I wouldn’t want to add unnecessary splintering to the SSN graph when were really talking about one single thing (if that remains to be the case).
It actually might be interesting to monitor this for a bit and see when each term is used and why. I am guessing ‘corona’ might occur more in less formal contexts for example. We could also monitor which term is being used the most/appears most frequently and then go with that.
Three Codebooks: Main codebook, Environment, Covid-19
We mainly use the big codebook until we identify core thematic areas that we want to devote more time to. We move coding to thematically dedicated codebooks (e.g. environment, covid), but cross-reference these for overlapping codes/definitions
Trouble-shooting: collaborative coding in a project without a clearly defined goal/ no clear theoretical framework can be frustrating: coding is often very descriptive as community members aren’t yet bringing in a lot of personal narratives and opinions, not a lot of qualifying language. It is totally fine to continue with descriptive codes, but it would help to have more 2nd and 3rd rounds of coding the same thread and scheduling meetings around emerging themes + develop more theoretical frameworks for analysis, developing more specific social semantic codes (memos also help). e.g. AI and ML – bring in fairness and accountability literature (that is less US-centric).
Going forward: more check-in meetings, discussing codes, identifying themes
Time spent coding: It depends on the thread: some threads are dry and descriptive and can be done quickly/only require one round, others take longer and require several rounds of coding. In general, max 2 hours per session.
Multiple codes per highlighted item are possible and encouraged (sometimes a little tricky to get the clicker to work).
Editing codes: its ok to make “mistakes”, to use codes as place-holders or change your mind about codes later all can be edited!
Focus on codes that have a lot of comments (particularly those with a range of community members – not just the community managers)
Make your own posts: it is encouraged that we add our own content, bring in new ideas, flag questions for discussion, participate in discussions etc. but only when/if we feel comfortable