Please do, I think the meeting will be still happening, just Richard will not be there I think
Jan, all this is correct, and intentional. Let me quote the white paper:
We can think of the [codes co-occurrence] network as a pattern of free associations, specifying the connections between the concepts encoded in ethnographic codes. Such patterns “represent a powerful and meaningful way of building network models of the mental lexicon” [Stella et. al 2019]. Here it is not a single person that is associating concepts, but a community of people in conversation. Accordingly, each edge represents not one, but k associations.
The model is robust to idiosyncratic convictions and random associations, because they will tend to occur only once or twice. By filtering out low co-occurrence edges, you are likely to be left only with robust associations – though the association is not of any one informant, but of the collectivity of informants as a whole.
So, don’t worry too much if you don’t get single interpretations exactly right: any mistake you make is likely to correct itself. If you “overcode” one post, you will create a lot of weak edges, which will be filtered out. If you “undercode” one post, its strong associations are likely to emerge elsewhere, and repair the error. Neat, eh? 
I am not sure this categorization is helpful. I could create exactly the same direct edge with
“The Church should take over the government, and install a theocracy!”
If the only two codes at play are church and government, then all the graph is saying is that someone in the community of informant is associating the two. To say more about this association, you could look to the neighboring codes collaboration and mutual support might mean that Church and government are allies, whereas conflict and political strife might indicate that they are not.
2 Likes
Dear All,
Here is our last meeting’s summary (08.07).
Our ethnographic material will consist of the following four data sets:
- The preliminary material collected through ER,
- Control group, made of the first bunch of ethnographic interviews with people expressing their political views in online public spaces,
- Common track interviews - a total of 30 interviews per country, 15 interviews with people seeking health advice online, and 15 interviews with people skeptical towards the pandemic/vaccinations.
- Case studies - individual tracks taking on various issues important in particular countries, 25 interviews per case.
Research questions remain with a focus on exploring political views, more on the topic in the new research outline: POPREBEL Ethnography: A new research outline - Google Docs
Meanwhile, work continues on putting together the final version of the codebook. The idea is to get rid of as many inessential codes as possible, to have them in dozens, not hundreds. While doing so, we need to keep in mind the following questions and issues:
- Think through emotions - what emotions came in the pilot study and the control group? What needs and problems?
- Pay particular attention to emotions such as hate vs anger, trauma vs outrage, as those emotions might be what motivates voters to go for populist parties/policies.
- Jiri created a list of intermediary (second level) emotion categories. Everybody should review them and share with other possible comments. We want to close this set of categories.
- Capture closer meaning with new codes, but do not obsess over details; it’s only natural that some contextual information will be lost with the analysis we choose.
- We should be done in 2 weeks’ time.
As soon as the new codebook takes shape, we will do an experiment to code 3 interviews in-vivo with the NVivo program to go for more social analysis, and code them using SSNA. If there will be anything interesting within the outcome of such double coding, we will keep the method, which can result even in a methodological paper.
Next meeting: 23.07, 9 AM
If I forgot to note down something, please comment!
3 Likes
thanks for this Mania! Next meeting 23.7. 9AM Jan’s time?
Yes, Jan’s time, which makes it CEE 3 pm!
Where are we?
Hey @amelia , are you joining us today?
https://us02web.zoom.us/j/89020808871?pwd=OUFZMSt6N0ZEUWI2a2RhNnZXcmNzZz09
Dear All, Wojt and I have been working long hours on the revised coding scheme. It is going well, but it is still some to go before we are 100% done. We have, however, made enough progress to have several important ideas to discuss with y’all. So, we are looking forward to seeing you this Friday (July 30) at 9:00 am EST (15:00 in East Central Europe). I hope @amelia and @alberto will be able to join us (though I am aware that we are in the middle of the vacation season). By tomorrow evening, I will post here some basic ideas we will propose to discuss. There will be a lot of information and we are excited to hear your critical reactions.
I also want to let you know that we have had a first, very preliminary signal from Brussels that our project extension request is in order and - most likely - will be formally accepted in the very near future. All my best, Jan
1 Like
I am on the road, so cannot make a firm promise, but will do my best.
Janie, can you remind us what our homework is for tomorrow’s meeting?
Dear All,
the most pressing task, by far, is the revised categorization of codes and the creation of the up-to-date codebook. Please invest as much time as you can in thinking about merging and pruning. The original plan of work looked like this:
Z Categories
- Actions and activities – (Jan)
- COVID 19 – ?
- Emotions – Jirka (discuss)
- Ideology – Amelia
- Institutions – Jan and Wojtek (isolate and categorize actors. The same for institutions)
- Social movements and historical events – Jan and Wojtek (1st) find actors
- People and identity – Richard
- Places - ?
- Problems and needs – Jan and Wojtek (includes now Resource needs)
- Social and political processes – Jan and Wojtek
- Values and beliefs – Amelia
You can see Wojt and I are doing this if you follow the second link below. Today I am working (partially with Wojt) on two sub-tasks:
-
Preparing my ideas for sharing with you tomorrow (you can follow this work and also record your own ideas here:Biweekly meeting (23 - Google Docs
-
Constructing the new system of codes for problems and needs, by far the biggest category. You can follow Wojt’s and my work here: (link with permission): Problems re-coded - Google Tabellen
Please write if you have any questions. Most likely I will be working the whole day.
Ciao,
Jan
Hello, my notes are in the last part of this document: Biweekly meeting (23 - Google Docs
The spreadsheet with our work is here: Problems re-coded - Google Tabellen
See you at 9:00 am EST on Friday!
Ciao,
Jan
https://us02web.zoom.us/j/89020808871?pwd=OUFZMSt6N0ZEUWI2a2RhNnZXcmNzZz09
We should be able to use this link again.
3 Likes
Hey, good morning! Will be there at 9:00. Wojt and I are working now on more re-coding.
Hi all — my apologies for missing the meeting, I have been very unwell the past week. Things looking up slowly. Will look at your progress and contribute my own ASAP. Stay well!
2 Likes
FYI, I talked to @matthias yesterday. In a day or two he will put up a proposal in the form of a well-structured GitHub issue, then I will ask @rebelethno to OK it. 
That proposal is ready, though it consists of a set of four separate enhancements listed below. Please @rebelethno have a look and tell me if that sounds suitable. The idea is to provide an efficient interface for moving annotations to a different code. That process would replace the current code splitting that starts with copying a code.
If it suits your purposes, we can implement this next week (between August 9 and 13).
1 Like