1/ Today Yahoo News published a story about a report from the Soufan Center on how China and Russia were fueling Qanon conspiracy theories. The report has a problematic methodology in regards to how it uses and perceives data, also the report has a weak grasp of QAnon.
2/ First lets get something clear, the report claims to be data driven, however, there is no explanation of the data sources (I am assuming it is @crowdtangle ). If that is the case then there should be a methodological explanation. If not that should be explain as its important 
3/ That data sample is a fraction of what I collected from January to August 2020 there were 1.65M posts in the 416 QAnon pages and Groups I was curating. Considering they claim a full year of collection I do not think their sample is large enough
4/ Also considering that Facebook has done a good job at cleaning QAnon from the platform. I would be interested to know what groups and pages where being collected post fall 2020 purge and post jan 6 purge.
5/ If Limbik used Crowdtangle did they use the search functionality? if so what were their search terms. Is Soufan aware of the limitations of search? If the curated QAnon groups and Pages, what was the threshold of inclusion as QAnon?
6/ This is the definition Limbik has of QAnon categorized, which can include almost anything that falls into the conspiracy sphere, anti-covid content or polarized political narrative. This definition shows a lack of understanding of what is a QAnon adherent
7/ Limbik also does not distinguish between a QAnon account, someone who is into conspiracy theorists, a MAGA supporter, QAnon adjacent. this should be scaled.
8/ The problem does not end with this but also with their definition of "foreign influence". By this definition, as a Canadian that posts about QAnon content I would be categorized in this foreign influence category. If I had a QAnon page or Group I would count.
9/ There is no evaluation of the account, Limbik is an enterprise service provider, it is a backbox in terms of methods and thresholds. data goes in an answer comes out but how we got to that answer is never explained. That is a terrible method of evaluation.
10/ Additionally, crowdtangle data and OSINT (as already explained by @3r1nG ) will not provide any level of confidence in regards to attributing the foreign nature of a Facebook account. Anyone can create a fake foreign account.
11/ I did a manual analysis of all 416 groups and pages I was curating to map out the transnational dimension of QAnon. The location of the administrators was the weakest factors in my evaluation as it was the most unreliable step as there is no clear attribution possible.
12/ What was important was the language being used in the chats, the socio-political narratives, the sources of the news, and the top level domains mostly reference. Even then my findings are my best qualitative guest. Data alone cannot determine this with facebook.
13/ The lack of a clear methodology, the poor thresholds, the lackadaisical definitions all combine to inform me that the data findings of this report are skewed from the get go.
14/ To that end I would caution everyone to not rely on the findings from this report that rely on the Limbik data. The Yahoo news story is cute, but behind the click bait title, the findings are unreliable as they stand.
15/ This is what most news sites are going to report on. However, the report fails to prove this in anyway. This report would require clear edits and reevaluations prior to claiming that foreign actors are fueling and promoting QAnon content.
16/ If the report claims that foreign actors are spreading narratives that are aligned with conspiracy theories that is one thing. But using soft power to fill data voids that mainstream sources will not occupy is a different thing from foreign actors administrating QAnon pages
17/ This section alone is, in my opinion, irresponsible. As very few will read the report in full, some may read the Yahoo article but most will simply read the clickbait title that is based on skewed methods, definitions and limited data.
18/ The next criticism I have is QAnon cannot be compared to ISIS, this needs to stop. There is a difference between religiously motivated violent extremism and ideologically motivated violent extremism
19/ Furthermore, ISIS has a clear command structure. QAnon is a an amorphous and decentralized movement. Q was not Baghdadi, QAnon influencers are not emirs. As @travis_view has said QAnon is a big tent conspiracy theory.
20/ Having spent 3 years researching ISIS and 3 years researching QAnon I can say with a high level of confidence that the recruitment and radicalization process is not the same between these two groups.
21/ Also the statement that "Individuals are radicalized in online echo chambers by false narratives and inspired by other attacks, such as the January 6, 2021 insurrection in the U.S." is incorrect.
22/ Data analysis and a qualitative analysis of QAnon communities on alt-tech platforms after Jan 6 showed an initial spike in membership due to the migration, however, this was followed by a drop in content and activity for a few weeks. only recently did activity pick up some.
23/ Jan 6 did not inspire growth in the movement. Additionally, other attacks do not attract people to QAnon as it does with other IMVE actors. this finding represents a flawed understanding of the QAnon community.
24/ The report never makes clear what is a "QAnon Narrative". Are the authors distinguishing between Canon QAnon narratives, neo-QAnon narratives, Qanon-adjacent narratives, or conspiracy theories or do they all fall under the same umbrella?
25/ The report also states that "Rather, the
ideological creed and commitment to violence
among QAnon adherents is likely to mutate in
the foreseeable future." To be clear QAnon is an ideologically motivated extremist movement, that can lead some adherents to violence.
ideological creed and commitment to violence
among QAnon adherents is likely to mutate in
the foreseeable future." To be clear QAnon is an ideologically motivated extremist movement, that can lead some adherents to violence.
26/ However, QAnon at its core is not committed to violence. yes it is anti-establishment and apocalyptic, which are risk factors for radicalization to violence. BUT Q and QAnon influencers were not committed to violence.
27/ The real risk factor for violence in QAnon is how deplatforming has pushed them into spaces inhabited by violent extremists, who will prey upon QAnon adherents and seek to further radicalize them into other ideologies/movements.
28/ Further in the report when discussing QAnon narratives, which again appears to be grounded in Limbik data, there is a clear demonstration that there is no distinction between QAnon and general conspiracy theories.
29/ What the report can conclude is that disinformation resonates with conspiracy theory believers and these individuals can radicalize and mobilize to violence. However, the QAnon dimension is tenuous and superficial at best.
30/ This again echoes what I stated above about foreign influence, all the report can claim is that individuals outside of the US who have posted conspiracy theory related content leverage sources, content or narratives that may be linked to a foreign source. END/
31/ I would add the only report that did good coverage of foreign actors and QAnon content stems from the work of @MelanieFSmith public-assets.graphika.com/reports/graphi…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
