Researchers at the University of South Carolina have been looking at Twitter again and decided harm reduction advocates are Big Tobacco shills or automated bots.
By @_Dave_Cross_ via @PVapes
Att: @UofSC
1/ (a thread ⬇️)
planetofthevapes.co.uk/news/health-st…
By @_Dave_Cross_ via @PVapes
Att: @UofSC
1/ (a thread ⬇️)
planetofthevapes.co.uk/news/health-st…
The abstract of this paper reports that "Nearly 42% of tweets showed a bot score greater than .43" and does so immediately following a sentence referring to 'pro-MRTP'(1) tweets.
1) Modified Risk Tobacco Product
2/
1) Modified Risk Tobacco Product
2/
While not explicitly stating these 'high bot rating' accounts are #BigTobacco shills, putting that note immediately after a sentence referring to the Pro-MRTP accounts would lead most readers to assume that those were the accounts with the high bot rating.
3/
3/
But let's look at this claim more closely. Since few accounts were explicitly identified, we can only do an analysis of the Twitter accounts explicitly mentioned in this graphic from page 6 of the paper.
4/
4/
In this spreadsheet detailing the results, we see there are 38 accounts mentioned, 35 of which are currently active. (See the 'TN Table' tab.)
Of those 35, 11 show a @Botometer rating of 0.43 (out of a maximum of 1) or greater.
5/
docs.google.com/spreadsheets/d…
Of those 35, 11 show a @Botometer rating of 0.43 (out of a maximum of 1) or greater.
5/
docs.google.com/spreadsheets/d…
The 3rd tab ('TN Results') summarises the data & hints at the reason why these researchers muddied the waters by not explicitly stating the breakdown of bot rankings between pro, neutral & anti #MRTP accounts.
6/
6/
While 'only' 50% of the Anti-MRTP accounts were deemed bots by the researchers' criteria, the rankings across all 4 were so high that on /average/, the /entire group/ ranked as bots.
7/
7/
The 'neutral' group, some of whom are media sites, was even higher, at an average of 0.55. An interesting inclusion was @FDATobacco itself, coming in at 0.56!
8/
8/
Now turning to the 'Pro MRTP' tweeters: The average ranking was 0.25 - significantly lower than the cut-off of 0.43.
Of the pro accounts at or above 0.42, two were openly & obviously tobacco companies, two were from consumer advocacy groups, ..
9/
Of the pro accounts at or above 0.42, two were openly & obviously tobacco companies, two were from consumer advocacy groups, ..
9/
.. and one, @AlexWodak is a tireless advocate for #HarmReduction. Alex was certainly not a bot on the many occasions I have met him.
@N_Zillatron was also mentioned, I have not had the pleasure of meeting Norbert in person, but have DM'd him and enjoyed his blogs. #NotABot!
10/
@N_Zillatron was also mentioned, I have not had the pleasure of meeting Norbert in person, but have DM'd him and enjoyed his blogs. #NotABot!
10/
@Clive_Bates asks “Who are all these imaginary bots?” and while I do not have a definitive answer to that, based on the small sample known from the graphic, the answer is pointing to: #ANTZ - anti-nicotine & tobacco zealots.
10/
10/
Of course, these @Botometer ratings should be taken with a grain of salt, but perhaps that was not understood by the 4 researchers involved in this study, only /one/ of whom (@alimaan) is even on Twitter, and not very active.
11/
11/
But perhaps Ali Zain can notify their co-researchers of the complete nonsense they're spouting, to help avoid embarrassing themselves (in public) in future.
12/
12/
"Conclusion: Continuous efforts are needed to surveil the industry’s attempts to create a climate of false consensus" should possibly read ".. surveil the /tobacco control/ industry’s .."
13/
13/
For the attention of (identified as 'bots'):
@AETOSWire
@SaharaReporters
@TobaccoFreeKids
@InsidePMI
@N_Zillatron
@AltriaNews
@FDATobacco
@exposetobacco
@AlexWodak
@AVABoard
@CASAAmedia
14/14
@AETOSWire
@SaharaReporters
@TobaccoFreeKids
@InsidePMI
@N_Zillatron
@AltriaNews
@FDATobacco
@exposetobacco
@AlexWodak
@AVABoard
@CASAAmedia
14/14
• • •
Missing some Tweet in this thread? You can try to
force a refresh
