Tim Pool on statistics: what the evidence says · JRE #1258
SUBJECT: STATISTICS
The short answer
Tim Pool claims a study found only about 5% of posts on Gab were hate speech compared to 2.4% on Twitter, characterizing this as a marginal difference. The figures Tim Pool cites trace to a widely referenced 2018 academic study (Zannettou et al., analyzing 22 million Gab posts from August 2016 to January 2018), which measured the prevalence of hate speech on Gab and found it to be much higher than on Twitter, though lower than 4chan's /pol/ board.
Not a true/false call. Every claim is logged with its sources; read the exhibits below.
Gab, a study was done, I talked about this last time where they found five percent of the tweet of the i don't say tweets but the posts on gab or hate speech compared to twitter's like 2.4 so it's a marginal increase
What the evidence says 01 / RECORD
The figures Tim Pool cites trace to a widely referenced 2018 academic study (Zannettou et al., analyzing 22 million Gab posts from August 2016 to January 2018), which measured the prevalence of hate speech on Gab and found it to be much higher than on Twitter, though lower than 4chan's /pol/ board. Pool's raw numbers (roughly 5% on Gab versus 2.4% on Twitter) are approximately consistent with that study's reported ratio. However, describing a roughly 2-times-higher rate of hate speech as merely a "marginal increase" understates what the same line of research found: Gab was identified as a platform predominantly attracting alt-right users, conspiracy theorists, and other trolls, sitting closer to overtly extremist forums than to mainstream platforms. Subsequent peer-reviewed research corroborates that Gab's hate speech problem is structural and worsening rather than marginal: one study found hate speech on Gab increasing over time with hateful users becoming more central to the platform's network, and another found Gab distinctively 'permits its users to post nearly anything, including hate speech,' unlike mainstream platforms, and has attracted a large base of self-identified far-right users. Overall, the raw statistics Pool cites are approximately consistent with the source research, but the 'marginal' characterization minimizes a meaningful disparity on a platform independently documented as a hub for hate speech and extremist content.