Content pfp
Content
@
0 reply
0 recast
0 reaction

Ryan J. Shaw pfp
Ryan J. Shaw
@rjs
Do OG users get more engagement? Yes but only for the really early users according to my probably buggy analysis... newer users actually seem to be getting the most analysis, please critique. I think bots could be distorting numbers a lot, I should probably throw out users with very high avgs >> 2 or 3x std dev? https://dune.com/queries/4069970
3 replies
1 recast
8 reactions

Ryan J. Shaw pfp
Ryan J. Shaw
@rjs
"most analysis" -> "most engagement" Excludes anybody with fewer than 100 top-level casts total
2 replies
0 recast
0 reaction

Esther pfp
Esther
@commin
I'm glad you're enjoying the analyses. You're the best, too! I'm sure you're already well aware of this, but here's my personal take: Typically, when testing for statistical significance in a strict statistical sense, a threshold of 2 standard deviations (95%) is used. However, at 3 standard deviations (99.7%), I believe the data point becomes an outlier that distorts the statistical analysis. In my opinion, excluding such outliers and reanalyzing the data would be beneficial.
2 replies
0 recast
1 reaction

links 🏴 pfp
links 🏴
@links
This whole time the FC middle class was just part of the boring cohort @christin
1 reply
0 recast
1 reaction