Psychographic models behind political microtargeting were not very accurate

Evidence

But I still remember something interesting that Kogan said to me during that conversation at the pub, back in the summer of 2016, long before the story broke. After I declined his offer to sell me data, he told me that his models were not very accurate anyway. The timing here is important: he told me this in private, long before all of this became public knowledge. Kogan never changed his story in the media and his testimony before the UK Parliament in 2018 reiterated that the modelling he did on the data did not prove useful in terms of predictive analytics. Indeed, respected researchers – including my own colleagues – have questioned the accuracy of these models and the science behind them. — Sander van der Linden, Foolproof, 506

Categorization

• QUESTION: This seeds genuine curiosity worth chasing: how poor quantitatively were Kogan’s personality-prediction models, and what does their inaccuracy imply for the whole Cambridge Analytica panic?
• XREF: connects to: Dimensional bipolar models misrepresent how people actually evaluate political blocs. — John T. Jost and Jim Sidanius, Political Psychology, 2031

Broader Connections

• Dimensional bipolar models misrepresent how people actually evaluate political blocs. #qualifies (0.75) — John T. Jost and Jim Sidanius, Political Psychology, 2031