published Feb 1st 2017, 8:04 am, by Cathy O’Neil

(Bloomberg View) —
Did the magic of psychological analytics — the big data version of “secret sauce” — launch Donald Trump into the presidency?

Don’t believe the hype.

Trump’s successful digital campaign has lately garnered a lot of breathless attention, notably in a Swiss magazine article that credits a small British company called Cambridge Analytica with giving him an edge in the cutthroat world of political messaging. The idea is that the company employed newly developed methods to derive peoples’ personality traits from their activity on Facebook and elsewhere.

There are a couple of problems with this narrative. First, that particular sauce wasn’t very secret. Second, everyone was doing it, including — or especially — Hillary Clinton.

Don’t get me wrong. Cambridge Analytica CEO Alexander Nix seems amply capable of hatching diabolical schemes, judging from this video in which he brags about helping Ted Cruz with a “cutting edge” model that defines people according to five personality traits. But everyone has something to sell, and in this case, Nix has a product that sounds a lot better if it seems extra sophisticated and devilishly clever.

In reality, personality tests have been around for decades and are in wide use. Big U.S. companies use them to filter job applicants. Or consider this publicly available 2011 paper, in which computer scientists from the University of Maryland explain how to infer the “big five” personality traits pretty effectively from Twitter data alone. Some secret sauce.

Clinton’s analytics team put much more money — and just as much, if not more data — into categorizing and manipulating American voters through targeted political ads. The Trump campaign reportedly focused on Facebook, tricking people into taking personality surveys and culling their likes. Compare that to Clinton’s database, which started with everything that Barack Obama’s campaign had. This would include Facebook data handed down from Obama’s famous analytics team, which employed an app — no longer available — that helped access information about Obama supporters’ friends.

In other words, it looks like Clinton’s campaign had more information, not less, than Trump’s. Obama invented this game. And although Democrats — with the help of Google executives — had a good head start, it wasn’t ever going to last.

To be sure, there’s plenty to be appalled at in this story. The asymmetry of information represented by this new generation of political ads, tailored to the exact personality type and browsing history of each voter, is scary enough on its face. It’s a threat to democracy that directly undermines the notion of an informed citizenry. It’s just not specific to Trump.

Granted, in some ways, Trump’s campaign might have taken the practice to a new level. A case in point: its self-described “voter suppression” efforts, which involved nonpublic Facebook “dark posts” (since suppressed by Facebook and the campaign) aimed at discrediting Clinton among specific groups of African American voters.

But when did campaigns ever play nice? Expect big data strategies on both sides to get dirtier in time for the midterm congressional elections. Whatever you may think about the two dominant political parties, neither has a monopoly on voter manipulation.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Corrects home country of magazine in third paragraph. It is Swiss, not German.
Cathy O’Neil is a mathematician who has worked as a professor, hedge-fund analyst and data scientist. She founded ORCAA, an algorithmic auditing company, and is the author of “Weapons of Math Destruction.”
To contact the author of this story: Cathy O’Neil at cathy.oneil@gmail.comTo contact the editor responsible for this story: Mark Whitehouse at mwhitehouse1@bloomberg.net