Facebook Remorse?

When we think candidate’s Facebook pages, we’re thinking about beer bong pictures and other adolescent behaviors that tell us if the candidate has basic character and discretion. Don’t we?

We’re certainly not thinking about predictive analytics designed to guess the probability of disease, pregnancy and the health issues  like alcoholism and migraines.  But if you think about what’s going on with big data these days, why wouldn’t that happen?

Read this and consider:

Candidate A’s resumé rose to the top because of LinkedIn Expert, the social network’s high-end professional service. LinkedIn developed technology to data-mine resumés for specific qualifications. Candidate A’s research on trade disputes between Korea and the USA caught everyone’s interest at a top technology consulting company. That’s why her “3D Resumé”  rose up to the top of the candidate pile.

The hiring manager does this pitch:

“She’s what you need for the transpacific trade issues you had mentioned. Her dissertation speaks for itself, she even learned Korean…”

He pauses.

“But?…” Asks the HR guy.

“She’s afflicted with acute migraine. It occurs a couple of times a month. She conceals it, but our data shows it could be problemlatic.”

“How did you learn that?”

“Well, she falls into this particular Health Cluster. In her Facebook wall, she sometimes refers to a spike in her sensitivity towards smells — a precursor to a migraine headache. In addition, each two weeks, we see a drop in the number of words she uses in Facebook posts, her vocabulary diminishes, and her tweets, usually sharp, become less articulate. That’s an obvious pattern for people suffering from serious migraine. In addition, the Zeo Sleeping Manager website and the stress management site HeartMath — both now connected with Facebook –  suggest she suffers from insomnia. In other words, we think you can’t take her into the firm. Our Predictive Workforce Analytics Modeling shows that she will cost you at least 20% more in lost days and productivity. Not to mention the patterns in her Facebook posts suggest a 60% chance for her to become pregnant in the next 12 months, according to our predictive models.”

“Not exactly a 100% certainty, but OK, let’s move to the next candidate”.

You might think I’m exaggerating with this tableau. But the fictitious Company could be using existing large quantitative research firms, combined semantic and data-mining information resources such as Recorded Future. This Sweden-based company, which has a branch in the USA,  provides real time analysis of thousands of sources (news services, social networks, blogs, government web sites). The firm offers clients the ability to predict a vast array of events (see this Wired story).

• Do we fully understand the impact of big data and HR? Probably, we do not. Are predictive analytic techniques that speak to the possibility of a candidate’s insomnia and migraines crooked cards?
• Is it right or is it wrong?  This is a probability score index.  Perhaps this index was modeled to give you a productivity and “ability to complete projects on time” score rather than talking about health issues.  Perhaps, the ability to be creative is under scrutiny?  Would that be more useful to Human Resources?
• What if one used a Psychology professional rather than metrics alone to give the hiring company access to plausible deniability?
• We would guess that high-end firms that already use this type of service will be the first to try predictive analytics in the hiring process. Perhaps we will eventually see the many uses of analytics litigated out in the courts.