Strata Week: Revolutionizing human resource management with work-force science

Big data replaces gut instinct in HR management

In a post at the New York Times, Steve Lohr took a look this week at a new data discipline: work-force science. The field pairs big data with human resources to help remove subjectivity and gut instinct from the hiring process and HR management. Lohr notes that in the past, studies conducted to understand worker behavior included a few hundred test subjects at most. Today, they can include thousands of subjects and far more data points. Lohr writes:

“Today, every e-mail, instant message, phone call, line of written code and mouse-click leaves a digital signal. These patterns can now be inexpensively collected and mined for insights into how people work and communicate, potentially opening doors to more efficiency and innovation within companies. Digital technology also makes it possible to conduct and aggregate personality-based assessments, often using online quizzes or games, in far greater detail and numbers than ever before.”

Lohr looks at several companies applying data-driven decision making to HR management. HR consulting startup Evolv, for instance, uses data science to help client companies make hiring decisions and develop effective management techniques. In a partnership with call-center operator Transcom, Lohr reports, Evolv enabled the center to hire fewer workers — 800 as opposed to 1000 — to end up with 500 workers who remained on the job for at least three months, reducing worker churn and improving customer service.

Google, unsurprisingly, is one of the larger companies applying data science to its human resource management. Prasad Setty, vice president for people analytics at Google, told Lohr that their practice of using SAT scores and college GPAs to screen candidates did not prove successful and has been discontinued; Google now conducts extensive surveys of its work force instead.

Max Nisen notes in a post on the same topic at Business Insider that Google’s data has shown their “most innovative and happy workers feel like they have autonomy and a strong sense of mission.” Such insights, Nisen reports, come from Google’s detailed employee tracking program that’s used to help optimize employee benefits, pay, and the workplace environment. “Data comes from the usual sources, surveys, employee feedback, and so on,” Nisen writes, “but also from experiments from the company’s People & Innovation Lab designed to figure out best practices. Google applies that wealth of data to both hiring and management.”

DataKind launches international U.K. chapter

U.S.-based non-profit DataKind announced this week it’s launching its first official chapter, DataKind UK. Kaitlin Thaney, Digital Science’s manager of external relationships and chair of the London Strata Conference, explained in a post at the Guardian why the group chose to expand to the U.K.:

“Last September, DataKind ran a “DataDive” in London the weekend before Strata Conference in London (note: I chair the event), connecting volunteer data scientists with three UK charities. Despite being told that there wasn’t as strong a data community in the UK as in the US, over 70 data scientists came out to spend the weekend helping these charities use data to maximize their impact, working with charities such as Oxfam, Place2Be and Keyfund.” (Editor’s note: links added.)

Thaney, who will serve on DataKind UK’s board of directors, says the group will be running DataDives throughout the year will serve “as a voice for proper data use in the social and government sector.” You can read more about DataKind UK on DataKind’s website and sign up to stay informed on the DataKind UK website.

German regulators fine Google a “paltry” sum for violating data protection laws

German regulators fined Google this week for collecting Wi-Fi data with its Street View cars. Karin Matussek reports at Bloomberg that Google was fined 145,000 euros, or $189,230. She reports that Hamburg data regulator Johannes Caspar said in a statement (PDF) that the Street View cars gathered data, including email content, passwords, photos and chat protocols, between 2008 and 2010, and he called the situation “one of the biggest data protection rules violations known.” Matussek points out that under German law, the maximum fine for violating data rules is 150,000 euros.

In a post at ArsTechnica, Cyrus Farivar notes the “paltry” amount of the sum. Quoting the New York Times piece on the fine, Farivar writes, “while the fine is the largest ever issued in Europe concerning Google’s actions, it ‘amounts to 0.002 percent of Google’s $10.7 billion in 2012 net profit.'” Caspar said in the statement that “[a]s long as violations of data protection laws are punishable by discount rates, the enforcement of data protection laws in a digital world with its high potential for abuse will be all but impossible.” Caspar suggested a maximum fine of 2% of a “company’s annual turnover” would “enable violations of data protection laws to be punished in a manner that would be felt economically.”

Tip us off

News tips and suggestions are always welcome, so please send them along.

Related:

• Raising the world’s data privacy IQ
• What does privacy mean in an age of big data?
• Data in the HR department
• Bringing big data to work
• More Strata Week coverage