To better place graduates, schools tap data to parse employers’ labor needs;

‘what’s a computer programmer?’


Despite Macomb Community College’s long ties to Michigan’s auto makers, the school’s leaders sometimes feel as though they aren’t speaking the same language as employers when it comes to matching its graduates with jobs.

Turns out they often aren’t: When the school’s engineering dean examined thousands of local job listings, he learned manufacturers used dozens of job titles for the same mix of skills Macomb calls mechatronics—a combination of mechanics, electronics and programming. The analysis, using real-time data scraped from online job boards, helped Macomb identify prospective employers for its graduates, based on the skill sets they specified.

“If I could wave a magic wand, what I’d like is to have clear definitions…from the company saying, ‘Here’s what we want a machinist to do, and if they can do these and these things, we’ll hire them,’ ” said Macomb President Jim Jacobs. “That doesn’t exist.”

Matching workers with jobs has never been easy, stymied by an information gap between employers’ specific needs and workers’ skill sets. Educational and training institutions are trying to narrow the divide using data gleaned from online job listings to more efficiently update and align their offerings to meet the demands of a swiftly changing labor market.

“The problem right now in the business community is these employers don’t speak with one voice,” said Peter Cappelli, the director of Wharton’s Center for Human Resources at the University of Pennsylvania, who has studied skills mismatch. “For the most part, they don’t get close to schools.”

It’s a problem mostly seen in the U.S., say labor-market experts, thanks to a rapidly evolving economy and a divide between the country’s educational institutions and employers that isn’t there in other advanced economies. In Germany and Denmark, for example, the two groups collaborate to ensure training and apprenticeships lead to jobs after graduation.

The gap has helped push U.S. job vacancies to 5.5 million, near historic highs. For most of the past year the number of job openings has exceeded the number of new hires, a reflection of employers’ difficulty in filling positions.

“Job change is higher here than in any other country in the world,” said Anthony Carnevale, director of the Georgetown University Center on Education and the Workforce. “It’s a very dynamic system and people need road maps now.”

So more educational and training institutions are turning to granular data from job-listing services for detailed and timely insights they can’t always get directly from employers. While the Labor Department gathers data on broad occupational groups, online job postings can be parsed by the skills, credentials and degrees in highest demand. Firms such as Burning Glass Technologies, LinkedIn and CareerBuilder analyze huge sets of job postings to assess demand for specific skills.

The federal government “tracks jobs like ‘computer programmer,’ ” said Burning Glass Chief Executive Matthew Sigelman. “What’s a computer programmer? I run a software company and I’ve never hired one.” Instead, Mr. Sigelman says, he seeks people who know specific programming languages such as Ruby or C++.

The Boston-based firm, founded in 1999, tracks an average of 3.6 million unique job openings scraped from roughly 40,000 job sites on any day. Its 7,500 subscribers, largely educators and workforce-development agencies, can break down listings by 70 different elements including skill clusters, location and certifications.

Northeastern University in Boston used a deep dive into the data to shape offerings at U.S. satellite campuses in Charlotte, N.C., Seattle and San Jose, Calif., and identify locations for its next hubs.

To identify skills and credentials in highest demand, a seven-person team layers traditional labor-market and population forecasts over real-time data on job openings in specific fields along with counts of how many people received relevant degrees.

A review of Bay Area job postings uncovered over 16,000 positions specifying data-analytics skills, so Northeastern in January started a two-month boot camp on data analytics for people who don’t want or need a full master’s degree in computer science.

“It’s not rocket science to see there’s an issue here,” said P.K. Agarwal, regional dean and CEO of Northeastern’s Silicon Valley campus.

Lone Star College’s associate vice chancellor, Linda Head, said real-time labor-market data has helped strengthen the school’s relationship with Texas employers and find new firms with which to partner. Since she started using it four years ago to evaluate the school’s 50 or so associate degree programs, she said, annual meetings with industry partners have gone from 2½-day brainstorming sessions to 4-hour powwows.

“They’re very excited to participate with us in meaningful conversations about hiring our students because…they know we’re not going to waste their time,” she said.

For example, in early 2013, she used Burning Glass data to identify demand for a new computed tomography program for medical imaging. After confirming the need with industry focus groups, Lone Star launched the program in January 2014.

“We can start an associate degree program in nine months, versus two years of meetings and going back and forth and back and forth,” she said, referring to the old process of course development.

But not every educational or training institution can move as swiftly as Lone Star. Initiatives designed to close the gap often move too slowly, especially when public dollars are at work, said Northeastern economist Alicia Modestino, who studies real-time labor market information.

When she tells career counselors that the future of job training should be “nimble and targeted, they laugh out loud,” she said. That’s because workforce grants are often directed by state or federal government priorities, such as whether a worker is a veteran or disabled, she said.

That points to a deeper mismatch between the nation’s training system and its swiftly evolving economy, one that’s especially acute in fields like technology.

Kris Stadelman, director of the NOVA Job Center in Sunnyvale, Calif., said she sees community colleges struggling to keep up with the rapidly changing tech sector in her area, which encompasses Silicon Valley.

“By the time [a new program is] approved, it’s already obsolete,” said Ms. Stadelman. “The market has changed dramatically, and the infrastructure hasn’t kept up.”

She often recommends laid-off engineers and tech workers pick up new skills through online training courses, but she’s unable to use government grant money to help workers cover the cost of for-profit offerings like online training resource or at one of the hands-on tech workshops in the Bay Area.

Michigan’s Macomb Community College, in a state hoping to reinvent its manufacturing sector, is trying to adjust more quickly. Granular job data allows its engineering school to cater to area employers’ needs by refining its curriculum so as to match up to 85% of their most commonly required skills.

“People need to appreciate the chaos that goes on in the private sector,” said Mr. Jacobs, the college’s president. “You have to ride the tiger to figure out what the hell is going on…It isn’t all that clear. Sometimes [businesspeople] don’t even know.”

Write to Anna Louie Sussman at [email protected] and Melissa Korn at[email protected]