Recently, Carnegie Mellon University economist Kate Anderson found herself thinking about the intriguing employment history of a friend from college — a biology major who was also adept at bicycle repair, mountaineering and driving a forklift.

He ended up working for remote scientific field expeditions, where it was uniquely valuable to have both the ability to move heavy things around a work site in Antarctica or Tanzania and the laboratory skills to know “which things would explode if you dropped them,” she said.

His experience is a peculiar example of the kinds of skill combinations that are increasingly important — but challenging to quantify in modern labor markets.

In a new study, Ms. Anderson, an assistant professor at the Tepper School of Business, developed a network model that can represent not just individual skills but their relationships to one another and the value of those combinations in a labor market.

Advertise with us

Her method helps overcome a problem found in more one-dimensional ways of characterizing workers’ skill sets that came from the manufacturing era when you could tell a lot just by knowing a worker’s trade and how long he had done it.

The same thing “doesn’t work very well when you’re talking about biochemists or software engineers,” she said.

In the paper in the journal Proceedings of the National Academy of Sciences, she illustrated her approach by applying it to 26,000 worker profiles and 365,000 job listings on the freelance job website Upwork.

She mapped networks representing skills that workers have and skills that employers require.

Within them, she was able to distinguish specialists from people with a diversity of skills. Some people with diverse skills were jacks-of-all-trades who used their skills independently in a wide range of jobs while others applied their various skills in combination at a single job.

Advertise with us

When she examined how various skill sets influenced wages, she found that workers with diverse skills made more than specialists. The people who used their diverse skills synergistically made the most.

“They offer something that is useful and rare,” she said.

Her findings have implications for both employers and workers, as well as the online job sites that try to match them.

For example, the network tool can be used to improve the algorithms that online job search platforms use to sift through candidates. The goal would be to better connect employers with the types of workers they are really looking for.

If it were translated into a more user-friendly format, the tool could also tell workers — or people who help them find jobs — how to showcase skills that are valuable in a particular labor market or what training they might pursue to fill a coveted gap. Ms. Anderson provides detailed algorithms in her paper, but it would require knowledge of network science to apply them.

Advertise with us

“One of the big results of this is that adding skills doesn’t just add up linearly because you could have combinations that are greater than the sum of their parts,” she said.

“The skills that are going to be most valuable for you to acquire are going to depend on the skills you already have.”

Laura Legere: llegere@post-gazette.com.

This story was originally published with images that are no longer available.