McCormick Magazine

Ethnographic engineering

Paul Leonardi learns by watching others work

Email::

paul leonardiAs an undergraduate, Paul Leonardi studied in Spain, learning the language, seeing the sites — and eventually asking himself a question that would shape his career.

“I noticed that many people in Spain were using cell phones, while in the United States at the time, I didn’t know anyone with a cell phone. People there would ask for my cell phone number, which I didn’t have, and I would ask for their e-mail address, which they didn’t have,” Leonardi says. “I wondered why it was that everyone I knew in the United States used e-mail while everyone in Spain seemed to use cell phones.”

What started as a simple question sparked an interest in the intersection of culture and technology: why we adopt some technologies and not others, and how we might be able to implement even better technologies. It was a question that led him to graduate school — first a master’s program and then a doctoral program in management science and engineering at Stanford — before it brought him to Northwestern, where he is the Breed Junior Professor of Design in McCormick’s Department of Industrial Engineering and Management Sciences and the Department of Communication Studies at the School of Communication.

In his quest to understand how technologies, cultures, and organizations interact and affect each other, Leonardi quickly focused on engineers. “I wanted to determine how engineers could develop technology better and how organizations could be designed to use technology better,” he says.

Much of his recent work has been in the auto industry. He studies how automotive engineers use finite element analysis and simulations to improve their efficiency and how organizational changes can help make new technologies more effective.

To begin his research, Leonardi watches engineers in action, following them to meetings, watching over their shoulders as they work, and conducting interviews and surveys, all the while taking copious notes. “As an ethnographer, I spend a lot of time observing people doing their everyday work to better understand their constraints and opportunities,” he says. “I go into a company and spend six months to a year observing engineers.”

Leonardi’s subjects are quite willing to share information and often provide meaningful insights. “If you think about it, people spend the majority of their lives at their workplace. How often is it that you go home and someone asks you with genuine interest, ‘What did you do today?’ and actually wants to know every little detail?” he says. “People relish the opportunity to talk about their work, to show you what’s good and what frustrates them on a daily basis.”

Leonardi uses this data to understand how organizations can better use new technologies. “Some of the most rewarding experiences for me are giving presentations to senior management at the companies I study,” he says. “I tell them, ‘Here’s what your engineers do every day, and here is why they perform these activities.’ They often look at me and say, ‘Are you kidding?’ I can show them the evidence, then make suggestions to develop new tools or to refocus the engineers’ work. Managers may think that the engineers are just afraid of new technology or stuck in old ways, but ethnography allows us to step back and look at the principled reasons why they might reject a technology.”

Leonardi is working to revitalize behavioral sciences research in industrial engineering and management sciences. Along with Noshir Contractor, the Jane S. and William J. White Professor of Behavioral Sciences, he is building strong connections with the School of Communication and with the Kellogg School of Management. He says behavioral science is an area that allows industrial engineers to make more meaningful contributions.

“My colleagues in IEMS need to make assumptions about how people behave — in order to create models,” Leonardi says. “We’re interested in gathering empirical data to provide a concrete understanding of what people are likely to do and why they might do those things, which makes those assumptions more accurate and, ultimately, will lead to more useful models.”

—Kyle Delaney