2 UW professors changed the way we look at workplace turnover

A few years ago, Terence Mitchell was trying to recruit a professor he admired to the faculty of the UW Foster School of Business. Both of them agreed that the job was a perfect fit. But the professor wouldn't budge.

“He said, ‘I’d love to come, but my son is a senior in high school next year, and I’ve got three doctoral students who are going to be finishing up in the next two years. I just can’t go anywhere,’” Mitchell recalls. “It didn’t have anything to do with the fact that he would’ve been more satisfied here than there. It was not a satisfaction decision. It was a ‘stuck’ decision.”

Mitchell was disappointed, but he wasn’t surprised. On the contrary, the recruit’s immobility was just one more example of a phenomenon Mitchell has spent the past two decades documenting: People often stay at their jobs, or leave their jobs, for reasons wholly unrelated to job satisfaction.

This may sound like common sense. But until Mitchell (at left above) and Thomas Lee (right), both professors at the Foster School of Business, laid it out in a series of influential papers, it was completely lost on the leading scholars of organizational behavior.

The “standard model” in the study of job turnover had long held that people who dislike their jobs tend to move on when they find better ones, and that people who are happy where they are tend to stay there. Mitchell and Lee demonstrated that leaving is more often the result of what they call “shocks”—sudden circumstantial changes—than dissatisfaction, and that the departure is sometimes abrupt.

Subsequently, they showed that staying has a lot more to do with “embeddedness”—the sort of personal connections that kept Mitchell’s recruit from coming to the UW—than with professional contentment. Their research has permanently changed the vocabulary of management research, and has helped make the UW one of the field’s epicenters.

“There’s probably no place where two bigger names exist side-by­side,” says Carl Maertz, a professor of management at St. Louis University and an expert on turnover. “There’s no place where you’ve got two people who have made such huge contributions to the big picture.”

In the world of business, turnover is both a timely and a timeless concern. Managers are always looking for ways to reduce attrition, because the cost of seeking, hiring, training and assimilating new employees is so high. One recent study determined that organizations lose 17 percent of their pre-tax income to employee withdrawal (a category that includes turnover as well as absence, lateness and withheld effort). Another estimated that the departure of a single experienced professional costs a management and consulting firm more than $1 million.

But the need to control turnover is about to become even more acute, as the baby boomers begin to retire. The American economy is now bracing for what Wendy Harman, ’06, an assistant professor of business at Truman State University, has called “the largest brain drain the world has ever experienced.”

“Certainly within the last 20 years we’ve seen a refocus on the topic [of turnover], mostly because organizations are very concerned about keeping their best and their brightest, and the competition for talent is extensive,” Mitchell said in a recent podcast interview for an Academy of Management Web site.

Yet for the longest time, researchers seemed to be satisfied with a primitive and not very nuanced understanding of why people leave—the standard model.

Mitchell and Lee were two exceptions. By the late ’80s, both had become frustrated with the field’s stagnation. “People were doing the same thing over and over again,” recalls Lee, who is also associate dean for academic and faculty affairs. “They were trying to get incremental tweaks, squeeze more information, out of a very common-sense, statistically valid idea: If you are job-dissatisfied and you have an alternative place to go, you have a high likelihood of voluntarily leaving. And to me it was very frustrating. So I thought about it and I thought about it. And then in 1990 I got tenure, and I thought, ‘Gosh, this is a good time to take a chance.'”

Around the same time, the field’s methodology was being transformed by a new tool called “meta-analysis”—a way of synthesizing the results of multiple studies in order to get a highly accurate sense of the relationship between two variables. “We did meta-analyses on things like job satisfaction and turnover, and the relationships were awful,” Mitchell says. Knowing how satisfied an ex-employee had been with a job, they learned, gave them roughly a one-in-10 chance of knowing why that person had left.

So Mitchell and Lee went hunting for other reasons people might quit, conducting broad-based surveys, in-depth interviews and reviews of existing literature. They focused principally on nurses and public accountants—people who can pretty much leave their jobs anytime they want because their services are needed everywhere.

(Nothing is certain, after all, except death and taxes.)

In 1994, Mitchell and Lee fired a shot heard ’round the management world—a paper in the field’s leading journal, modestly titled ”An Alternative Approach: The Unfolding Model of Voluntary Employee Turnover.” In it, they argued that the standard model was merely one of four “paths” by which people left their jobs, and that the other three all involved “shocks”—precipitating events that might be expected or unexpected, positive or negative.

Path 1, they wrote, involves a plan. The employee has decided to work until some anticipated date or event comes to pass—a pregnancy, admission to graduate school, the start of ski season. Conversely, the “shock” in Path 2 is something genuinely shocking—what Lee calls an “abhorrent event,” such as pressure from a boss to do something illegal, a fight with a co-worker or a catastrophe unrelated to work. Most of the people who left their jobs in New York City after Sept. 11, 2001, Lee notes, were leaving via Path 2.

Path 3 is akin to the standard model, but with the dissatisfaction triggered by a shock—a feeler from a prospective employer, an encounter with a more successful colleague. In effect, the employee doesn’t realize he or she is dissatisfied until the shock forces a stock-taking. “It’s not job dissatisfaction, but relative dissatisfaction,” says Lee. The standard model itself is demoted to Path 4.

In the 13 years since Mitchell and Lee introduced the unfolding model, it has assumed a central place in the scholarly conversation about turnover. Additional research has bolstered it, built upon it and modified it to varying degrees, but never really challenged it. “There really isn’t a competing model right now,” Maertz says.

In the wake of this success, Mitchell and Lee began thinking about what the next step ought to be, and in 1996 they had something like an epiphany: Mitchell has been working at the UW continuously since 1969, Lee since 1983. “We looked at each other and said, ‘What do we know about turnover?’” Lee recalls. “What we know about is staying.”

Here they saw an area in which they might make an even more meaningful contribution. Historically, staying was non­existent as a subject of serious inquiry. If it was discussed at all, it was discussed only as an epiphenomenon of leaving:

“Why did you stay? Because you didn’t leave,” is how Lee describes it. Having given the standard model such an extensive remodeling, he and Mitchell felt it was time to update this simplistic formula as well. Once again, they went out and began interviewing people.

“We’d say, ‘Are you unhappy? Or do you like your job?’ ” Mitchell recalls. ”And the thing we’d hear over and over again was, ‘You know, it really doesn’t matter. Sometimes I’m happy about it. Sometimes I’m not so happy about it. The problem is, I can’t leave. I’m stuck.’ And that’s not accounted for by the traditional model either, which says that satisfied people stay and dissatisfied people leave. It’s not necessarily satisfied people who stay, it’s often stuck people who stay.”

Mitchell and Lee were careful to use the more neutral­sounding term ’embedded’ when they published their first paper on the subject in 2001. Not everyone who’s stuck in a job is unhappy with it. On the contrary, Mitchell says, job satisfaction correlates strongly with staying, because people have positive feelings about many of the things that tie them down—perks, salary, long-time friends, comfort with their professional identity. But embeddedness itself—that web of personal and professional connections it would be too traumatic to break—is the best predictor, better by far than satisfaction, of whether a person will stay or go.

So, for the second time, Mitchell and Lee had introduced a paradigm-changing idea to the field of organizational behavior.

“Embeddedness is liable to make at least as many citations in the future as the unfolding model has,” Maertz says.

“There were tons of papers at the last Academy of Management conference that had ’embeddedness’ in their titles—like 20 or something.”

Maertz adds, for the record, that he considers both models somewhat incomplete—he and a colleague published a paper not long ago, for example, arguing that the unfolding model should be expanded to include not just four paths, but eight. But it’s a measure of Mitchell and Lee’s success, he says, that even scholars who disagree with them are disagreeing with them on their own terms.

Their research is also having an impact on the business world. Not long ago, a company hired Mitchell and Lee to help institute “embeddedness strategies,” such as mentoring and incentives.

Likewise, Mitchell and Lee have helped companies apply the lessons of the unfolding model. One major bank was losing too many employees after just a few months. Mitchell and Lee discovered that these short-timers were, in effect, taking the money and running—they had come in with a plan to complete an initial three-month training program, collect the $2,000 training bonus and quit. At Mitchell and Lee’s recommendation, the company began withholding that bonus until employees had been on the job at least a year.

If that sounds like a simple solution to a simple problem, it is. But as Lee and Mitchell have spent their careers demonstrating, there’s a great deal of difference between the simple and the simplistic. Lee says he can’t speak for other disciplines, but in business virtually all groundbreaking work has one thing in common: ”After somebody says it, it’s so obvious. But until somebody says it, it’s not obvious at all.”

As a fan of such elegant simplicity, Lee says he’s particularly proud of his own job description—the short version of it, anyway.

”I’m a professor,” he says. “People expect me to have all these highfalutin’ polysyllabic words. And they say, ‘What do you study?’ And I say, ‘Why people leave, why people stay.’ And they say, ‘Oh.’”

Business suits

When a person and a job are well matched, business professors like the UW’s Terence Mitchell and Thomas Lee call it “fit.” And the perfect fit can be just as hard to come by in a career as it is in a suit of clothes—harder, even, because there are no fitting rooms or tailors with tape measures to tell you what you need.

But there is Career Discovery Week. Presented by the UWAA and the UW Career Center, this annual career extravaganza, with more than a hundred sessions on all three UW campuses, is designed to help students and alumni figure out what’s going to fit. The January 2008 installment will feature presentations on everything from “Careers in the Film and Entertainment Industry” to “Careers in Urban Studies” to “Why Software Developers Have the Best Jobs in the Country.”

Now in its ninth year, Career Discovery Week was conceived by a small group of academic advisers and career counselors as an opportunity for students to “learn more about careers of interest by hearing firsthand from UW alumni and friends working in those fields,” says Don Gallagher, associate director for career services at the UW Alumni Association.

But over the years it has grown into what Gallagher calls “one-stop shopping” for career-minded students and alumni alike, with panel presentations, networking events, career fairs and seminars designed to help participants improve their job search skills. Some of the programming is now targeted exclusively to UW alumni, both recent graduates looking to kickstart their careers and experienced professionals who are considering a change.

And Career Discovery Week has grown in duration, too—the “week” now begins in mid-January and runs for more than a month. All of this expansion has happened pretty organically, Gallagher says. “Growth happens when you combine an original great event concept with dozens of active campus partners.”

Virtually all Career Discovery Week events are free. Ses­sions begin Jan. 9, with the greatest number slated for the week of Jan. 28.

The UW Alumni Association offers a variety of year-round career services to its members, including the Husky Career Network, an exclusive online network of more than 5,000 successful UW alumni and friends who have volunteered to share their advice and contacts with fellow Huskies.