If you think the workplace looks a lot different than it did a decade ago, it can be mind-blowing to consider the changes that have happened over the past century. In the area of turnover, there are a lot of things that have evolved, including industrial psychologists’ view on the factors that predict if you’ll stay or leave your job.
For the 100th anniversary of the Journal of Applied Psychology (JAP), Professor of Management Peter Hom and his research partners — Thomas W. Lee of the University of Washington, Jason D. Shaw of Hong Kong Polytechnic University, and John P. Hausknecht of Cornell University — reviewed hundreds of employee turnover studies conducted in the past century for “One Hundred Years of Employee Turnover Theory and Research.”
“We wanted to give a sense of what we know thus far and what [industrial psychologists] should focus their research on in the future,” Hom says. “We think we know a lot about turnover and what a manager can do about it, like monitor employee morale and try to address dissatisfaction, especially those reasons that tend to be correlated with turnover.”
For Hom, who became an industrial psychologist because of his interest in job satisfaction and how people find meaning in their work, the answers to turnover haven’t always been straightforward.
“It was a common belief among some that more satisfied people are more productive,” he says. “And so that was the assumption we widely tested by the time I entered graduate school — and the findings were quite mixed.”
When Hom began his work in the 1970s, employee turnover research was in the third of six epochs identified in the JAP review:
1917: birth of turnover research — JAP published its first studies documenting how employers stemmed quits with pay hikes and scholars speculated about why employees leave.
1920s-1960s: formative years — Researchers examined the demographic and psychological correlates of turnover, while exit interviews identified common reasons for leaving.
1970s: foundational models — Research by March and Simon and Mobley launched theory-driven research.
1980s: theory testing — Studies began looking at how family and kinship can affect turnover, and researchers began to look at who was quitting as well as the idea of organizational commitment.
1990s: unfolding model — A radically new theory, this model disputed assumptions about job dissatisfaction and specified four distinct turnover paths.
2000s: 21st Century Research — Research introduced the theories of job, community, and occupational embeddedness; and looked at turnover at the collective vs. individual level.
While each era had notable developments, from looking at costs and causes in the early years to developing models based on job satisfaction, it has been in the past 40 years that the research has changed the most.
One big shift that occurred in turnover research for Hom evolved out of a conversation and collaboration with fellow researchers Roger Griffeth, Tom Lee, and Harry Mitchell over a spring training baseball game in Arizona.
Lee and Mitchell convinced him that in researching only why people leave, he was missing half the picture. They told him he should also focus on why people stay, because why people stay may be for very different reasons than why they leave.
“That clicked finally,” Hom says. “They identified things we had hardly considered or had downplayed. One thing that was unique about their work was the idea of community embeddedness. The idea that people stay because their family wants them to stay or they stay because of community. Most turnover people focus mostly on the job and the organization.”
The idea of embeddedness — all the factors that keep you in your job, which can range from liking the town or your co-workers to being the spouse who carries health insurance — that Lee and Mitchell put forward in 2001 resonated widely among academics. “If you’re embedded in your job, either because your family wants you to stay or you like the sunshine, that’s a good thing. You tend to be more stable. But together we speculated, that if you’re embedded for the wrong reasons, for example, the money is too good to leave, you actually may not be a good employee, and you might ultimately leave once you have the opportunity.”
That new theory meant that it wasn’t enough anymore for managers to make sure people were happy with the aspects of their jobs they could control, like pay, schedule, or workload. They also need to consider factors in the community, employees’ families, and co-worker relationships among other things.
Some other changes over the past few decades that have challenged turnover thinking include:
Baby boomers — As baby boomers age out, the workforce is shrinking, says Hom. This is not only making turnover more rampant, it’s making it more important as an employer to keep the people you have. “Unfortunately, college graduates are not majoring in the same degrees in the STEM field. And therefore, if you get one, you want to keep them — or try to keep the baby boomer you have.”
Benefits trend — As more companies move away from defined benefits like pension plans and toward 401(k)s, it makes turnover more unpredictable, Hom says. There’s no set retirement time, so retirement is becoming more like turnover, he says. “Executives are nervous because they’re noticing that turnover rates for the 401(k) people are much higher than for employees with pension plans, because once you’re vested, you can take your money with you.”
Following the leader — Hom says one trend in turnover today is looking at how startups tend to be formed. “Startups often form by turnover when people leave companies to follow a leader,” he says. For example, Intel was created when Robert Noyce and Gordon Moore left Fairchild Semiconductor and many of their subordinates left to be with them. Hom is also looking into the theory that the same is true in large companies and what variables would moderate the leader departure effect.
Tips for understanding turnover today
So, what lessons can managers and employees learn from turnover trends? Here are a couple things Hom calls out:
- Expect idiosyncrasies — “People in our field talk about evidence-based management, and we don’t want practitioners to reinvent the wheel,” he says. “But the kinds of dissatisfactions that drive turnover might vary across companies or across the nation.” Particular features that motivate turnover tend to be unique, for example, in accounting, where there’s high turnover at the two-year mark because people get their CPA and then move on. In academia, when the question of tenure approaches at five or six years, turnover can peak.
- Use turnover tools — Employers can use validated selection tools, such as personality testing, to screen out applicants who might be more likely to quit. They also can use ongoing surveys to track employee attitudes over time, which can be a predictor of turnover.
- Find a good fit — Recent research has found that “fit” matters more than researchers once thought. Hom suggests the idea of comprehensive job previews, so employees can learn a lot about the job and its qualities before they start to help avoid reality shock and disillusionment, which are both common in new jobs.
- Embeddedness can be stressful — Employees who are “reluctant stayers,” meaning they’re embedded in a job they don’t like, can cause psychological and emotional distress. “If you’re embedded in a job where you’re abused by your boss, you actually have psychological ailments,” Hom says. “If you’re stuck because you can’t get a better job elsewhere or the pension is too good or you lose all your vacation time, you suffer mental stress.”
- Focus on the family — Hom says that psychological research of the past suggests turnover is a one-person decision, but it’s not — especially if you’re married. “It’s a decision that can affect others, so you have to be mindful that you just can’t quit because you’re dissatisfied,” he says. “It’s going to have to be a joint decision. You might have to quit jobs you like or stay in jobs you don’t like.” For example, leaving might impact the career of your partner or your family’s insurance coverage.