Our Colleagues’ Decisions May Influence Us More Than We Realize

But peer influence doesn’t stop at childhood. It extends to the workplace as well, shaping how professionals make decisions.

Indeed, a growing body of research shows that professionals in a variety of industries are influenced by those around them. Physicians, for example, are influenced by their peers when writing prescriptions; sales teams become more productive when working alongside productive peers; and biomedical scientists are shaped by their peers’ innovation and entrepreneurial choices.

So, which factors in the workplace are driving this peer influence?

Jillian Chown, an associate professor of management and organizations at Kellogg, teamed up with Carlos Inoue of Gies College of Business to answer this question in the context of a high-stakes work setting: the maternity wards and operating rooms of Brazilian hospitals. Specifically, they examined how medical doctors trained in obstetrics and gynecology make the call to perform a birth via Caesarean section versus vaginal delivery. They centered their study on the public health system in Brazil, which has one of the highest rates of caesarian births in the world; 56 percent of births are delivered via C-section, significantly higher than the rate of 15 percent recommended by the World Health Organization.

Chown and Inoue investigated how much peer influence shapes these critical medical decisions.

“We realized that if physicians have this tendency to be more or less aggressive in terms of opting to perform a C-section, that tendency might spread to their colleagues around them,” Chown says. “Maybe in some instances it’s good and in some it’s bad, but right now we just don’t know a lot about how peer influence happens in this type of context, or its consequences.”

To that end, the researchers analyzed detailed data covering more than five million births performed by 16,500 physicians across 915 public hospitals in Brazil.

They found that when doctors worked alongside a peer who tended to perform more C-sections than expected, their own likelihood of performing the procedure also increased. Furthermore, doctors were more susceptible to a peer’s influence under two circumstances: if they had similar work responsibilities to that peer, and if they had a greater variety of day-to-day responsibilities than most doctors in the same role.

“The idea that expert decision-making is malleable is somewhat disconcerting,” Chown says. “That it could be so swayed by contextual factors like who a person is working alongside, or how their work is structured, raises some interesting and potentially troubling questions about the nature of expertise and decision-making in these types of jobs.”

Quantifying physician practice styles

Brazilian public hospitals turned out to be an exceptionally well-suited setting for studying peer influence. The rich, detailed data from the Brazilian Ministry of Health allowed the researchers to closely track physicians’ work tasks (captured in claims records), as well as the details and circumstances of births from 2012 to 2017.

Using this highly detailed dataset, the researchers built a statistical model to determine how likely a C-section was for each birth. The model incorporated factors including the mother’s age, whether she’d had prior C-sections, whether the pregnancy included multiple babies, and fetal position. The model confirmed what clinical guidelines suggest: older mothers, those with prior C-sections, and those carrying multiple babies were more likely to receive a C-section.

Then, Chown and Inoue constructed a second statistical model that estimated each physician’s personal tendency to perform C-sections. This model controlled for the characteristics of the mother and pregnancy, allowing them to assign each physician a “practice style” score between zero and one—with higher values indicating a stronger preference for C-sections.

“We could take, for example, the hundred births that a given doctor performed during our sample period and say, ‘We would’ve predicted 50 of these to be vaginal deliveries based on medical factors, but for this doctor, only 20 ended up being delivered vaginally,’” Chown explains. “That told us some doctors were systematically more aggressive about performing C-sections while others were more conservative.”

The ingredients of influence

Through their analyses, the researchers found that physicians were more likely to perform C-sections when working a shift with a colleague who was more aggressive in their practice style (i.e., they were more likely to perform a C-section than their model would predict). Specifically, physicians were one percentage point more likely to perform a C-section for every 0.06 increase in their proximate colleagues’ practice-style scores. Over the five-year study period, this translated to an additional 53,000 C-sections out of nearly five million births.

In addition, two key factors magnified this peer influence: task variety and task similarity.

First, physicians who performed a wide variety of medical procedures—including clinical consultations and other surgeries—were more susceptible to peer influence. Chown and Inoue measured task variety based on the extent to which physicians’ work was spread across different tasks, where a higher score indicates a larger variety of tasks. The effect of peer influence was 73 percent higher (4 percentage points) when task variety was one standard deviation higher.

Why? The researchers theorize that doctors with a broad range of responsibilities may feel less certain about any single type of procedure, making them more likely to be influenced by their colleagues.

Second, the more similar a doctor’s day-to-day tasks were to those of their peers, the stronger the peer influence.

The researchers measured this by creating a “task vector” (or representation of tasks) for each physician and then measuring how similar those task vectors were for physicians working on the same shift. The effect of peer influence was 95 percent higher (3 percentage points) for every one-standard-deviation increase in task similarity. In other words, the more closely physicians’ day-to-day tasks resembled the tasks of a peer, the more likely they were to be influenced by that peer. This pattern, the researchers say, may stem from the increased opportunity that colleagues with similar duties have in sharing information and practice styles.

Finally, peer influence was particularly strong when the best course of action—C-section versus vaginal delivery—was uncertain. In these ambiguous cases, it appeared that physicians were actively seeking input from peers to guide their decisions. “When work is uncertain,” the researchers write, “it can heighten an individual’s motivation to seek out others who can provide advice.”

Influence as a force for good

Peer influence can be a powerful force—shaping not only individual decisions, but also broader workplace practices. Whether its effects are positive or negative often depends on how organizations structure their work and interactions. And Chown believes that it’s often up to the leaders to decide which it will be.

“One thing leaders can do is be intentional about who is influencing whom,” she says. “By carefully structuring teams and assigning roles, organizations can encourage positive peer influence and potentially limit the spread of less-desirable behaviors.”

For example, in maternity wards, if the goal is to reduce unnecessary C-sections, hospital administrators could “ensure that doctors whose practice styles favor vaginal deliveries are more involved in training new physicians or are assigned to shifts where they will have the greatest influence,” Chown says. Conversely, “physicians with a stronger preference for C-sections might be assigned to high-risk cases where it’s known ahead of time that a C-section will likely be required.”

These insights can apply to many workplaces beyond healthcare, from corporate teams to law firms and investment banks. Managers can strategically shape peer influence by designing work environments that foster positive learning and best practices.

By thoughtfully harnessing the power of peer influence, Chown says, organizations can shape behaviors, improve decision-making, and drive better outcomes—not just in medicine, but across industries.