Fear of appearing prejudiced can inhibit accurate performance feedback to women

PULLMAN, Wash. – Evaluators who want to avoid appearing prejudiced may overcorrect and give women inflated performance feedback, new research indicates, which is a practice that could ultimately hinder their ability to improve and advance.

A Washington State University-led research team investigated the connection between overly positive performance reviews and “protective paternalism,” the belief that women need to be handled carefully and shielded from harm.

While it may be well-intentioned, evaluating women this way could hold them back by causing them to miss growth opportunities that might otherwise advance their careers, said Leah Sheppard, lead author of the research published in the Journal of Business and Psychology.

“If women don’t get honest feedback, they’re only going to fall further behind. That’s the clear, negative outcome,” said Sheppard, a WSU Carson College of Business researcher. “It’s never going to be the right thing, or even the kind thing to do, ultimately, to spare somebody from the obvious areas in which they need to improve.”

Previous research has found that reviewers tend to tell “gendered white lies” and inflate the feedback they give to women. To better understand this tendency, Sheppard and colleagues at Colorado State University first looked to see if the inflation was apparent in examples of real-world reviews.

Since internal workplace evaluations are usually confidential and difficult to access, the researchers used publicly available music reviews of men and women solo artists. Analyzing three years of reviews from Rolling Stone magazine, they found that the word count devoted to positive feedback was higher for reviews of women than men on albums given the same star rating.

In a second two-part study, they delved into when and why this occurs with a group of 486 participants, divided roughly in half by gender. First, participants took a survey designed to measure their motivation to appear nonprejudiced toward women. This included statements assessing internal motivation, such as “Being nonprejudiced toward women is part of my self-concept,” and external motivation: “Because of today’s politically correct standards, I try to appear nonprejudiced towards women.”

Next, participants viewed a performance description of a hypothetical, gender-neutral employee and then picked from a list of positive and negative statements as representing the most honest, accurate feedback for the employee.

About a week later, the participants viewed the same performance description, but this time, the employee had a male or female name. Participants were asked how they would feel when delivering feedback, including whether they felt they should protect and shield the employee from harm. Finally, participants were asked to select from the same feedback statements used in the ungendered scenario.

The researchers found that when participants expressed a stronger external motivation to avoid appearing prejudiced, they reported feeling a greater desire to protect the woman employee from harm. This “protective paternalism,” in turn, predicted delivering feedback that was more positive than they had given in the ungendered scenario. This was true whether the participants themselves were women or men.

The findings suggest that if not implemented well, some diversity efforts could backfire, the researchers noted.

If a diversity, equity and inclusion program is creating an atmosphere where people are afraid to criticize the job performance of certain groups of people, that could ultimately hurt the careers of the people those efforts are most intended to help, said Sheppard.

“We are starting to see more conversations around what we’re doing with DEI programs. Are we actually making a better space for people that have been historically marginalized? If we’re not doing that, and we’re actually just instilling fear in people, we’ve got to kind of go back to the drawing board and figure it out,” she said.

For workplace performance evaluations, Sheppard suggested that future research could investigate whether giving reviewers a short text on the importance of negative feedback might help prevent the delivery of inflated feedback.

Another potential solution might be to reframe the entire evaluation process itself.

“We have to do a lot better at performance reviews, and maybe even get away from some of the language traditionally used around it: so instead of a scary ‘performance appraisal,’ maybe it’s coaching or mentoring,” said Sheppard. “It could be a way to normalize mistakes. Everybody has areas in which they falter and could improve.”