When Our Work Is Disrupted, the Story We Tell Matters

So perhaps it is no surprise that large-scale disruptions like the Covid-19 pandemic can have an even bigger impact. One study has shown, for example, that increasing childcare duties because of daycare and school closures during the pandemic reduced the amount of time women scientists with young children were able to dedicate to their research by 30 to 40 percent.

Even in ordinary times, women—who disproportionately serve as caregivers in American society—are often underrepresented in up-or-out fields. So when the pandemic hit, many worried that the disruption might reshape the composition of these professions. In response to this concern, many higher-education institutions introduced the Covid-19 Impact Statement—a document in which faculty being reviewed for tenure have the chance to explain how the pandemic affected their productivity.

But not everyone was convinced that the Covid-19 Impact Statement would be an unmitigated boon for workplace equality.

“People were talking about how the solution may be these impact statements, but as a researcher who studies the expression—or interruption—of bias, I was really concerned,” says Lauren Rivera, a professor of management and operations at Kellogg. After all, might highlighting someone’s caregiving responsibilities potentially reinforce the idea that they are less devoted to their work?

Rivera teamed up with Northwestern’s Katherine Weisshaar, an associate professor of sociology, and the University of Toronto’s András Tilcsik to investigate how these Covid-19 Impact Statements were perceived. To put the tool to the test, the researchers evaluated the results of an original survey experiment completed by over 600 full professors in science, technology, engineering, and math (STEM) fields, with a focus on the biological and physical sciences.

They found that the Covid-19 Impact Statement could, in fact, help level the playing field for caregivers in academia. They also confirmed that it does not appear to induce bias in the way Rivera thought it might. Instead, the tool shines a light on the ways in which personal narratives, and how they’re contextualized, can help cut through bias.

“We know from other disciplines that the stories we tell matter, and they affect how people view us,” Rivera says. “And what we show in this paper is that they are powerful tools that can be used, in the right circumstances, to increase equity.”

Comparing work disruptions

Rivera was well positioned to investigate the nuances of the Covid-19 Impact Statement and similar tools. Through her research, she had long analyzed the ways in which workplaces evaluate merit—and how these evaluations can perpetuate social inequalities. She was well aware, for example, of past research that has shown how disclosing information about caregiving responsibilities can backfire.

“We’ve seen that in white-collar professional occupations in the U.S., and in academic science in particular, there’s a stigma against anything in your life that is not total devotion to science and work,” she says.

The researchers chose to disseminate their survey among professors in STEM disciplines in large part because these fields tend to require research work in laboratories, many of which shuttered right alongside childcare centers and schools during the pandemic. In the context of Rivera’s and her coauthors’ study, those lab closures were crucial because they enabled the researchers to compare caregiving challenges with a totally different kind of productivity disruption.

This comparison became a central part of the experimental survey’s design. First, the STEM professors who responded to the researchers’ survey read a tenure application from a fictional assistant-professor candidate. The candidate’s profile was intentionally designed to read as “on the cusp,” Rivera says—potentially ready for tenure, but not necessarily so. The fictional application noted that the candidate had been allowed to submit a Covid-19 impact statement with their application.

Next, professors who responded to the survey were asked to weigh in: Should this candidate be granted tenure?

The researchers distributed three slightly altered versions of the profile in their survey. In one version, the candidate did not include an impact statement; in another, they included a statement describing a childcare disruption; and in a third version, they included a statement about a laboratory disruption. The researchers also varied the candidate’s perceived gender by changing their name: some of the applications presented a candidate named Jennifer Nelson; others, Michael Nelson.

The survey posed a few additional questions as well, requesting that the professors assess the candidate’s current and future productivity, the candidate’s commitment to their work relative to similar faculty members, and the extent to which they would advocate to tenure the candidate if the committee disagreed over the decision. There was also an open-ended answer section, where professors could offer more-extensive explanations about their decisions.

“One of the reasons we included the open-ended section is because, in a survey, you can administer various measures, but you’re making assumptions about what matters to people,” Rivera says. “We wanted to include a space where respondents could express their own rationales, and capture things that we, as researchers, might not have expected.”

Making a statement

Across the board, a clear pattern emerged: the inclusion of any Covid-19 Impact Statement yielded significantly more positive tenure recommendations compared with statement-less applications.

The professors evaluated the fictional candidates on a scale of 1 (“definitely do not promote”) to 6 (“definitely promote”). On average, candidates with no statement received a score of 4.54, between “maybe promote” and “probably promote.” But candidates who included a statement detailing a lab closure received a higher average score of 4.73, and those who included a statement pertaining to childcare closures received an average score of 4.87. There was a statistically significant difference between the scores of candidates with a statement compared with those without a statement, but not between the different types of statements. A caregiving-related disruption, in other words, was deemed just as justifiable as a lab closure.

The same patterns held true regardless of candidates’ gender.

Similarly, the survey’s follow-up questions didn’t yield notably different outcomes for the two types of statements, as the researchers had initially suspected they might. “It is possible that a Covid-19 statement detailing extensive childcare obligations could yield a more-favorable tenure recommendation but generate a more-negative perception of a faculty member’s long-term research trajectory or commitment to science,” the researchers write in the paper. But in the end, their results revealed almost the opposite.

Indeed, the researchers found that, in the open-ended-answer section, the professors ended up using similar, understanding language when describing their decision to grant tenure to candidates with an impact statement. They underscored that the pandemic was now “in the past,” and that the disruption it had caused was “legitimate.” They also emphasized that the problems it caused were “outside of the candidates’ control” and were “external” to those candidates.

“It was almost like this particular cognitive framing enabled people to interpret the presence of disruption in a way that did not activate caregiving and gender biases,” Rivera says.

The stories we tell

Rivera and her colleagues continue to research the impacts that framing hardships and disruptions as temporary factors outside of one’s control can have in the workplace.

They plan to look beyond Covid-19 Impact Statements and at a wider variety of settings, keeping the following question in mind: “Under what conditions do the stories we tell about these disruptions level the playing field—or tilt the playing field more?”