The second panel featured three presentations on the science of disrupting stereotypes in everyday life. Patricia Devine, the Kenneth and Mamie Clark Professor of Psychology at the University of Wisconsin–Madison, spoke about the ways in which individuals can overcome unintentional stereotyping. Mark L. Hatzenbuehler, an associate professor of sociomedical sciences and sociology at Columbia University’s Mailman School of Public Health, discussed stereotyping and stigma in social structures and institutions. Julie Sweetland, a sociolinguist and vice president for strategy and innovation at the FrameWorks Institute, spoke about her organization’s strategic approach to disrupting stereotypes. After the presentations, Rebecca Stoeckle, a vice president in the Health and Human Division of the Education Development Center, moderated an open discussion among the speakers and the workshop participants.
University of Wisconsin–Madison
Devine started by describing her interest in the paradoxical situation where individuals believe that prejudice is wrong, yet they continue to hold unconscious stereotypes. Stereotypes come to mind unintentionally and automatically, she said. As a result, stereotypes influence our impressions of people, the judgments we make about others, and the way
we interact with others. This creates a moral dilemma for people who renounce prejudice when they realize that they unintentionally contribute to bias simply by virtue of their subconscious thinking. Because of such stereotypes, even well-meaning individuals might constrain others’ opportunities, diminish the personal experiences of others, or otherwise unfairly disadvantage others, she said.
Unconscious Bias and Habits of Mind
Devine described unintentional biases as “habits of mind” that arise spontaneously and may be responsible for perpetuating bias even among people whose personal values reject prejudice. As an example, Devine cited the experience of social psychologist Tom Pettigrew who talked to American southerners about race issues. Pettigrew said that many white southerners confessed that although they no longer feel prejudice toward black people in their minds, they still feel “squeamish” when asked to shake hands with a black person (Goleman, 1987). “These feelings are left over from what they learned in their families as children,” he said.
Devine offered one of her student’s experiences as another example that emphasizes why unconscious biases are concerning. Devine said that the student came to the University of Wisconsin to study how to overcome bias. As it turns out, she said, there are many traffic accidents on the street corner outside the psychology building in Madison: students often cross against the light, and too often get hit by cars when doing so. One day, her student witnessed another student getting hit by a car and ran over to render assistance. Simultaneously, another woman came to help and began barking commands: “Don’t move the head. Call 911.” Devine’s student looked up at the woman and asked, “Are you a nurse?” As it turns out, the woman was a doctor. Devine explained that her student’s reaction to the experience was extremely negative. The student felt that she had violated her own morals, values, and principles, while also making the woman feel diminished in her role as a physician.
Overcoming Unintentional Bias
Devine posed the question, “Can we do anything about these habitual responses?” Her answer was “yes,” and she explained that using the “habit metaphor” is helpful when thinking about unconscious biases. She used a personal example—her habit of biting her nails—to illustrate her point: Although she had the motivation to stop biting her nails, that motivation alone did not provide her with steps toward breaking the habit. In order to break a habit, a person first needs to be aware of when that habitual behavior appears. Devine realized that her habitual nail-biting
occurred when she was writing and her fingers were poised above the keyboard. But even though she is now aware of when she is most likely to bite her nails, she said, simply being aware of the habit is also not sufficient to stop the behavior. To overcome habits, specific strategies are needed, which might include alternative responses or behaviors. In her own case, Devine said, she would place her fingers on the keys as a tactile reminder and an alternative response to her nail-biting habit.
Devine said that she has found that many people already have the motivation to change their stereotyping responses. As with the nail-biting example, after the initial motivation to break a habit is in place, it is necessary to help a person become aware of when he or she is likely to show bias. Devine said that this can be done by using the Implicit Association Test (IAT),1 which reveals people’s tendency to express automatic associations and unconscious stereotypes such as linking black people with negative information, linking white people with positive information, linking men with ideas of science or career, or linking women with ideas of the humanities or family life. Taking the IAT can be a very palpable and sometimes threatening experience for people, she said. The implicit biases that the IAT reveals show that we grow up in a culture where it is extremely difficult to avoid learning stereotypical associations commonly attached to major social groups. Devine said that media influences (e.g., television and the Internet) and observations of social interactions in everyday life activate stereotypical associations quite frequently. Children articulate these types of implicit associations as early as 3 or 4 years old—long before they make conscious decisions about their own personal values and beliefs. Essentially, people are set up to fall into the stereotyping predicament, she said.
Devine teaches about the kinds of processes and constructs that lead to the perpetuation of bias, such as the cognitive bias known as expectancy bias. Stereotypes, she said, tell us what to expect in others. Often stereotypes are prescriptive in that they include assumptions about how others should behave, and people become uncomfortable when others violate those expectations. This becomes a self-fulfilling prophecy, Devine said. “We interact with others in ways that draw out the very behaviors we expect to see.” Biases like these are very subtle, which is why they continue to be perpetuated over time. Devine emphasized that “knowledge is power” when it comes to implicit biases and stereotypes. Being aware of biases makes it possible to understand when a person is most vulnerable to acting on biases and when a person’s biases might manifest themselves in everyday life.
Intervention to Break Prejudice Habits
Devine has developed a toolkit intervention that includes several strategies for overcoming bias, including stereotype replacement, individuating others, taking perspective, considering situational explanations, and seeing opportunities for contact. She highlighted two of these strategies: individuating others, which involves seeking more information about others and taking others’ perspectives into account; and considering explanations for other people’s behavior that might be different from habitually biased expectations. Devine explained that her intervention builds on motivation, awareness, strategy, and effort. Though we might have the motivation for, awareness of, and strategies for disrupting stereotypes, we still need to work at actually disrupting the biases we hold, she said. Fortunately, she added, none of these tools are particularly difficult to implement, and, in fact, they make up a toolkit with synergistic effects.
Devine described her habit-breaking intervention aimed at disrupting stereotypes.2 She has found that people in the habit-breaking intervention group increase their concern about discrimination as soon as 2 weeks and for as long as 2 years after her intervention training, as compared to a control group that does not undergo the intervention. Devine has found that how participants become aware of their own biases and when they are likely to show bias are extremely important because these details become cues for individuals to use the habit-breaking toolkit to reduce their own biases. The awareness gained in the intervention is not limited to personal bias: it also helps trainees become aware of biases held by friends, family members, and the media. After training, people are able to recognize biased behaviors and label them as wrong, and this capability becomes a springboard for combating bias in the larger world. Devine has found that trainees in the intervention group are much more willing to challenge a biased behavior and point out that behavior as wrong even 2 years after the training.
Devine said she is often asked, “Why do you focus on individuals when stereotyping is such a big problem, and when there are systematic issues that are involved in addressing bias?” She explained that focusing on individuals is a major part of the solution to disrupting negative stereotypes, although it will not be the whole solution. She emphasized that people become agents of change first in themselves, and then in the
world around them. Devine described her research conducting cluster-randomized controlled trial designs in the science, technology, engineering, mathematics, and medicine (STEMM) departments at the University of Wisconsin (Devine et al., 2017). She said that these departments have historically suffered from an underrepresentation of women in the faculty. Half of the departments received training on the habit-breaking model of disrupting stereotypes, while half did not receive training. The departments that received training, she said, increased their efficacy in addressing gender bias, in both the commitment to addressing bias and the self-reported actions taken to reduce bias. Two years after the intervention, the departments that received training are hiring about 15 percent more women than was true at the outset of the research. This shows that individuals can be powerful agents of change both in themselves and in their environments, Devine said.
In closing, Devine said that prejudice is a habit that needs to first be broken in individuals and then dealt with in the larger social context. “It is a habit that can be changed to create a better overall, more welcoming, and inclusive environment for those who have historically been targeted by bias,” she said.
Mark L. Hatzenbuehler
Hatzenbuehler began by observing that “when thinking about stigma, we often imagine concrete events and experiences that happen to people such as hate crimes, bullying, or being the target of [negative] stereotypes.” However, he said, while these types of experiences are important, they represent only the “tip of the iceberg” in terms of how stigma operates in the world. Stigma is much broader than these concrete events and experiences; it is promulgated and reinforced through social institutions, laws, policies, and social norms and attitudes held about members of stigmatized groups. Such structural forms of stigma operate just below the surface of daily life, with profound implications for the health and well-being of members of stigmatized groups, Hatzenbuehler said.
Although structural stigma is embedded in multiple institutions (e.g., schools, the criminal justice system, media, and hospitals), Hatzenbuehler said he focused his remarks on institutional laws and policies for two reasons. First, laws and policies are clear targets for interventions aimed at reducing the structural forms of stigma. Second, research shows that public policies not only reflect social norms and attitudes held toward members of stigmatized groups, but they also help to shape those norms and
attitudes (Donovan and Tolbert, 2013). For instance, longitudinal studies have shown that when state governments pass laws banning smoking or banning same-sex marriage, prejudicial attitudes toward smokers and sexual minorities increase, respectively (Kreitzer et al., 2014; Pacheco, 2013). Hatzenbuehler and his colleagues have proposed that social policies have at least three different effects on stigma processes: policy can invigorate stigma (thereby producing harm), interrupt stigma (thereby mitigating harm), or willfully ignore the interests of stigmatized groups, which can exacerbate harm (Link and Hatzenbuehler, 2016). Hatzenbuehler presented illustrative examples of each of these three ways in which social policies affect health, using evidence from his research group.
To illustrate how policy can invigorate stigma and produce harm for members of stigmatized groups, Hatzenbuehler used the example of a quasi-experimental study that documented negative mental health consequences for lesbian, gay, and bisexual (LGB) respondents following state-level bans on same-sex marriage.
Between 2001 and 2005, 16 states passed constitutional amendments banning same-sex marriage. These events occurred between two data collection periods of a nationally representative survey called the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC).3 Respondents were interviewed in 2001 and re-interviewed in 2005, enabling researchers to examine changes in the prevalence of psychiatric disorders among LGB individuals before and after the bans. Hatzenbuehler and colleagues found that before same-sex marriage bans were instated in 2001, the prevalence rates of mood disorders were nearly identical between LGB individuals living in states that later passed a ban and LGB individuals living in states that did not (Hatzenbuehler et al., 2010). In 2005, following the passage of these bans, there was a 37 percent increase in the prevalence of mood disorders among LGB respondents living in states that banned same-sex marriage (see Figure 4-1). In contrast, no significant change was found in the prevalence of mood disorders among LGB individuals living in states that did not pass a ban. Furthermore, there was no significant change in the prevalence of mood disorders among heterosexual individuals living in states that banned same-sex marriage during the study period (see Figure 4-2), documenting specificity of the findings.
The second way that social policy can affect stigma processes is to interrupt them and thereby mitigate harm, Hatzenbuehler said. As an example, he described his research on state-level policies related to hate crime laws and employment nondiscrimination acts that specifically included sexual orientation as a protected class status (Hatzenbuehler et al., 2009). In this study, states with no protective policies were compared with states with at least one protective policy. Using data from the NESARC, state-level policy information was linked to individual-level mental health outcomes. The researchers found that LGB individuals living in states with no protective policies were nearly two and a half times as likely to meet criteria for dysthymia, a chronic mood disorder, than heterosexual individuals in those same states (see Figure 4-3). In contrast, LGB adults living in states with protective policies were no more likely to meet the criteria for dysthymia than were heterosexual individuals living in the same state. This pattern was similar across generalized anxiety disorder and posttraumatic stress disorder, as well as for psychiatric comorbidity, which is defined as the presence of two or more co-occurring psychiatric disorders.
The third way that social policies can affect stigma is by ignoring the interests of stigmatized groups through policy inaction. As one example of policy inaction, Hatzenbuehler noted that at the time of his presentation, no federal employment nondiscrimination act exists that includes sexual orientation and gender identity as a protected status. In the majority of the country, lesbian, gay, bisexual, transgender, and questioning (LGBTQ) people can therefore be legally fired based on their sexual orientation or gender identity, he said, because some 30 states do not include these factors as a protected status in their laws. Policy inaction can also occur when a policy is enacted but is then implemented selectively or not at all. Hatzenbuehler used the rollout of the Americans with Disabilities Act of 1990 (ADA)4 to illustrate this form of policy inaction. There are many ways in which the ADA had created positive outcomes for individuals with disabilities, he said. Nevertheless, a report released nearly two decades after the ADA’s passage noted various problems with the implementation of the law, which could ultimately undermine its efficacy (National Council on Disability, 2007).
Structural Stigma and Health Inequities
Hatzenbuehler described three ways in which structural stigma via laws and policies may be producing health inequalities. First, structural stigma appears to increase various risk factors of poor psychological health. To illustrate this, he described a study in which he and colleagues rated European countries, in part, on the number of policies that conferred protections based on a minority sexual orientation (e.g., relationship protection, hate crime laws protections, and employment nondiscrimination protections) (Pachankis et al., 2016). This rating was then linked to individual-level outcomes among men who have sex with men (MSM) living in Europe. MSM living in countries with fewer protective policies were more than twice as likely to report concealing their sexual orientation compared to MSM living in countries with more protective policies. Hatzenbuehler said that hundreds of studies exist showing that concealment of one’s stigmatized identity, including sexual orientation, is associated with a host of negative mental and physical health outcomes (Pachankis, 2007). This example suggests one psychological mechanism linking structural stigma to poor health, he said.
Second, structural stigma can also affect physiological health. Hatzenbuehler and his colleagues have conducted research showing that
4 Americans with Disabilities Act of 1990, Public Law 101-336, 101st Cong. (July 26, 1990).
structural stigma activates the body’s physiological stress response system, as measured by cortisol (Hatzenbuehler and McLaughlin, 2014). If chronically activated over the lifespan, research shows that this activation of the physiological stress response system can lead to poorer health outcomes. Hatzenbuehler said this research suggests a physiological mechanism explaining the relationship between structural stigma and health.
A third way in which structural stigma may affect health was suggested by research that finds structural stigma may undermine the effectiveness of health interventions. Researchers reanalyzed a meta-analysis of 78 studies that focused on individual-level HIV prevention interventions to improve condom use among African Americans (Reid et al., 2014). These interventions were conducted in communities across the United States that held varying levels of prejudice toward African Americans. In communities with the highest levels of prejudice, the effect size of the interventions approached zero, indicating that the interventions were not effective. In contrast, as prejudice levels decreased, the effects of the interventions became much more robust. These data provide some empirical evidence for why some health interventions might flourish in some communities and languish in others, Hatzenbuehler said, and they suggest that structural stigma is one mechanism underlying that difference.
While the research that Hatzenbuehler presented focused on stigma surrounding sexual orientation, he said he believes that his findings can be used to disrupt other kinds of stereotypes as long as interventions are appropriately adapted to different stigmatized conditions or characteristics, including mental illness and disability. Hatzenbuehler also said that focusing on structural stigma in laws and policies alone is not a sufficient target for interventions that seek to disrupt negative stereotypes. Stigma and stereotypes are multifaceted constructs, he said, so they require multipronged and multilevel interventions. Hatzenbuehler concluded by emphasizing the robustness of the relationship between social policies and health outcomes, and he noted that this relationship has been demonstrated across multiple methodological approaches, stigmatized groups, and health outcomes, as well as by multiple research groups. Taken together, he said, this work suggests that structural stigma represents one important target for multi-level interventions aimed at disrupting stereotypes.
Sweetland introduced the work of the FrameWorks Institute,5 a nonprofit organization that conducts and shares research to provide evidence-based strategies for reframing social and scientific issues. “The stories that we tell about social issues are the dress rehearsals for the policies that our society will eventually endorse,” she explained.
Sweetland described FrameWorks’ recent studies investigating the communications aspects of aging, in which an interdisciplinary team of researchers conducted individual and group interviews to assess people’s attitudes toward aging. These studies have found that Americans hold multiple and conflicting models of what it means to age. The most dominant association is that aging is a process of physical and mental decline that ends in dependency, an idea considered unpleasant by most people. In turn, Sweetland noted, people distance themselves from aging. They consider older people as different or separate from the rest of society and think of older people as “others.” Americans often defined older as 10 years older than themselves, no matter their age, and typically do not want to identify as aging. The tendency to “other” older people can dampen support for public policy issues that have to do with aging because members of the public do not want to associate themselves with the idea of being old, she said.
Reference Terms for Older Adults
The FrameWorks Institute ran an experiment to explore the public’s associations of common reference terms for older adults with assessed level of competence. Respondents consistently ranked words like “senior,” “elder,” and “senior citizen” as being the least competent (see Figure 4-4), and respondents associated these terms with frailness, feebleness, and not being able to use a cell phone. The terms “older person” and “older adult” were consistently associated with high competence and ideas like wisdom, savvy, and self-sufficiency. At the time of this research, Sweetland noted, the term recommended by many leading aging organizations was “older adult.” However, when respondents were asked how old an older adult is, the average answer was 54 years old, and some answers
ranged as low as 30 years old. By contrast, an older person was expected to be in his or her mid-60s, and senior citizens were expected to be around ages 68 and 69.
In order to successfully characterize people in later life, the FrameWorks Institute recommends using the terms “older person” or “older people” because these terms cue that an individual is over the age of 60 while also prompting associations of competence. Sweetland cautioned against using the term “older adult” when talking about individuals who are 65 years or older because the public pictures a much younger person. The words we use can have significant and non-obvious effects on people’s understanding, she said, and it is necessary to use the available evidence-based methodologies so that positive cues are used. Positive cues, such as the term “older people,” can move the public toward more positive, productive understandings of age and away from negative stereotypes.
Communicating About Ageism
One of the most important issues that advocates can raise right now, Sweetland said, is the topic of ageism. She noted that FrameWorks research has found that most Americans have not heard of this term before, making it all the more important that communicators take the opportunity to frame the issue conscientiously. It is therefore necessary, she said, to elevate the issue of ageism by talking about it more frequently and explicitly. “There are three legs to the stool of supporting the public’s understanding of ageism: naming and defining it; explaining the mechanism, the process by which it happens, by explaining implicit bias; and offering examples,” she said.
Sweetland recommended specific wording for a definition of ageism: “discrimination based on prejudices about age.” This definition, she noted, allows room for interpersonal, structural, and institutional forms of ageism. It is also necessary, she said, to explicitly state that ageism can have a negative impact, and to avoid examples of ageism that might be
dismissed as “not that bad.” She noted that the most frequent form of ageism is jokes about age. Nevertheless, when presenting ageism as a social problem, leading with the issue of ageist jokes can backfire by making individuals more likely to dismiss ageism as an issue of little importance.
Sweetland cautioned against comparing ageism to other forms of identity-based discrimination, because when FrameWorks researchers tested language that compared ageism to sexism or racism, it backfired. People reasoned that because they had heard of sexism and racism but had not heard of ageism, ageism could not be as prevalent or as harmful as the more familiar forms of discrimination. Connecting ageism to sexism and racism led people to downplay its possible significance. Thus, we need to define ageism on its own terms, without comparing it to other “isms,” Sweetland said.
Ageism should also not be labeled as a civil rights issue, she said. When ageism was labeled a civil rights issue, FrameWorks found that people’s focus quickly became narrowed to actions that would be handled through lawsuits. Instead, talking to people about implicit bias and learned expectations was more successful in lending a sense of importance and urgency to the issue of ageism.
Instead, Sweetland said, the most effective strategy for elevating the issue of ageism is stating that ageism has a negative impact and using carefully chosen examples. FrameWorks found that making ageism concrete by using the example of workplace discrimination is an effective way to build public understanding of how and why ageism is harmful to individuals and society. In particular, because the public is familiar with the idea of workplace discrimination in other areas, it is a useful starting point for illustrating the impact of ageism. Sweetland said that FrameWorks analyzed public reaction to a current policy proposal—namely, changing job applications so that applicants are not required to disclose the year that they graduated from high school or college—and found that this example prompted members of the public to broaden their thinking about structural stigma and stereotypes. She noted that examples of age discrimination in health care were also found to be effective in raising public awareness.
Finally, Sweetland said that it was important for communicators to help the public understand how age discrimination comes about, and that raising awareness of implicit bias can be an effective strategy. The explanation of implicit bias that FrameWorks recommends says that “Because we are all exposed to negative messages about older people, our brains automatically form judgments about people based on their age.” For example, she pointed out that older characters in movies are often portrayed as forgetful, grouchy, or frail. As these stereotypical characteristics become ingrained in our understanding of old age, it can affect our deci-
sions and assumptions without us realizing it. This explanation should be followed with examples of what can be done to disrupt implicit bias or prevent it from causing unfair treatment of older people, Sweetland said.
Sweetland closed with a quote:
Neither revolution nor reformation can ultimately change a society. Rather, you must tell a more powerful tale, one so persuasive that it sweeps away the old myths and becomes the preferred story, one so inclusive that it gathers all the bits of our past and present into a coherent whole, one that even shines some light into our future so that we can take the next step. (Illich, 2007)
She said, “the ideas that we have brought together today could be really helpful in shining a light on a different way of talking about aging, and I invite you to change society by telling an alternative story.”
A workshop participant reflected on the different types of discrimination that individuals might face (e.g., discrimination based on age, ability, race, income, or sexual orientation) and asked how discrimination against older persons or those with disabilities might vary based on other demographic factors and whether this should affect how we think about the ways in which discrimination occurs. Devine answered that we do need to think about those other factors because no one is a member of just one group and we all live with different experiences of intersectionality. Hatzenbuehler added that some believe having multiple stigmatized identities can cause additively worse outcomes. This is sometimes true, he said, but not for all outcomes: some research has shown a protective aspect to having multiple identities. Hatzenbuehler gave this example: “If you are African American and identify as gay, you might draw upon your experience growing up in a family that inculcated a positive sense of identity, and thus you can buffer the negative experiences or risk factors of discrimination based on sexual orientation.” The ways in which intersectionality play out are complicated, and while intersectionality does not always predict negative outcomes, the complexity makes it difficult to tailor interventions across different groups, he said. Sweetland added that keeping intersectionality in mind might offer the fields of aging and disability different ways to think about possible coalitions that could be very powerful. She encouraged the audience to think about how to con-
nect the movement against ageism and ableism to other movements as a way to begin creating change.
James C. Appleby of The Gerontological Society of America asked if the issue of workplace discrimination could be addressed by teaching people who screen resumes and job applicants how to flag their own biases and therefore avoid them. He asked Devine if there is a way to use her interventional work to get at workplace discrimination. Devine replied that it was absolutely possible and that many organizations around the United States are already interested in doing so. She noted that in her STEMM study, she did not actually know the mechanism causing more women to be hired in the intervention group. She asked: Were the people in charge of hiring doing a more careful analysis of the resumes they received? Did they make a concerted effort to bring in more women and then discover that these women were worthy of being hired? Is it because the department’s climate changed and thus became more effective in recruiting women? She believes that more detailed information on these processes could lead to the development of even more effective tools to interrupt bias.
Julie Bynum of Dartmouth University asked how stereotypes and bias around aging and disability might affect physicians’ heuristics and decision-making processes and inquired about interventions to address implicit bias in decision making. Devine answered that the key issue is to recognize when heuristics would lead a physician to provide worse instead of better care. She suggested specific things that physicians could do to ward off unintentional bias, including the use of actionable checklists and ways to interact with patients and their families to ensure that the physician is getting necessary information for decision making. This kind of intervention helps people realize how often they are acting on implicit biases and helps them to slow down and have more thoughtful decision-making processes, Devine said. “If you can name it, you can tame it,” she said, in regard to identifying bias that might lead someone astray. She once again emphasized that habit breaking is something that one has to do in all domains of life.
Margaret Campbell of Campbell & Associates Consulting brought up the tendency to reify social categories through stereotypes and asked the panelists to comment on their thoughts about getting beyond stereotypes to inclusion. Sweetland responded by providing two framing tips: (1) moving from using they/them language to us/we labels is extremely important in mitigating stereotypes; and (2) she has found it effective to openly communicate with an insistence and reminder that a just society includes and welcomes the participation of all of its members, making sure to note that right now we are marginalizing, excluding, and minimizing the contributions of older persons. She said that simply building this
kind of description into communication makes people want to be more inclusive. Devine added that becoming aware of the way in which stereotypes actually constrain the opportunities for older people, whether in the job context or in daily interactions, can be a revelation for some people, which can then lead them to make efforts to become more inclusive and create a welcoming environment. Devine described the University of Wisconsin–Madison Challenge6 aimed at promoting inclusion and overcoming bias on campus. She said she has found that people who undergo stereotype intervention training are more likely to join the challenge and take active steps to mitigate stereotyping.
Kathy Greenlee of the Center for Practical Bioethics emphasized another aspect of framing aging that often goes unnoticed: society often describes aging as an individual experience, and many messages tell older people that they are responsible for their own future. This is true, she said, but it prevents us from getting to a collective sense of “we.” She said that there is a much better sense of we in the disability field and that the aging field needs to create more cohesion among stakeholders and older people in order to mobilize people en masse around issues of aging.
6 For more information, see https://campusclimate.wisc.edu/join-the-uw-challenge (accessed December 19, 2017).