Rogerian Essay

Review

www.thelancet.com Vol 393 February 9, 2019 579

Working toward gender diversity and inclusion in medicine: myths and solutions Sonia K Kang, Sarah Kaplan

Women’s representation in science and medicine has slowly increased over the past few decades. However, this rise in numbers of women, or gender diversity, has not been matched by a rise in gender inclusion. Despite increasing representation, women still encounter bias and discrimination when compared with men in these fields across a variety of outcomes, including treatment at school and work, hiring, compensation, evaluation, and promotion. Individual and systemic biases create unwelcome environments for women, particularly for those who additionally identify with other traditionally devalued groups (eg, women of colour). This Review draws on several decades of research in the field of management and its cognate disciplines to identify five myths that continue to perpetuate gender bias and five strategies for improving not only the number of women in medicine, but also their lived experiences, capacity to aspire, and opportunity to succeed. We argue for a move away from a singular focus on interventions aimed at targeting individual attitudes and behaviour to more comprehensive interventions that address structural and systemic changes.

Introduction The year 2017 marked the first time in history when the number of women enrolling in US medical schools exceeded the number of men.1 When this historic cohort of female medical students enters the workforce, what kind of work environment will they encounter? Despite the record numbers of women entering fields across science, technology, engineering, mathematics (STEM), and medicine,2 women continue to experience dis­ advantage, discrim ination, and gender­based violence in their home and work lives,3–9 a reality that is too often amplified for women of colour, of low socioeconomic status and social class, and of advanced age, and women who do not identify as heterosexual, are disabled, or belong to other traditionally devalued groups.10–16 In medicine, these inequities manifest for women as everyday experiences of sexism, which includes exposure to sexist jokes in class; sexual harassment by clinicians, faculty, or patients; weaker reference letters than men for medical school faculty applications; lower income than men; channelling into lower paid areas of medicine such as family practice; and a decreased likelihood of being addressed by one’s professional title than men.17–22 While we may be making progress on the numbers—an increase in gender diversity—true progress on improving women’s sense of belonging and inclusion is critically lagging.

The disconnect between diversity and inclusion is not unique to medicine. This inability to see the problem as beyond merely the number of women in the field occurs in a wide variety of academic and professional domains.23–26 In this Review, we offer insights from decades of research on diversity and inclusion in the fields of manage ment and its cognate disciplines, inc­ luding psychology, sociology, and economics, to propose suggestions for improving not only the number of women in medicine, but also their lived experiences, capacity to aspire, and opportunity to succeed. We acknowledge that most of this work has been done in Europe and North America and therefore might not be

generalisable to all contexts. Moreover, any effective intervention would have to be tailored not only to country cultures and laws, but also to the specific organisations and departments in which these interventions are being made. Thus, the solutions proposed here should not be viewed as general fixed principles, but rather as a starting point for making more localised change. We start by debunking five myths that are commonly encountered when examining diversity and inclusion practices, and conclude by offering five research­supported solutions to bring about equity by design, an approach that we argue is particularly well suited to the medical field.

Five myths about diversity and inclusion Diversity and inclusion policies and practice are becoming nearly ubiquitous in organisational settings. Finding an organisation or institution without a written statement outlining their commitment to diversity is now rare, and billions of dollars each year are used in the efforts to increase the representation of women and minorities.27,28 To create lasting change and to prevent the current focus on diversity and inclusion from becoming another ineffectual trend in management, it is important to ensure that efforts are evidence­based and do not rely on common myths that might instead perpetuate the problems they are trying to solve. The five myths uncovered in this Review are not mutually exclusive—often being inter­ twined or potentially conflicting—but represent the most common (albeit inaccurate) assump tions people make about achieving diversity and inclusion.

Myth 1: other people are biased, not me The first myth that should be debunked is the idea that bias is a problem unique to only a few individuals: namely the racists, sexists, and bigots among us.29 However, research on the human brain and how it makes sense of the world suggests not only that all of us are biased, but that we must be biased to survive.30,31 Cognitive biases and heuristics are shortcuts that allow us to interact meaningfully with people, objects, and tasks

Lancet 2019; 393: 579–86

Institute for Management and Innovation, University of Toronto Mississauga, Mississauga, ON, Canada (S K Kang PhD); Institute for Gender and the Economy, Rotman School of Management, University of Toronto, Toronto, ON, Canada (S K Kang, Prof S Kaplan PhD); and Mack Institute for Innovation Management, Wharton School, University of Pennsylvania, Philadelphia, PA, USA (Prof S Kaplan)

Correspondence to: Dr Sonia K Kang, Rotman School of Management, University of Toronto, ON M5S 3E6 Canada [email protected]

For more on the #LancetWomen initiative see https://www. thelancet.com/lancet-women

Review

580 www.thelancet.com Vol 393 February 9, 2019

without having to exhaust our insufficient attentional resources to decipher every sensory signal.32,33 Whenever you encounter a person, for example, your brain rapidly engages in a series of calculations to interpret that person’s relevance to you by placing them within a social category.32 The first automatic calculations regard age, race, and gender.34–36 Because of this perceptual primacy, gender has come to frame the way we see the world; it is an implicit or unconscious bias that serves as a foundation upon which stereotypes, expectations, and norms have been created.37 Social categorisation is an inevitable part of our perceptual experience, such that the stereotypes we hold about different social groups will alter our perceptions of, and reactions to, individual group members.33 Further, when it comes to devaluing women’s contributions in masculinised settings, women can be just as biased as men, meaning that people of all gender identities can perpetuate gender bias in organisations.5,38 Rejecting the idea that only some people are biased is a crucial first step to personally engaging with the problem of discrimination so as to bring about change.

Myth 2: the key to controlling bias is controlling how people think Most of the efforts made to minimise bias in organisations has focused on controlling or eradicating the biases that exist in our minds. Implicit bias training is an example of such efforts. Testing for implicit bias via the Implicit Association Test (IAT)39 has become commonplace, and has risen in popularity along with implicit bias or diversity training. Despite the millions of dollars spent on administering the IAT and training people to act without bias, the evidence that this kind of training actually changes organisational outcomes is scarce.40,41 More commonly, diversity training (especially when done alone and not in combination with other organi sational interventions) has produced a host of unintended consequences.42 It has been shown to be associated with reduced diversity,43 worsened behaviour toward minority co­workers,44 and the creation of the illusion of fairness such that those who claim to have experienced discrimination are less likely to be believed.27 Instructing people to avoid the use of stereotypes can paradoxically lead to increased activation of those stereotypes,45 and attempts to increase the awareness of stereotype prevalence can inadvertently normalise stereotyping and discrimination (such as, if everyone uses stereotypes it must be okay).46 Eradicating these innate human biases is difficult and likely to be impossible. Although educating people about these biases and providing education on how to recognise them is an important first step, we must go further to create systems and environ ments in which bias and stereotyping are either less likely to become initiated, or are prevented from resulting in discrimination even when they are active.

Myth 3: under-representation of women is a pipeline problem The representation of women across a variety of STEM fields and within medicine has been slowly increasing, albeit at different rates within these fields and across nations.47,48 If we look to the research on child development and psychology, girls perform to an equal or better standard than boys in STEM topics, and report high interest in pursuing careers in STEM.49–52 Therefore, the pipeline of female trainees and candidates itself is healthy. The real problem is brought about by the pressures that push women out of the pipeline.18,20,21,53–58 Research shows that discrimination exists against women at each stage of professional life, from recruitment and selection, to recommendation, evaluation, promotion, training, and compensation.3,59,60 These effects are often exacerbated for women of colour or for those who possess other devalued intersectional identities.22,61–66 Women are conferred less respect and status, experience greater workplace hostility and harassment, are disproportionately punished for errors, and experience higher amounts of invisible and uncompensated labour, particularly in terms of emotional labour, than men.54,67–74 It is often argued that women choose to opt out of certain careers or opt in when lower status roles are available due to motherhood. However, research suggests that the effects of this so­called motherhood penalty are structured by discrim inatory dynamics.75,76 Further, if motherhood fully explains women’s under­representation in STEM and medicine, then we would not also observe under­representation of men of colour, but we do.77 Therefore, it is not the case that women are entering the pipeline in too few numbers, but rather that a confluence of factors is pushing them out.

Myth 4: promoting diversity contravenes meritocracy One of the most commonly cited explanations that people provide for rejecting diversity initiatives is that their organisations are meritocratic.78 The arguments are that if women were equally qualified, they would be hired and promoted, and that any diversity initiatives aimed at righting the imbalance would compromise quality. However, an abundance of research evidence shows our so­called meritocracies are not so meritocratic.79,80 Studies that control for underlying quality show that a signal of female gender by itself leads to devaluation: for equal curriculum vitae in which only the name is different, Brian is more likely than Karen to be seen as hireable; with equal business cases for a startup company in which only the video narration voice differs, the male narration is deemed twice as investable as the female narration; female postdoctoral applicants have to be 2·5 times more productive than the average male applicant to be hired; and computers named James are valued nearly 25% more than computers named Julie.56,58,81–84 So, if anything, underlying biases appear to be causing the current meritocratic systems to bypass many highly capable women and members of other minority groups. We are

Review

www.thelancet.com Vol 393 February 9, 2019 581

drawing heterosexual, white men from much further down the distribution of talent than we are for other social categories.

Myth 5: we have to fix the women Related to the gender bias in organisations, most programmes attempting to address gender inequality focus on so­called fixing the women, by teaching them such skills as how to lean in, negotiate better, stand up straight, adopt powerful postures, talk more in meetings, and be more assertive, to name a few.85–91 Although not universally the case, many of these solutions are themselves highly biased, in that they train women to act more like men because the actions of men are more valued and perceived as the correct way to succeed. What is neglected in this approach is the backlash women often experience when engaging in these behaviours. Competent women with agency have been shown to experience backlash for violating expectations of warmth and so­called feminine niceness.92–95 Research also shows that women are punished more severely than men for mistakes or failures, and that these negative effects can negatively affect perceptions of other female co­workers.72,96,97 Thus, attempts to fix the women will continue to be counterproductive within a system that is rigged against them.

Five solutions for achieving gender equality in medicine We turn next to a series of suggestions based on management research for improving the experience of women in medicine. This research shows that inter­ vening at the level of the individual is difficult, perhaps even impossible, and that attempts to alter the behaviour of individuals have not proven effective on their own. The introduction and application of stereotypes and other biases are most common under stress (eg, when time is restricted, when information processing demands are high, or when we are under pressure to perform). Accordingly, perhaps no better environment for un­ intentionally enabling bias exists than the medical workplace, where demands are high, resources can be low, and the pressure to perform the right action quickly can be overwhelming. Under these conditions, individual change must be embedded within a structure that is designed to enable progress. Management research about the medical profession has shown that practitioners can easily overlook or misin terpret cues in crisis situations, supporting the assertion that educating about and attempting to control bias are not enough.98 We also know from progress in evidence­based medicine that structural solutions such as behavioural guidelines might be an effective means for physicians and other health­care practitioners to over come habits and biases.99,100 There­ fore, we argue that medicine is particularly well suited to interventions that target organisational change by designing for equality, and we outline five potential

solutions in this context. These solutions are presented as a starting point for the innovation of refined and targeted solutions that consider departmental, organisational, regional, and national contextual needs and constraints. We should also recognise that we cannot achieve equality inside organi sations without also achieving equality in people’s lives outside of work. Structural inequalities in society should also change. The interventions that we will outline therefore focus on only one aspect of achieving workplace gender equality in medicine, but it is the aspect over which most people in the medical profession have more control.

Solution 1: treat gender equality as an innovation challenge The general approach to working toward gender equality, and diversity more broadly, has been largely rooted in attitudes and values. Justifications for gender equality are often discussed in terms of the so­called business case, focusing on how equity, diversity, and inclusion are economically productive, rather than focusing on them as the right thing to do.101 Instead of initiating action toward finding a solution, such discussions on gender equality become stalled when trying to define the nature of the problem and in determining whether or not it is worthy of solving. Further, this type of framing can ultimately cause more harm than good.101 To make progress in achieving gender equality, we must declare the discussion on whether and why we should pursue equality to be over. From this point, within contexts where this is possible (ie, within some contexts even the recognition of gender equality as a basic human right is still an ongoing challenge), we can switch our focus onto experimentation and innovation.102 As with any organisational initiative, gender equality should be approached with an open and scientific attitude, and the willingness to experiment and measure outcomes. Because the challenge of achieving equality is complex and multifaceted, openness to failure and the willingness to change tactics is a must, as is transparency via measurement and reporting, so that momentum and accountability for change remain high.

The most promising solutions are probably behavioural and systemic changes that focus on creating a climate for change, an approach widely supported by the so­called nudge theory (with the idea of identifying easy to implement and economical ways to change people’s behaviours by structuring their choices),103 rather than those focusing only on changing individual attitudes or values. Gender­inclusive workplace cultures are those that create a positive social climate for people of all gender identities, and can be cultivated through such practices as inc reasing the representation of women and gender non­binary people in leadership, by use of gender inclusive photos and pronouns in organisational communications, and adopting anonymous evaluation practices that minimise the potential for bias by

Review

582 www.thelancet.com Vol 393 February 9, 2019

eliminating gender cues such as names and pro­ nouns.43,104–107 That said, no quick­fix solution is available to offer, and actual change will only follow from the repeated application of commitment, courage, and many iterations of innovative experimentation. Further, just as we should think about solutions for gender diversity innovatively, we should also update our conceptualisation of gender itself to expand beyond the traditional male and female binary, so as to encompass the range of identities that represent gender diversity.

Solution 2: change institutional norms Norms are the conventional patterns of behaviour that are considered acceptable by a social group.108–110 People of all gender identities are under pressure to conform to gender norms (eg, women are expected to be kind and nurturing; men are expected to be competent and strong).111–113 These norms have a powerful influence on our behaviour.114,115 They result in women being socialised into more communal medical specialties (eg, family medicine) and men being socialised into more agentic specialties (eg, surgery).21 Over time, these norms have strong effects on other measurable outcomes beyond behaviour. As family medicine has become more femin ised over time, for example, the pay gap between family medicine and other specialties has widened consider ably.19,116 Fortunately, because we are a fundamentally social species, changing perceptions of norms also changes behaviour.117 Theories on the social influence of norms predict that if everyone else in an organisation appears to value diversity, we are more likely to act like we value diversity ourselves. Conversely, if expression of prejudicial attitudes or engagement in discriminatory behaviour (as observed in the so­called iron man surgery culture118) is considered normative, these practices will become embedded within a social environment. The most important source of such normative change is that of a group’s leaders.119,120 The behaviour of those at the head of a group have a powerful influence on the people further down the group, and therefore the communication and behaviour of hospital administration and senior staff, for example, must show a commitment to diversity for others to follow suit.

Solution 3: create a culture in which people feel personally responsible for change One of the reasons that diversity training programmes can be so spectacularly unsuccessful is that they challenge people’s sense of autonomy, self­determination, and control.42,121 Just as humans are inherently prone to bias, so too do we have an inherent drive toward autonomy, which can lead us to resist initiatives that we feel are forced upon us.122,123 People react negatively to perceived coercion, and overbearing diversity program mes can therefore go wrong and actually make organisations less inclusive. The three most common diversity programmes of the past 30 years, mandatory diversity training, mandatory testing for job applicants, and grievance systems, are associated

with decreases in the representation of white women, as well as black, Hispanic, and Asian men and women.43 Better results are seen with diversity pro grammes that capitalise on people’s need for autonomy, increase contact between diverse groups, encourage personal engagement, and include all members of the organisation rather than only those who are part of the group targeted for intervention. Examples of successful diversity pro­ grammes are men toring programmes, which effectively increase represen tation among minority women in particular, and the establishment of diversity task forces.43 Even more effective are sponsorship pro grammes in which sponsors become personally invested in their protégé’s career success, take risks to champion them for recognition and advancement, and actively embed them in powerful networks.124,125 Combined targeted recruitment and mentoring programmes, in which sponsors are given personal responsibility for recruiting and fostering the success of under­represented minorities can also be effective. These types of programmes are promising not only because they support individual autonomy and engagement, but also because they circumvent the hesitation to participate because an individual believes he or she is not biased. Attempts to control specific attitudes and behaviours can go wrong if individuals feel that their autonomy is threatened, but meaningfully engaging them in organisational change can help to circumvent this type of backlash.121

Solution 4: implement behavioural guidelines and action plans People often encounter difficulty translating their goals into action. This issue can be remedied through a type of planning known as implementation intentions, which links anticipated acute situations to goal­directed responses (eg, “whenever situation x happens, I will initiate the goal­directed response y”).126,127 These kinds of systems are already very common in medical workplaces (eg, code systems clearly link specific situations to a prescribed set of responses). To advance toward the goal of gender equality, one suggestion is to put more emphasis on behavioural rather than attitudinal guidelines for promoting diversity and inclusion. For example, consider what might happen if a hospital decides to combat gender bias during the search for new attending physicians. A typical first step would be clearly outlining the goal (eg, to eliminate gender bias within search committees looking for attending physicians) and the rationale (eg, eliminating gender bias will benefit our team, patients, organisation, and other various stake holders) This step is where most plans for improving diversity end. People are left with an abstract set of values and goals, but no distinct action plan for achieving them and no indication of how progress will be measured and success identified, which research over the decades has shown is an essential part of effective goal­setting.128 Worse still, in some cases abstract or attitudinal diversity goals or statements have been

Review

www.thelancet.com Vol 393 February 9, 2019 583

unsuccessful and end up doing more harm than good.27,28,42,129–131

To move from abstract plans to actions and avoid unintended consequences, organisations must clearly lay out the specific steps that will be taken to enact their values and goals, and specify the indicators that will be used to measure success, while also taking into con sideration the many barriers that stand in the way of individual behavioural change.132 In the attending physician example, some of these guidelines might include ensuring that at least a third of hiring committee members are women, with success being identified as a steady state of 50:50 men and women on committees within 3 years; ensuring that only standardised questions and structured interviews are used, and using a statistically significant increase in the number of women hired over a 3 year period as an indicator of success. Without specific behavioural guide lines to inform prac­ tice, diversity and inclusion are often unable to advance beyond attitudes and goals. Examples of resources to equip managers with the tools to interrupt bias are the toolkits available via Bias Interrupters, an initiative of The Centre for Worklife Law at the University of California Hastings College of the Law (San Francisco, CA, USA), and the Bias Busting Strategies worksheets created by the Engendering Success in STEM consor­ tium. These tools are available for individuals and organisations, list clear behaviours and responses that can be initiated when bias is encountered, and provide a starting point from which to build specific action plans for a given environmental context and set of goals.

Solution 5: create organisational accountability for change Just as individual change must be embedded within supportive organisational structures, so too must organisational efforts be embedded within larger systems that support and monitor progress toward diversity and inclusion goals.43,121 Methods for holding individuals, teams, and the organisation as a whole accountable for change can help by measuring and keeping progress on track toward essential milestones, and by signalling the importance of the initiative.103 The common maxim what gets measured gets done applies just as well to diversity initiatives as it does to any other type of initiative an organisation might wish to set up. Without ongoing data collection and trans parency, whether time and resources are being expended effectively or just wasted is impossible to tell. However, to avoid backlash associated with threats to autonomy, organisational accountability must be des­ igned empathetically and with room for failure.3,29,121 Failure is part of the experimental process, and learning from mistakes allows us to refine, redesign, and retest. Organisational initiatives embedded within account­ ability frameworks such as affirmative action plans and work processes adhered to and promoted by diversity committees and task forces and diversity managers or

departments have shown great promise in increasing and sustaining diversity.121 Again, the most promising approach is comprehensive, in which individual, struc­ tural, and organisational initiatives are combined in the push for progress.

Conclusions The available evidence is clear: decades of policies and billions of dollars aimed at changing individuals have not been successful in bringing about gender equality. We have made progress, albeit uneven across the world and specialties, in the number of women entering and working in medicine, but true progress on inclusion remains an elusive goal. By understanding more about how bias works and dismissing the myths that have held us back for so long, we can turn our attention and resources toward structural and systemic interventions that have more promise for success. Contributors SKan wrote the first draft. SKap collaborated on outlining the ideas and on revisions of the preliminary draft.

Declaration of interests SKap reports advising the Department of Medicine, University of Toronto, Toronto, ON, Canada prior to writing this review. SKan declares no competing interests.

References 1 Association of American Medical Colleges. More women than men

enrolled in U.S. medical schools in 2017. Dec 18, 2017. https://news. aamc.org/press­releases/article/applicant­enrollment­2017/ (accessed Sept 14, 2018).

2 Bowley R. A snapshot of progress among women in the workforce. March 6, 2018. https://blog.linkedin.com/2018/march/6/a­snap shot­of­progress­among­women­in­the­workforce (accessed Sept 14, 2018).

3 Bohnet I. What works: gender equality by design. Cambridge, MA: The Belknap Press of Harvard University Press, 2016.

4 Eagly AH, Carli LL. Through the labyrinth: the truth about how women become leaders. Boston, MA: Harvard Business Review Press, 2007.

5 Heilman ME. Gender stereotypes and workplace bias. Res Organ Behav 2012; 32: 113–35.

6 Parker K, Funk C. Gender discrimination comes in many forms for today’s working women. Dec 14, 2017. http://www.pewresearch.org/ fact­tank/2017/12/14/gender­discrimination­comes­in­many­forms­ for­todays­working­women/ (accessed Sept 14, 2018).

7 WHO. Violence against women. Nov 29, 2017. http://www.who.int/ news­room/fact­sheets/detail/violence­against­women (accessed Sept 14, 2018).

8 World Bank. Women, business and the law 2018. Washington, DC: World Bank Group, 2018.

9 UN Women. Discrimination against women persists around the globe hampering development. July 5, 2012. http://www.unwomen. org/en/news/stories/2012/7/discrimination­against­women­persists­ around­the­globe­hampering­development (accessed Nov 16, 2018).

10 Dasgupta N, Stout JG. Girls and women in science, technology, engineering, and mathematics: STEMing the tide and broadening participation in STEM careers. Policy Insights Behav Brain Sci 2014; 1: 21–29.

11 Duncan C, Loretto W. Never the right age? Gender and age­based discrimination in employment. Gend Work Organ 2004; 11: 95–115.

12 Hunter ML. “If you’re light you’re alright”: light skin color as social capital for women of color. Gend Soc 2002; 16: 175–93.

13 Ortiz SY, Roscigno VJ. Discrimination, women, and work: processes and variations by race and class. Sociol Q 2009; 50: 336–59.

14 Sanchez­Hucles JV, Davis DD. Women and women of color in leadership: complexity, identity, and intersectionality. Am Psychol 2010; 65: 171–81.

For more on Bias Interrupters see https://biasinterrupters.org/

For more on the Bias Busting Strategies worksheets see http://successinstem.ca/ resources/

Review

584 www.thelancet.com Vol 393 February 9, 2019

15 Shields SA. Gender: an intersectionality perspective. Sex Roles 2008; 59: 301–11.

16 Williams JC, Multhaup M, Mihaylo S. Why companies should add class to their diversity discussions. Sept 5, 2018. https://hbr.org/2018/ 09/why­companies­should­add­class­to­their­diversity­discussions (accessed Sept 18, 2018).

17 Beagan B. Micro inequities and everyday inequalities: ‘race,’ gender, sexuality and class in medical school. Can J Sociol 2001; 26: 583–610.

18 Files JA, Mayer AP, Ko MG, et al. Speaker introductions at internal medicine grand rounds: forms of address reveal gender bias. J Womens Health (Larchmt) 2017; 26: 413–19.

19 Glauser W. Why are women still earning less than men in medicine? CMAJ 2018; 190: E664–65.

20 Sanfey H, Crandall M, Shaughnessy E, et al. Strategies for identifying and closing the gender salary gap in surgery. J Am Coll Surg 2017; 225: 333–38.

21 Trix F, Psenka C. Exploring the color of glass: letters of recommendation for female and male medical faculty. Discourse Soc 2003; 14: 191–220.

22 Williams J, Phillips KW, Hall EV. Double jeopardy? Gender bias against women of color in science. Jan, 2014. https://worklifelaw. org/publications/Double­Jeopardy­Report_v6_full_web­sm.pdf (accessed Nov 1, 2018).

23 Nishii LH. The benefits of climate for inclusion for gender­diverse groups. Acad Manage J 2012; 56: 1754–74.

24 Roberson QM. Disentangling the meanings of diversity and inclusion in organizations. Group Organ Manag 2006; 31: 212–36.

25 Shore LM, Randel AE, Chung BG, Dean MA, Holcombe Ehrhart K, Singh G. Inclusion and diversity in work groups: a review and model for future research. J Manage 2011; 37: 1262–89.

26 Tienda M. Diversity ≠ inclusion: promoting integration in higher education. Educ Res 2013; 42: 467–75.

27 Kaiser CR, Major B, Jurcevic I, Dover TL, Brady LM, Shapiro JR. Presumed fair: ironic effects of organizational diversity structures. J Pers Soc Psychol 2013; 104: 504–19.

28 Kang SK, DeCelles KA, Tilcsik A, Jun S. Whitened résumés: race and self­presentation in the labor market. Adm Sci Q 2016; 61: 469–502.

29 Chugh D. The person you mean to be: confronting bias to build a better workplace and world, 1st edn. New York, NY: Harper Business, 2018.

30 Macrae CN, Milne AB, Bodenhausen GV. Stereotypes as energy­saving devices: a peek inside the cognitive toolbox. J Pers Soc Psychol 1994; 66: 37–47.

31 Macrae CN, Bodenhausen GV. Social cognition: thinking categorically about others. Annu Rev Psychol 2000; 51: 93–120.

32 Fiske ST, Lin M, Neuberg SL. The continuum model: ten years later. In: Chaiken S, Trope Y, eds. Dual­process theories in social psychology. New York, NY: Guilford Press, 1999: 231–54.

33 Macrae CN, Bodenhausen GV. Social cognition: categorical person perception. Br J Psychol 2001; 92: 239–55.

34 Brewer MB, Lui LN. The primacy of age and sex in the structure of person categories. Soc Cogn 1989; 7: 262–74.

35 Ridgeway CL. Framed before we know it: how gender shapes social relations. Gend Soc 2009; 23: 145–60.

36 Schneider DJ. The psychology of stereotyping. New York, NY: Guilford Press, 2004.

37 Greenwald AG, Banaji MR. Implicit social cognition: attitudes, self­esteem, and stereotypes. Psychol Rev 1995; 102: 4–27.

38 Ellemers N, Rink F, Derks B, Ryan MK. Women in high places: when and why promoting women into top positions can harm them individually or as a group (and how to prevent this). Res Organ Behav 2012; 32: 163–87.

39 Greenwald AG, McGhee DE, Schwartz JL. Measuring individual differences in implicit cognition: the implicit association test. J Pers Soc Psychol 1998; 74: 1464–80.

40 Oswald FL, Mitchell G, Blanton H, Jaccard J, Tetlock PE. Predicting ethnic and racial discrimination: a meta­analysis of IAT criterion studies. J Pers Soc Psychol 2013; 105: 171–92.

41 Singal J. Psychology’s favorite tool for measuring racism isn’t up to the job. Jan 11, 2017. https://www.thecut.com/2017/01/psychologys­ racism­measuring­tool­isnt­up­to­the­job.html (accessed Sept 22, 2018).

42 Dobbin F, Kalev A. Why diversity programs fail. July 1, 2016. https://hbr.org/2016/07/why­diversity­programs­fail (accessed Sept 22, 2018).

43 Kalev A, Dobbin F, Kelly E. Best practices or best guesses? Assessing the efficacy of corporate affirmative action and diversity policies. Am Sociol Rev 2006; 71: 589–617.

44 Sanchez JI, Medkik N. The effects of diversity awareness training on differential treatment. Group Organ Manag 2004; 29: 517–36.

45 Macrae CN, Bodenhausen GV, Milne AB, Jetten J. Out of mind but back in sight: stereotypes on the rebound. J Pers Soc Psychol 1994; 67: 808–17.

46 Duguid MM, Thomas­Hunt MC. Condoning stereotyping? How awareness of stereotyping prevalence impacts expression of stereotypes. J Appl Psychol 2015; 100: 343–59.

47 Hill C, Corbett C, St. Rose A. Why so few? Women in science, technology, engineering, and mathematics. 2010. https://eric.ed. gov/?id=ED509653 (accessed Sept 22, 2018).

48 Miller DI, Eagly AH, Linn MC. Women’s representation in science predicts national gender­science stereotypes: evidence from 66 nations. J Educ Psychol 2015; 107: 631–44.

49 Halpern DF, Benbow CP, Geary DC, Gur RC, Hyde JS, Gernsbacher MA. The science of sex differences in science and mathematics. Psychol Sci Public Interest 2007; 8: 1–51.

50 Hyde JS, Lindberg SM, Linn MC, Ellis AB, Williams CC. Gender similarities characterize math performance. Science 2008; 321: 494–95.

51 Pomerantz EM, Altermatt ER, Saxon JL. Making the grade but feeling distressed: gender differences in academic performance and internal distress. J Educ Psychol 2002; 94: 396–404.

52 Spelke ES. Sex differences in intrinsic aptitude for mathematics and science?: a critical review. Am Psychol 2005; 60: 950–58.

53 Carnes M, Bartels CM, Kaatz A, Kolehmainen C. Why is John more likely to become department chair than Jennifer? Trans Am Clin Climatol Assoc 2015; 126: 197–214.

54 Hall W, Schmader T, Aday A, Croft E. Decoding the dynamics of social identity threat in the workplace: a within­person analysis of women’s and men’s interactions in STEM. Soc Psychol Personal Sci 2018; published online May 2. DOI:10.1177/1948550618772582.

55 Hall W, Schmader T, Aday A, Inness M, Croft E. Climate control: the relationship between social identity threat and cues to an identity­safe culture. J Pers Soc Psychol 2018; 115: 446–67.

56 Moss­Racusin CA, Dovidio JF, Brescoll VL, Graham MJ, Handelsman J. Science faculty’s subtle gender biases favor male students. Proc Natl Acad Sci USA 2012; 109: 16474–79.

57 Shapiro JR, Williams AM. The role of stereotype threats in undermining girls’ and women’s performance and interest in STEM fields. Sex Roles 2012; 66: 175–83.

58 Steinpreis RE, Anders KA, Ritzke D. The impact of gender on the review of the curricula vitae of job applicants and tenure candidates: a national empirical study. Sex Roles 1999; 41: 509–28.

59 Botelho TL, Abraham M. Pursuing quality: how search costs and uncertainty magnify gender­based double standards in a multistage evaluation process. Adm Sci Q 2017; 62: 698–730.

60 Heilman ME, Eagly AH. Gender stereotypes are alive, well, and busy producing workplace discrimination. Ind Organ Psychol 2008; 1: 393–98.

61 Berdahl JL, Moore C. Workplace harassment: double jeopardy for minority women. J Appl Psychol 2006; 91: 426–36.

62 Clancy KBH, Lee KMN, Rodgers EM, Richey C. Double jeopardy in astronomy and planetary science: women of color face greater risks of gendered and racial harassment. J Geophys Res Planets 2017; 122: 1610–23.

63 Espinosa L. Pipelines and pathways: women of color in undergraduate STEM majors and the college experiences that contribute to persistence. Harv Educ Rev 2011; 81: 209–41.

64 Krieger N, Waterman PD, Hartman C, et al. Social hazards on the job: workplace abuse, sexual harassment, and racial discrimination— a study of black, latino, and white low­income women and men workers in the United States. Int J Health Serv 2006; 36: 51–85.

65 Randolph DS. The meaning of workplace discrimination for women with disabilities. Work 2005; 24: 369–80.

66 Sayer G, Granleese J. Gendered ageism and “lookism”: a triple jeopardy for female academics. Women Manage Rev 2006; 21: 500–17.

Review

www.thelancet.com Vol 393 February 9, 2019 585

67 Babcock L, Recalde MP, Vesterlund L, Weingart L. Gender differences in accepting and receiving requests for tasks with low promotability. Am Econ Rev 2017; 107: 714–47.

68 Fitzgerald LF. Sexual harassment. Violence against women in the workplace. Am Psychol 1993; 48: 1070–76.

69 Guy ME, Newman MA. Women’s jobs, men’s jobs: sex segregation and emotional labor. Public Adm Rev 2004; 64: 289–98.

70 Mitchell SM, Hesli VL. Women don’t ask? Women don’t say no? Bargaining and service in the political science profession. PS Polit Sci Polit 2013; 46: 355–69.

71 Morris JA, Feldman DC. The dimensions, antecedents, and consequences of emotional labor. Acad Manage Rev 1996; 21: 986–1010.

72 Sarsons H. Interpreting signals in the labor market: evidence from medical referrals. Nov 28, 2017. https://scholar.harvard.edu/files/ sarsons/files/sarsons_jmp.pdf (accessed Nov 1, 2018).

73 Thomas­Hunt MC, Phillips KW. When what you know is not enough: expertise and gender dynamics in task groups. Pers Soc Psychol Bull 2004; 30: 1585–98.

74 Winslow S. Gender inequality and time allocations among academic faculty. Gend Soc 2010; 24: 769–93.

75 Benard S, Correll SJ. Normative discrimination and the motherhood penalty. Gend Soc 2010; 24: 616–46.

76 Correll SJ, Benard S, Paik I. Getting a job: is there a motherhood penalty? AJS 2007; 112: 1297–339.

77 Griffith AL. Persistence of women and minorities in STEM field majors: is it the school that matters? Econ Educ Rev 2010; 29: 911–22.

78 Thomas KM, Mack DA, Montagliani A. The arguments against diversity: are they valid? In: Stockdale MS, Crosby FJ, eds. The psychology and management of workplace diversity. Malden, MA: Blackwell Publishing, 2004: 31–51.

79 Castilla EJ. Gender, race, and meritocracy in organizational careers. AJS 2008; 113: 1479–526.

80 Castilla EJ, Benard S. The paradox of meritocracy in organizations. Adm Sci Q 2010; 55: 543–676.

81 Brooks AW, Huang L, Kearney SW, Murray FE. Investors prefer entrepreneurial ventures pitched by attractive men. Proc Natl Acad Sci USA 2014; 111: 4427–31.

82 Posard MN. Status processes in human­computer interactions: does gender matter? Comput Human Behav 2014; 37: 189–95.

83 Wennerás C, Wold A. Nepotism and sexism in peer­review. Nature 1997; 387: 341–43.

84 Wennerás C, Wold A. Nepotism and sexism in peer­review. In: Wyers M, Barbercheck M, Giesman D, Öztürk HÖ, Wayne M, eds. Women, Science, and Technology, 2nd edn. New York, NY: Routledge, 2010: 50–56.

85 Cuddy AJ, Schultz, SJ, Fosse NE. P­curving a more comprehensive body of research on postural feedback reveals clear evidential value for power­posing effects: reply to Simmons and Simonsohn (2017). Psychol Sci 2018; 29: 656–66.

86 Rennung M, Blum J, Göritz AS. To strike a pose: no stereotype backlash for power posing women. Front Psychol 2016; 7: 1463.

87 Babcock L, Laschever S. Women don’t ask: negotiation and the gender divide. Princeton, NJ: Princeton University Press, 2003.

88 Bowles HR, Babcock L. How can women escape the compensation negotiation dilemma? Relational accounts are one answer. Psychol Women Q 2013; 37: 80–96.

89 Brescoll VL. Who takes the floor and why: gender, power, and volubility in organizations. Adm Sci Q 2011; 56: 622–41.

90 Akinola M, Martin AE, Phillips KW. To delegate or not to delegate: gender differences in affective associations and behavioral responses to delegation. Acad Manage J 2018; 61: 1467–91.

91 Moss­Racusin C, Rudman LA. Disruptions in women’s self­promotion: the backlash avoidance model. Psychol Women Q 2010; 34: 186–202.

92 Eagly AH, Mladinic A. Are people prejudiced against women? Some answers from research on attitudes, gender stereotypes, and judgments of competence. Eur Rev Soc Psychol 1994; 5: 1–35.

93 Phelan JE, Moss­Racusin CA, Rudman LA. Competent yet out in the cold: shifting criteria for hiring reflect backlash toward agentic women. Psychol Women Q 2008; 32: 406–13.

94 Rudman LA, Glick P. Feminized management and backlash toward agentic women: the hidden costs to women of a kinder, gentler image of middle managers. J Pers Soc Psychol 1999; 77: 1004–10.

95 Rudman LA, Glick P. Prescriptive gender stereotypes and backlash toward agentic women. J Soc Issues 2001; 57: 743–62.

96 Egan ML, Matvos G, Seru A. When Harry fired Sally: the double standard in punishing misconduct. National Bureau of Economic Research. March 2017. https://www.nber.org/papers/w23242.pdf (accessed Nov 1, 2018).

97 Kennedy J, McDonnell M­H, Stephens N. Does gender raise the ethical bar? Exploring the punishment of ethical violations at work. Rochester, NY: Social Science Research Network. April 26, 2016. https://papers.ssrn.com/abstract=2770012 (accessed Sept 24, 2018).

98 Christianson MK. More and less effective updating: the role of trajectory management in making sense again. Adm Sci Q 2017; published online Dec 21. DOI:10.1177/0001839217750856.

99 Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med 2009; 360: 491–99.

100 Gawande A. The checklist manifesto: how to get things right, 1st edn. New York, NY: Picador, 2010.

101 Kaplan S. Beyond the business case for social responsibility. Acad Manage Discov 2019; published online Jan 3. DOI:10.5465/ amd.2018.0220.

102 Kaplan S. Because it’s 2017: gender equality as an innovation challenge. 2017. http://www.rotman.utoronto.ca/Connect/Rotman­ MAG/Back­Issues/2017/Back­Issues­­­2017/Fall2017­Inequality/ Fall2017­FreeFeatureArticle­Kaplan (accessed Sept 14, 2018).

103 Thaler RH, Sunstein CR. Nudge: improving decisions about health, wealth, and happiness, revised and expanded edition. New York, NY: Penguin Books, 2009.

104 Dasgupta N, Asgari S. Seeing is believing: exposure to counterstereotypic women leaders and its effect on the malleability of automatic gender stereotyping. J Exp Soc Psychol 2004; 40: 642–58.

105 Hall WM. Interpersonal triggers and cultural moderators of social identity threat. PhD thesis, University of British Columbia, 2016. 1–171.

106 Murphy MC, Steele CM, Gross JJ. Signaling threat: how situational cues affect women in math, science, and engineering settings. Psychol Sci 2007; 18: 879–85.

107 Rosenthal L, Levy SR, London B, Lobel M, Bazile C. In pursuit of the MD: the impact of role models, identity compatibility, and belonging among undergraduate women. Sex Roles 2013; 68: 464–73.

108 Gollwitzer P, Bargh JA. The psychology of action: linking cognition and motivation to behavior. New York, NY: Guilford Press, 1996.

109 Hirsh JB, Kang SK. Mechanisms of identity conflict: uncertainty, anxiety, and the behavioral inhibition system. Pers Soc Psychol Rev 2016; 20: 223–44.

110 Miller DT, Prentice DA. The construction of social norms and standards. In: Kruglanski AW, Higgins ET, eds. Social psychology: handbook of basic principles. New York, NY: Guilford Press, 1996: 799–829.

111 Brescoll VL, Uhlmann EL. Attitudes toward traditional and nontraditional parents. Psychol Women Q 2005; 29: 436–45.

112 Carli LL. Gender and social influence. J Soc Issues 2001; 57: 725–41. 113 Prentice DA, Carranza E. What women and men should be,

shouldn’t be, are allowed to be, and don’t have to be: the contents of prescriptive gender stereotypes. Psychol Women Q 2002; 26: 269–81.

114 Cialdini RB, Goldstein NJ. Social influence: compliance and conformity. Annu Rev Psychol 2004; 55: 591–621.

115 Cialdini RB, Kallgren CA, Reno RR. A focus theory of normative conduct: a theoretical refinement and reevaluation of the role of norms in human behavior. In: Zanna MP, ed. Advances in experimental social psychology. Cambridge, MA: Academic Press, 1991: 201–34.

116 Vogel L. Pay gap growing between family doctors, specialists. Sept 28, 2017. https://cmajnews.com/2017/09/28/pay­gap­growing­ between­family­doctors­specialists­109­5508/ (accessed Sept 26, 2018).

117 Miller DT, Prentice DA. Changing norms to change behavior. Annu Rev Psychol 2016; 67: 339–61.

118 Kellogg KC. Challenging operations: medical reform and resistance in surgery. Chicago, IL: University of Chicago Press, 2011.

119 Bass BM, Avolio BJ. Transformational leadership and organizational culture. Public Adm Q 1993; 17: 112–21.

120 Taggar S, Ellis R. The role of leaders in shaping formal team norms. Leadersh Q 2007; 18: 105–20.

Review

586 www.thelancet.com Vol 393 February 9, 2019

121 Dobbin F, Schrage D, Kalev A. Rage against the iron cage: the varied effects of bureaucratic personnel reforms on diversity. Am Sociol Rev 2015; 80: 1014–44.

122 Legault L, Gutsell JN, Inzlicht M. Ironic effects of antiprejudice messages: how motivational interventions can reduce (but also increase) prejudice. Psychol Sci 2011; 22: 1472–77.

123 Ryan RM, Deci EL. Self­determination theory and the facilitation of intrinsic motivation, social development, and well­being. Am Psychol 2000; 55: 68–78.

124 Hewlett SA. (Forget a mentor) find a sponsor: the new way to fast­track your career. Boston, MA: Harvard Business Review Press, 2013.

125 Patton EW, Griffith KA, Jones RD, Stewart A, Ubel PA, Jagsi R. Differences in mentor­mentee sponsorship in male vs female recipients of National Institutes of Health grants. JAMA Intern Med 2017; 177: 580–82.

126 Gollwitzer PM. Implementation intentions: strong effects of simple plans. Am Psychol 1999; 54: 493–503.

127 Gollwitzer PM, Sheeran P. Implementation intentions and goal achievement: a meta­analysis of effects and processes. In: Zanna M, ed. Advances in experimental social psychology. Cambridge, MA: Academic Press, 2006: 69–119.

128 Locke EA, Latham GP. Building a practically useful theory of goal setting and task motivation. A 35­year odyssey. Am Psychol 2002; 57: 705–17.

129 Kang S, DeCelles K, Tilcsik A, Jun S. The unintended consequences of diversity statements. March 29, 2016. https://hbr.org/2016/03/ the­unintended­consequences­of­diversity­statements (accessed Sept 24, 2018).

130 Trawalter S, Driskell S, Davidson MN. What is good isn’t always fair: on the unintended effects of framing diversity as good. Anal Soc Issues Public Policy 2016; 16: 69–99.

131 Unzueta MM, Knowles ED, Ho GC. Diversity is what you want it to be: how social­dominance motives affect construals of diversity. Psychol Sci 2012; 23: 303–09.

132 Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA 1999; 282: 1458–65.

© 2019 Elsevier Ltd. All rights reserved.

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.