Skip top navigation Skip to content

Print   Email  Share

Select Task Force Meeting of June 15, 2015 - Workplace Harassment: Examining the Scope of the Problem and Potential Solutions

Written Testimony of Mindy Bergman, Associate Professor of Psychology, Texas A&M University

Thank you for this opportunity to address the EEOC's Select Task Force on the Study of Harassment in the Workplace. I am excited to share the knowledge that has developed about workplace harassment in the field of industrial-organizational psychology, which we call "the science for a smarter workplace" ( The convening of this Task Force by Commissioners Feldblum and Lipnic shows the strong commitment and forward-thinking leadership that we have all come to know and respect from the EEOC.

My name is Mindy Bergman. I am an Associate Professor in the Department of Psychology at Texas A&M University and have been promoted to Full Professor, effective September 1, 2015. I am also a Faculty Fellow in the Mary Kay O'Connor Process Safety Center in the College of Engineering at Texas A&M University. I am also the incoming Associate Head for Undergraduate Studies in the Department of Psychology, effective September 1.

I earned my PhD in 2001 in industrial-organizational psychology from the University of Illinois at Urbana-Champaign. I teach courses in organizational psychology, introductory psychology, and human sexuality. I am a co-investigator on Texas A&M's NSF-funded ADVANCE grant, an institutional transformation grant focused on the advancement of women in STEM fields in academia. I also serve in a variety of service and consulting capacities for upper administration at the University because of my expertise in measurement, survey construction, and diversity. In the over 15 years that I have been in this field, I have published roughly 40 scholarly articles, mostly in the specialty of occupational health psychology (i.e., the study of how workplace structures and events influence the health and well-being of individuals). I have conducted research on a variety of types of harassment and discrimination, including harassment based on sex, race, and linguistic diversity. Most of my research on these topics have used large scale survey designs, although I also use qualitative paradigms and experimental methods to investigate these topics. It is this research that forms the basis of my testimony today.

Before beginning my substantive remarks, I want to note that I am a psychologist, not an attorney. Although my research is informed by the legal perspective, my research is a scientific endeavor. I am attempting to uncover and understand the experiences of people who are harassed, people who harass, and people who work with them. Not all of the experiences of harassment that I study would rise to an actionable level in a court of law or at the EEOC, although certainly some of them do. That does not mean that the experiences we study as psychologists are unimportant. In fact, because we are not examining only those cases that rise to legal definitions and have sufficient evidence to be actionable, we psychologists have a good sense of the range of behaviors that are occurring in organizations, how people react to them, and how people are affected by them.

Most of my comments are based in the sexual harassment literature, but some are also based in literature on harassment due to race/ethnicity and to linguistic diversity. No surprise to anyone on this Task Force, there's a lot that we do know about harassment in the workplace:

  • Harassment is pervasive.
  • Harassment is damaging if not devastating to the people who experience it.
  • Harassment is costly to organizations.
  • Harassment can happen to anyone, but it is more likely to happen to people from particular demographic backgrounds.
  • Harassment is not caused by just "a few bad apples."
  • Harassment types based on different characteristics have much in common with each other.
  • Harassment is underreported.

The first three points--harassment is pervasive, damaging to individuals, and costly to organizations-is well known to the Task Force and one of the reasons why this Task Force was convened. Estimates of the lifetime prevalence of harassment vary, but the best estimates come out of the sexual harassment literature. In the sexual harassment literature, lowest estimates are around 50% of women and the highest are approximately 90% of women; for men, the numbers are lower but still serious, with estimates ranging from 25% experiencing harassment at some point in his lifetime up to 75%. Estimates of racial harassment are also quite high, with estimates ranging from lows of 10% to highs of 75% of workers experiencing racial harassment in the past year. So, there is strong evidence that harassment is prevalent. There is also considerable evidence that harassment is costly to individuals and the organization. Harassment damages individual target's occupational, psychological, and physical well-being. Harassment also pushes good workers to leave an organization, which is costly to the organization in terms of lost productivity and employee replacement costs.

The next four points are ones that I'd like to address further in my oral testimony.

Harassment can happen to anyone, but it is more likely to happen to people from particular demographic backgrounds

There are people of all races, sexes, sexual orientations, national origins, religions, disabilities, and so on, who have experienced harassment based on a protected characteristic. Despite this, we also know that some people are more vulnerable to being targeted for harassment than are others. Generally, the more characteristics a person embodies that are minority or lower power markers, the more likely the person is to be harassed. So, for example, we know that ethnic minority women are more likely to experience sexual harassment than are Caucasian women and more likely to experience racial/ethnic harassment than are ethnic minority men. Of course, there are some exceptions to this general rule-for example, research often shows that Asian women and men are less likely to experience harassment than are other minority groups, and this might be because of the stereotypes of Asians as the "model minority" in the United States or because we lack stereotypes suggesting that Asians are physically threatening or aggressive (unlike stereotypes about other ethnic minority groups in the US).

Relatedly, we know that there is a considerable correlation between experiencing sexual harassment and experiencing racial/ethnic harassment. People who experience one are more likely to experience the other, and higher levels of one is associated with higher levels of the other.

These findings point to the importance of considering intersectionality. Intersectionality refers to the multiple identities that people hold and the patterns of dominance and oppression, of privilege and discrimination, that are linked to these multiple identities. The intersections of identity tell us who is most at risk. They also suggest that the experiences of people who share one identity (e.g., sex) but not another (e.g., race) are likely to have different rates and experiences of harassment overall.

My research on intersectionality has found that even though rates of harassment vary across demographic groups, the effects of harassment are the same once the rate of harassment is accounted for. My research shows that many of the mean differences in well-being outcomes that are seen across demographic groups can be explained by differences in exposure to harassment. Importantly, these differences in well-being are NOT caused by differences in vulnerability (i.e., being "more bothered") to those events. As an example, it is the case that women experience more sexual harassment than men, not that women are "more sensitive" to sexual harassment; thus, although on average women have worse well-being and job attitudes than their male counterparts, these differences can be explained by the differences in exposure to harassment. In my research, I have replicated this effect for men and women as well as across five racial/ethnic groups for both sexual and racial/ethnic harassment.

Harassment is not just caused by a few bad apples.

Undoubtedly, there are some people who are more likely to harass than are others. John Pryor from Illinois State University illustrated this beautifully with a series of quasi-experiments about the sexual harassment proclivities of men. But what seems to be more important to predicting harassment are organizational and interpersonal factors, in addition to the demographic issues described above. So what are these causes of harassment? One of the most important is organizational climate[1], which is the extent to which particular behaviors are expected and formally and informally rewarded in the workplace. Climate is an important driver of harassment because it is the norms of the workplace; it basically guides employees, as Charles O'Reilly and Jennifer Chatman say, to know what to do when no one is watching. When an organization is more tolerant of harassing behavior, more harassing behavior occurs.

Organizational climate is one of the most important drivers of harassment. Workplaces that tolerate harassment have more of it and workplaces that are less tolerant of harassment have less of it. This is a circular problem, because when harassment occurs and organizational leaders do not take it seriously, then the message is that harassment is tolerated, so then it becomes even more OK to harass-and when harassment is taken seriously and shut down, then the message is that harassment is not tolerated. This is a great example of a leadership issue when it comes to harassment. As Benjamin Schneider makes clear in his classic paper, "The people make the place," leaders are major contributors to organizational climate. They set the tone for what is expected in the workplace. If leaders only talk about ending harassment and do not actually act against harassment-by contributing resources to combatting harassment, taking harassment complaints seriously, and remedying substantiated complaints-then the climate will likely continue to be permissive of harassment.

Related to the work I described above, my intersectionality research also shows that climates that allow for harassment disproportionately allow for that harassment to occur against people who have multiple indicators of low(er) power and/or minority status. Thus, my work shows minorities are in a position of vulnerability, such that they are more likely to be harassed in a climate that tolerates harassment than are other people. This is compounded by being a multiple minority, and it is worsened as climate gets worse.

I should also note that I am a psychologist, not a sociologist[2], so I look more at causes of harassment that are more proximal and more immediate-who is in the workplace now, what are the norms now, what are the stereotypes that people hold now. But our colleagues in sociology could better speak to the sociocultural and historical factors that influence harassment, especially the patterns of who is harassed and under what conditions. Without getting into too much detail and going beyond my expertise, I can say this: given the historical and sociocultural conditions in our great nation, it is not a surprise that women are more likely to be harassed than men, that African-Americans, Native Americans, and Hispanics are more likely to be harassed than Caucasians, or that immigrants and English as a second language speakers are more likely to be harassed than domestic-born Americans and English as a first language speakers. These kinds of sociocultural patterns have played out in numerous ways and in numerous arenas of society beyond the workplace for decades and decades in the US and eons and eons throughout history.

Harassment "types" based on different characteristics have much in common with each other

The issues I've reviewed thus far bring us to an important question: to what extent are different kinds of harassment similar? That is, are the causes and consequences of harassment based on sex, race, linguistic diversity, and other protected classes similar? The answer appears to be that they have much in common. Louise Fitzgerald, Charles Hulin, and Fritz Drasgow-my mentors at the University of Illinois when I was a graduate student-proposed a model of sexual harassment that has been tested empirically several dozen times now. Their model is psychological, so it begins by conceiving of sexual harassment as a stressor because it is unwanted and threatening. In their model, sexual harassment is caused by organizational climates that tolerate harassing behaviors and the job-gender context (i.e., the historical and contemporary gender-related employment patterns in the occupation and the historical and contemporary gendered stereotypes about the occupation). Their model also shows that sexual harassment negatively affects psychological, physical, and job-related well-being.

My research extends this model of sexual harassment to other forms of harassment (Figure 1). Essentially, my work proposes that all harassment is a stressor with negative effects on a person's psychological, physical, and job-related well-being. Further, my work proposes that there are many climates related to many kinds of harassment and that these climates are interrelated but not the same-such that organizations that tolerate sexual harassment are more likely to tolerate racial harassment, religious harassment, and so on, but that they are not perfectly correlated. Further, my work argues that there are job contexts that describe the historical and contemporary (a) stereotypes about jobs and (b) employment patterns in jobs, based on many kinds of demographic characteristics. Finally, my extension of this model indicates that some people are more at risk for harassment because of the demographic characteristics that they embody.

Figure 1: General model of harassment causes and consequences

My work demonstrates that this general model holds for racial/ethnic harassment; I also have evidence that preliminarily supports the general model for harassment based on language use (e.g., speaking Spanish in a predominantly English speaking workplace). Thus, there are many commonalities across different kinds of harassment and other kinds of harassment.

Harassment is underreported

The complaints that arrive at the EEOC are undoubtedly just the tip of the iceberg. My own work in sexual harassment and racial/ethnic harassment and discrimination suggests that fewer than 25% of people who experience harassment report it to anyone in power (i.e., own supervisor, harasser's supervisor, etc.). Our results also show that reporting at best does not make things worse and at worst leads to retaliation, minimization of complaints, and additional injury to the reporter. Compared to non-reporters, reporters of sexual harassment had worse job and psychological well-being; this worsening of outcomes could be traced directly to how well the organization responded to reports. Reports that were met with retaliation and minimization of the complaint were dissatisfying, damaging responses from the organization whereas remedies were satisfying, helpful responses from the organization.

These results stand in stark contrast to the Supreme Court opinions in Burlington v. Ellereth and Faragher v. Boca Raton. It is actually unreasonable for employees to report harassment to their companies because minimization and retaliation were together about as common as remedies and created further damage to people who had already been harassed. Further, because remediating the situation did not make the person whole-that is, did not overcome the damage caused by harassment-and helpful vs. hurtful responses were each found about 50% of the time, reporting is a gamble that is not worth taking in terms of individual well-being.

Relatedly, the climate that contributes to the harassment also contributes to whether or not a person will report harassment. That is, as climates are more tolerant of harassment, they are also less encouraging of reporting. Thus, the organizations that have the worst problems might be the least likely to recognize it because (a) their climates make it normative and acceptable to harass and (b) reports that behavior is harassing are less likely to occur.

For what it's worth, harassment isn't the only critical event that goes underreported in organizations. In the workplace safety literature, Tahira Probst has conducted a considerable amount of research on underreporting of injuries and critical incidents. Her work is instructive here because she shows there are several reasons why people do not report workplace injuries and incidents. One is that there are negative consequences to the individual when reporting. Another is that reporting is onerous; it is time consuming, stressful, and sometimes difficult to do.

So, extrapolating from the safety literature as well as the literature on harassment reporting, organizations can work to end harassment by making reporting safer and lesser onerous and by making sure that the organizational members who receive harassment reports are well-trained, helpful, and responsive. This does not mean that the organization must side with the complainant; instead, the organization must take the complaints seriously, treat the complainant with dignity, and protect the complainant from retaliation (regardless of the outcome of harassment investigation).

What evidence do we still need to gather?

Despite the considerable scientific, political, social, and legal attention to the problem of harassment in the workplace, there is still much work to do and much to learn. In the following, I briefly outline a few of the questions that I believe are pressing in order to make strides against harassment in the workplace.

  • Differences across harassment types: Although I have presented evidence that harassment based on different types of protected classes have many commonalities and that the different harassment types are strongly related, there is a need to determine where differences lie. This is especially important when considering prevention and intervention.
    1. Do we need general or specific interventions? Although there appear to be common causes and effects of harassment across types, we don't know if the types of interventions that will be efficacious in ending these different types of harassment will also be similar. Further, we don't know if there is a need for specific interventions (e.g., training to prevent racial harassment, training to prevent sexual harassment), if specific interventions generalize across harassment types (e.g., does training to prevent sexual harassment also prevent other kinds of harassment, such as religious harassment or disability harassment?), and/or if general interventions are efficacious (e.g., training to prevent harassment based on characteristics generally).
  • Long term effects of harassment on well-being, employment, and economic security and propserity: Although we know quite a bit about the effects of harassment over a several year period as well as the effects of harassment on turnover intentions, we do not know much about how harassment today affects a person's employment, well-being, and economic security and prosperity 10, 20, or 50 years in the future. Leaving a job can be costly to a worker in the short term (lack of health care, lack of salary or wages, etc.), but it might also be costly in the long-term. Employees who turnover because of harassment might have difficulty securing additional employment because it might be harder to obtain a good reference. They also might not have learned the full skill set at the job they were in, making it more difficult for them to be promoted at the pace they could have been. Further, there are psychological costs to being harassed and there are psychological costs to being unemployed and underemployed. Thus, a vicious negative cycle might arise from harassment experiences. Long-term longitudinal studies of workers and how harassment influences their (much) later work success and mental health are needed.
  • Harassment among teen workers: The EEOC's Youth at Work program is one of the United States' visible signs of commitment to ending harassment of teen workers and helping teens know their rights at work. What we do not know, however, is how these early work experiences shape later work experiences. In addition to the long-term effects described just above, the experiences of teens are particularly important. The following research questions explain why. Both should be examined via long-term longitudinal studies of working teens, tracking them into their adult years.
    1. What are the developmental effects of harassment of teens? Teens are still developing physically and psychologically, so harassment of these youth might be particularly dangerous to their long-term well-being. Additionally, as these youth are still growing and developing, there might be physical decrements that occur because of workplace harassment. The psychological strain from harassment could reduce psychological resources available for other important life domains (e.g., schooling), negatively influencing achievement in other areas as well.
    2. How do the early work experiences of harassment of teens influence their work-related expectations? Teens are often in their first work role within organizations. The psychology of memory tells us that first experiences set the stage for later experiences; later experiences are judged against and guided by the earlier experiences that people have. So how do these early career experiences influence the expectations that teens have, when later they are adults, in the workplace? Further, how do these early work experiences influence teen choices in career paths? Additionally, teens are still learning the limits of interpersonal interactions inside and outside organizations and other institutions. Could early harassment experiences-whether targeted, perpetrated, or witnessed-make later perpetration more common? Or reporting of harassment less likely?
  • Allies: We know very little about allies (i.e., people who are not themselves part of a particular group that is marginalized, mistreated, or misrepresented but who speak up or act on behalf of the perceived interests of that group or some of its members). Public opinion polls provide some sense of how people feel about workforce participation, equality attitudes, and the like. But public opinion polls don't really tell us much about ally behavior. Some questions we should be addressing about ally behavior:
    1. How effective is ally behavior? It is unknown whether ally behavior helps targets of harassment. When allies act on behalf of others, are they giving these others what they need? We need to discover what and when ally behavior is supportive emotionally as well as professionally.
    2. What conditions encourage allies to actually speak up or act? There are probably many people who hold positive attitudes and have good intentions, but do not speak up due to (for example) fear of retaliation, belief that they are the only ones who think this way, concerns over spending their few social credits when they might be needed later, personal exhaustion or distraction, or other reasons. How can we create a workplace that is conducive to ally behavior? How do we empower allies to speak up?
    3. How do we create allies? Recent research out of Rice University by Rachel Trump and Mikki Hebl suggests that who sends messages about being allies matters; their research shows that men were more persuaded by arguments about gender inequality made by a man than they were by the same arguments made by a woman.
  • Perpetrators: We know some things about perpetrators. For example, in the sexual harassment literature, perpetrators are more likely to be men regardless of whether the target is male or female and they tend to be more powerful than the target. We also know that some personality variables influence the likelihood to harass. But there's a lot we don't know. Here are some questions that we should be able to answer, but we can't or haven't so far, possibly because it is very difficult to study perpetrators. (Frankly, many people do not want to admit to perpetrating harassment when completing organizationally-sponsored surveys, and experiments do not let us see how these harassment unfolds in the field.)
    1. Are there serial perpetrators? The guess is "probably yes." So, assuming that there are serial perpetrators, what proportion of harassment is perpetrated by these persons? This is an important question because if a large proportion of harassment is perpetrated by a small number of persons, then a simple intervention is to fire these people.
    2. What signals from climate indicate that it is a "safe" time to harass? Most of our research focuses on broad climate and broad harassment experiences-that is, climate over a period of time and harassment over a period of time. But what are the moment-to-moment signals in the environment that indicate that perpetration should happen now? Such information might provide insights into changing workplace behavior, work spaces, and work practices that could reduce harassment "opportunities."

Concluding remarks

Thank you again for the opportunity to address the Task Force on this pressing matter. I hope that my remarks are useful and informative as the Task Force conducts its business. Many academics never get the opportunity to speak this directly to our national leadership on the topics that we study so closely-and it is something that many of us dream of. It is a true honor to share my knowledge with you and contribute to the important work of the EEOC.

References supporting this testimony

Bergman, M.E., & Drasgow, F. (2003). Race as a moderator in a model of sexual harassment: An empirical test. Journal of Occupational Health Psychology, 8, 131-145.

Bergman, M.E. & Henning, J.B. (2008). Sex and ethnicity as moderators in the relationship between sexual harassment climate and sexual harassment. Journal of Occupational Health Psychology, 13, 152-167.

Bergman, M.E., Langhout, R.D., Palmieri, P.A., Cortina, L.M., & Fitzgerald, L.F. (2002). The (un)reasonableness of reporting: Antecedents and consequences of reporting sexual harassment. Journal of Applied Psychology, 87, 230-242.

Bergman, M.E., Palmieri, P.A., Drasgow, F., & Ormerod, A.J. (2007). Racial and ethnic harassment and discrimination: In the eye of the beholder? Journal of Occupational Health Psychology, 12, 144-160.

Bergman, M.E., Palmieri, P.A., Drasgow, F., & Ormerod, A.J. (2012). Racial/ethnic harassment and discrimination, its antecedents, and its effect on job-related outcomes. Journal of Occupational Health Psychology, 17, 65-78. doi: 10.1037/a0026430

Bergman, M.E., Watrous, K.M., & Chalkley, K.M. (2008). Identity and language: Contributions to and consequences of speaking Spanish in the workplace. Hispanic Journal of Behavioral Sciences, 30, 40-68.

Barratt, C., Bergman, M.E., & Thompson, R.D. (2014). Masculinity and femininity as predictors of workplace discrimination and support for female federal police officers. Sex Roles, 71, 21-32.

Buchanan, N.T., Bergman, M.E., Bruce, T. A., Woods, K.C., & Lichty, L. F. (2009). Unique and joint effects of sexual and racial harassment on college students' well-being. Basic and Applied Social Psychology, 31, 267-285.

Cunningham, G.B., Bergman, M.E., Miner, K.N. (2014). Interpersonal mistreatment of women in the workplace. Sex Roles, 71, 1-6.

Fitzgerald, L.F., Drasgow, F., Hulin, C.L., Gelfand, M.J., & Magley, V.J. (1997). Antecedents and consequences of sexual harassment in organizations: A test of an integrated model. Journal of Applied Psychology, 82, 578-589.

Fitzgerald, L.F., Hulin, C.L., & Drasgow, F. (1995). The antecedents and consequences of sexual harassment in organizations: An integrated model. In G. Keita & J. Hurrell, Jr. (Eds.), Job stress in a changing workforce: Investigating gender, diversity, and family issues (pp. 55-73). Washington DC: American Psychological Association.

Glomb, T.M, Munson, L.J., Hulin, C.L., Bergman, M.E., & Drasgow, F. (1999). Structural equation models of sexual harassment: Longitudinal explorations and cross?sectional generalizations. Journal of Applied Psychology, 84, 14-28.

Langhout, R.D., Bergman, M.E., Cortina, L.M., Fitzgerald, L.F., Drasgow, F., & Williams, J.H. (2005). Sexual harassment severity: Assessing situational and personal determinants and outcomes. Journal of Applied Social Psychology, 35, 975-1007.

Mazzeo, S.E., Bergman, M.E., Buchanan, N.T., Drasgow, F., & Fitzgerald, L.F. (2001). Situation specific assessment of sexual harassment. Journal of Vocational Behavior, 59, 120-131.

O'Reilly, C.A., & Chatman, J.A. (1996). Culture as social control: Corporations, cults, and commitment. Research in Organizational Behavior, 18, 157-200.

Ostroff, C., Kinicki, A. J., & Muhammad, R.S. (2012). Organizational culture and climate. In I.B. Weiner, N.W. Schmitt, & S. Highhouse (Eds.), Handbook of psychology: Industrial and organizational psychology (2nd ed., Vol. 12, p. 643-676). New York, NY: John Wiley.

Probst, T.M., Brubaker, T.L., & Barsotti, A. (2008). Organizational injury rate underreporting: The moderating effect of organizational safety climate. Journal of Applied Psychology, 93, 1147-1154.

Probst, T.M., & Estrada, A.X. (2010). Accident under-reporting among employees: Testing the moderating influence of psychological safety climate and supervisor enforcement of safety practices. Accident Analysis and Prevention, 42, 1438-1444.

Probst, T.M., & Graso, M. (2013). Pressure to produce = pressure to reduce accident reporting? Accident Analysis & Prevention, 59, 580-587.

Pryor, J.B. (1987). Sexual harassment proclivities in men. Sex Roles, 17, 269-290.

Schneider, B. (1987). The people make the place. Personnel Psychology, 40, 437-454.

Schneider, B., & Reichers, A.E. (1983). On the etiology of climates. Personnel Psychology, 36, 19-39.

Trump, R.C.E., Parker, N., Hebl, M., & Nittrouer, C.L. (2015, April). Engaging men: How men can serve as allies to women. In L.R. Martinez & E.N. Ruggs (Chairs), Bystanders, allies, and advocates: Recognizing "others" in workplace crisis situations. Symposium conducted at the 30th annual conference for the Society for Industrial and Organizational Psychology, Philadelphia, PA.

Wasti, S.A., Bergman, M.E., Glomb, T.M., & Drasgow, F.  (2000). Generalizability of an integrated sexual harassment model: A cross-cultural comparison.  Journal of Applied Psychology, 85, 766-778.



[1] The definition of "organizational climate" might sound like "organizational culture" to people who are familiar with the latter term. In industrial-organizational psychology and management, we make a distinction between climate and culture. Culture is deeply held values and underlying ideologies; these values and ideologies are often implicit, making it difficult for people to articulate these deeply held beliefs because they form a lens through with they interpret the world. Culture supplies the "why" for the decisions that organizations make. Climate, in contrast, is more surface than culture and supplies the "what" of decisions that organizations make. Climate is more easily articulated by employees, because it is a reporting of what behaviors are expected, rewarded, and supported at work. Climate and culture are highly related, as the "what" is expected behavior from climate is justified by the "why" from culture; conversely, the "what" of climate helps illuminate the "why" of culture. Most non-academics who talk about "organizational culture" are referring to both organizational culture and organizational climate as defined in the academy. Ostroff, Kinicki, and Muhammad (2012) nicely articulate this distinction; their Figure 24.1 is especially useful for understanding the interplay between culture and climate.

[2] Psychology is the science of mind and behavior; it focuses primarily on human cognition, attitudes, beliefs, behaviors, and well-being. Sociology is the science of the development, structure, and functioning of human societies and institutions and how these societies and institutions interact. Psychology and sociology inform each other, but generally psychology is interested in the individual and interpersonal (e.g., romantic relationships) levels of analysis whereas sociology is interested in the institutional and societal levels of analysis.