Tuesday, August 12, 2014

sexual harassment in science

I usually save this site for posting original posts and use my tumblr to post about new articles and such - but this was an incredible piece that I need to share!

http://www.nytimes.com/2014/08/12/science/harassment-in-science-replicated.html

CreditKatherine Streeter
Continue reading the main storyShare This Page
As an undergraduate student in biology, I spent several weeks in Costa Rica one summer with an older graduate student on a research project deep in the cloud forest. It was just the two of us, and upon arriving at our site, I discovered that he had arranged a single room for us, one bed.
Mortified but afraid of being labeled prudish or difficult, I made no fuss. I took the lodge owner aside the next day and requested my own bed. The problem ended there, and my graduate student boss never made any physical advances.
Reflecting back, I’m struck by how ill equipped I was to deal with this kind of situation, especially at 19. My university undoubtedly had a harassment policy, but such resources were thousands of miles away. I was alone in a foreign country and had never received any training on my rights and resources in the field.
I’d forgotten about this experience from two decades ago until I read areport published July 16 in the journal PLOS One. Kathryn Clancy, an anthropologist at the University of Illinois at Urbana-Champaign, and three colleagues used email and social media to invite scientists to fill out an online questionnaire about their experiences with harassment and assault at field sites; they received 666 responses, three quarters of them from women, from 32 disciplines, including anthropology, archaeology, biology and geology.
Audio
Almost two-thirds of the respondents said they had been sexually harassed in the field. More than 20 percent reported being sexually assaulted. Students or postdoctoral scholars, and women were most likely to report being victimized by superiors. Very few respondents said their field site had a code of conduct or sexual harassment policy, and of the 78 who had dared to report incidents, fewer than 20 percent were satisfied with the outcome.
The findings are depressingly similar to the data some colleagues and I collected this year from an online questionnaire sent to science writers. We received responses from 502 writers, mostly women, and presented our results at M.I.T. in June during Solutions Summit 2014: Women in Science Writing, a conference funded by the National Association of Science Writers.
More than half of the female respondents said they weren’t taken seriously because of their gender, one in three had experienced delayed career advancement, and nearly half said they had not received credit for their ideas. Almost half said they had encountered flirtatious or sexual remarks, and one in five had experienced uninvited physical contact.
Given their voluntary nature, neither report can be expected to tell us the true incidence of sexual discrimination and harassment among scientists and science writers. Still, the volume of responses sends an unmistakable message: Four decades after Title IX outlawed sex-based discrimination in public education and 23 years after Anita Hill pushed sexual harassment into the limelight, bias and harassment continue to hinder women’s progress.
Dr. Clancy says she decided to collect data after being overwhelmed with responses to a post she published on her blog at Scientific American in 2012. A female student, “Hazed,” recounted life in her graduate program:
“My body and my sexuality were openly discussed by my professor and the male students,” the woman wrote. “Comments ensued about the large size of my breasts, and there was speculation about my sexual history.” Her professor, she said, “often joked that only pretty women were allowed to work for him, which led me to wonder if my intellect and skills had ever mattered.”
Comments and emails poured in, Dr. Clancy said: “One story quickly became two stories, and quickly became what felt like 100.”
Similarly, our survey of writers grew out of well-publicized harassment accusations against a prominent male editor who was a mentor to many female writers. Those incidents led women to come forward with their stories of discrimination throughout the profession.
In academia, accusations of sexual harassment or assault are usually handled internally, Dr. Clancy says, and this can create powerful incentives to cover up bad behavior, especially among perpetrators with tenure and power. “I’ve heard too many stories about the professor who isn’t allowed to be in a room with X, Y and Z anymore,” she said. Sometimes perpetrators even benefit by getting out of dreaded teaching assignments while keeping their jobs.
Harassment among science writers spawned a hashtag,#ripplesofdoubt, to describe how harassment undermines women. Some women who had been passed over for jobs wondered if they had been rejected for their looks rather than their work. Others worried that they might not have attained their positions on merit.
Indeed, data suggest bias in mentoring decisions. In a study published this year, a team of researchers led by Katherine L. Milkman at the University of Pennsylvania sent identical letters, purportedly from students, to more than 6,500 professors at 259 universities asking to discuss research opportunities. Professors were more likely to respond to email from “Brad Anderson” than from fictitious aspirants with names like Claire Smith or Juan Gonzalez. Such bias perpetuates discrimination.
“Our world is small and our resources are scarce,” said another author of the PLOS One report, Julienne Rutherford, a biological anthropologist at the University of Illinois at Chicago. If women are dissuaded or excluded from even a handful of opportunities, she continued, the loss to science is enormous.
Last year, at the annual conference of the National Association of Science Writers, I joined five leading female science writers to presentdata we had collected on gender disparities in bylines, top-level jobs, awards and salaries, and to recount personal stories of times when our gender had stood in the way of our careers.
Afterward, long lines formed at the microphones as people in the audience stood up to share their stories. Young women told of being harassed by sources. Seasoned journalists recalled male bosses with wandering hands.
Men rose to offer support. The director of a prominent science writing program said that the next time one of his students confided she was being harassed in an internship, he was going to intervene. (Apparently it had not occurred to him before.)
Most men are not creeps, and they have a powerful role to play here. During a field trip at a journalism conference a few years ago, I had an engaging conversation with a keynote speaker. As we parted, he told me, in front of two other men, “Your husband shouldn’t let you out of the house.”
The two bystanders brushed off this insulting attempt at a compliment. It was easier for them to let it go than to call out a friend, and their behavior said it was all right to treat me like that.
Whether harassment or discrimination takes place at a field site in Costa Rica or in a conference room, the problem will not be solved with new rules archived on unread websites. The responsibility for pushing back should not rest solely with the victims. Solutions require a change of culture that can happen only from within.
It will take chief executives, department heads, laboratory directors, professors, publishers and editors in chief to take a stand and say: Not on my watch. I don’t care if you’re my friend or my favorite colleague; we don’t treat women like that.

Why are there fewer women in the sciences? A look into sex differences in mentor-protégé relationships

          Why are there fewer women in the sciences? This question has recently appeared in the headlines of many popular media sources, including The New York Times, and has been the focus of an incredible amount of scientific studies (Pollack, 2013; Halpern et al., 2007; Ceci & Williams, 2010). Many studies have examined innate cognitive differences and early socialization factors as the root of these differences (Halpern et al., 2007; Ceci & Williams, 2010). Despite significant differences in these early factors, there seems to be no drastic effects on women’s early participation in science related fields overall:[1] throughout high school, women tend to outperform men in math in science classes, and enroll in the same number of science and math classes as men (Halpern et al., 2007; Ceci & Williams 2010). Furthermore, according to the National Science Foundation (2013), women earn about 50.3% of all science and engineering bachelor degrees. After graduation, however, the attrition rate of women from science related fields drastically increases (Halpern et al., 2007). This suggests that the substantial decrease in the number of women in the field must be more heavily related to factors that become increasingly relevant during and after women’s undergraduate years, like the development of mentor-protégé relationships. In this post I will investigate potential sex differences in access to, benefits of, and level of support in mentor-protégé relationships and ways in which they may facilitate the attrition of female participation in the sciences.

Mentor-protégé relationships: Definitions and benefits
Throughout the literature, the definition of a mentor, as well as the mentor-protégé relationship, varies (Jacobi, 1991). Despite general differences, there are many common underlying themes present in the definitions used across studies – definitions all focus on an established personal connection between a protégé and a more senior mentor that is reciprocal, and highlight similar ways in which the relationship functions. The primary functions of these relationships are to influence and foster protégé success through emotional and psychological support, direct assistance, role modeling, and professional development (Jacobi, 1991).
The support of mentor-protégé relationships during undergraduate years, and throughout the course of one’s profession, is instrumental in later success. Many studies have documented the positive effects of such relationships, including: higher compensation, more promotion opportunities, increased chance of early success, and increased career satisfaction (Young et al., 2006). These relationships, explicitly during undergraduate years, could also work to keep women engaged in male-typical majors, like the sciences, and work to buffer women from forms of discrimination that may cause them to leave the field altogether (Jacobi, 1991; Ragins and Cotton, 1999). While the benefits of protégé-mentor relationships may work to decrease the attrition rate of women in the sciences, often times women have limited access to mentors, establish different mentor relationships than their male peers, and tend to have differing outcomes as a result (Young et al, 2006; Jacobi et al., 1991; Milkman et al., 2014; Johnson, 1989; Bushardt et al., 1991; Ragins and Cotton, 1991; Klabflisch and Keyton, 1995).
             
Sex differences in factors of mentor-protégé relationships
            Johnson (1989) found that women were less likely than their male peers to have a mentor during their undergraduate years. Furthermore, many studies document a significant difference in the way professors respond to particular students’ initiation of mentor-protégé relationships (Young et al., 2006). A more recent study that investigated differences in professors responses to student inquiries in terms of sex and ethnicity/race (Milkman et al, 2014). In this study, researchers sent out 6,500 emails to faculty at 259 top institutions across the country pretending to be students – every email was exactly identical with the exception of the name (the name was either male or female, white or nonwhite). Within the emails, “students” would discuss their interest in the professor’s particular field of study, ask if they had any potential insights, and whether they would be willing to discuss potential research opportunities. Professors across 89 disciplines responded to emails that contained a white male name more often than they responded to emails that contained a white female name, or a male or female nonwhite name. In particular, professors in the life sciences, health sciences, and physical and mathematical sciences, responded to emails sent by women of all ethnic and racial backgrounds (with the exception of black male and females) significantly less often than they responded to their ethnicity/race matched male counterparts. These findings suggest that men, and particularly white men, tend to have greater access to potential mentors than women do within the sciences. Lack of responses to informal mentoring inquiries may cause women, as well as those of color, in the sciences to lose interest or leave the field because they are unable to compete with their male peers with greater access to the benefits of mentor-protégé relationships. 
Women who attempt to establish mentor-protégé relationships report encountering obstacles, like lack of responses to inquiries, more often than their male peers (Young et al., 2006; Ragins and Cotton, 1991). Ragins and Cotton (1991) surveyed employees (229 women and 281 men) from three research and developmental organizations about their experience with mentors and their perceived obstacles in obtaining a mentor-protégé relationship. They found that men and women reported no significant difference in extensive mentoring experience,[2] however men experienced more moderate levels of mentoring than women did. In addition, women reported having a more difficult time obtaining a mentor than males did, even after controlling for age, experience, rank, and position. While it is possible that the observed difference in perceived obstacles may be due to women overestimating difficulty, the findings may also suggest that the women in the study may have had to work harder to obtain the same mentor experiences as their male colleagues.
Perceptions of protégés may contribute to a mentor’s decision in selecting a protégé, and may serve as an obstacle to gaining access to a mentor. Mentors often assess, implicitly or explicitly, a student’s fit within the field in question and potential before pursuing a mentor-protégé relationship (Young et al. 2006). As the success of protégés often leads to rewards within a mentor’s career, including increased peer and manager recognition, and failed protégés, who are unable to accumulate success or drop out of the field, diminish their mentor’s success, selecting the right protégé is important for a mentor (Young et al. 2006). As such, strong negative stereotypes about women in science, including diminished capabilities in visuospatial and quantitative abilities, may decrease the likelihood that a mentor will select a female protégé. The accuracy of such perceptions does not entirely matter as the perceptions themselves may also work as a barrier for women seeking access to informal mentoring.
In addition to perceptions of the protégé, Scandura and Ragins (1993) found that the gender of the field is an important factor in predicting access to mentor support (as cited by Young et al. 2006). If the field is more masculine (defined as having at least 70% male representation), female protégés tend to have less access to mentors. Decreased access to mentors in male-dominated fields may simply stem from there being fewer women in higher tier positions to serve as mentors for female protégés.  
As there are more male scientists than female scientists, female students have less access to same-sex mentors as their male peers do. As a result, females who develop mentor-protégé relationships within these fields tend to have a male mentor (Young et al., 2006). Many studies have shown that female protégés tend to actually benefit more from having a male mentor than they do having a female mentor (although data is inconsistent within the literature), however, cross-sex mentor-protégé relationships may be more difficult to establish than same-sex mentor-protégé relationships (Young et al., 2006).
Several of studies have shown that male mentors and female protégés tend to report fear that mentorships may be perceived as having some sort of sexual undertone (Young et al. 2006; Bushardt et al., 1991; Ragins and Cotton, 1991). This fear often prevents the initiation of such relationships. Bushardt et al. (1991) suggests that these fears also prevent established cross-sex mentor-protégé dyads from engaging in informal interactions, such as meeting behind closed doors or eating meals with one another alone. Decreased access to informal social encounters due to social reasons prevents cross-sex mentor-protégé dyads from establishing closer and long-term bonds with one another, factors that promote later career outcomes typical of such relationships. Perhaps as a result, several studies have shown that males tend to have longer-term, more enriching mentorships than their female peers (Young et al., 2006; Ragins and Cotton, 1999).  
While male-female cross-sex mentor-protégé relationships are harder to establish, in male-dominated industries and in more masculine positions, they tend to yield the highest benefits. Dougherty et al. (2013) examined differences in mentor relationships and outcomes in male dominated fields. They based their research on the signaling theory, which proposes visibility of the relationship with high status mentors, namely males in more senior positions within a company, signal the protégé’s high potential and lead to increased career outcomes. The authors posited that females should benefit the most under the framework of this theory. While males should benefit the most when we consider all other factors, signaling theory posits because those kinds of relationships are more common, the tokenism of the senior-male and female protégé is more salient and amplifies the signal that she is a contender for upward mobility (Dougherty et al., 2013).
 In their first study, Dougherty et al. (2013) collected questionnaires from male and female professionals who graduated from an undergraduate state business school program 11 years earlier. The questionnaire collected information on the subject’s mentor-protégé relationship experience, their gender, and their career outcomes (salary and career satisfaction). All respondents (n=356) used in the data analysis were employed in full-time positions, 98% self-identified as Caucasian, and they all worked within male-dominated business firms. After controlling for confounding factors (years with employer, number of company changes, level of education, socioeconomic status, organization size, protégé job level, etc.), they found that females with experience in mentor-protégé relationships with a senior-male mentor did not report higher career outcomes than males with similar relationships as they had expected; women and men with senior-male mentors actually fared about the same. However, there was a greater difference in career outcome reports between females with and without senior-male mentors (those who had senior-male mentors had greater career outcomes) than between males with and without senior-male mentors. This suggests that the role of a senior-male mentor may offset potential sex differences that may deter or prevent women from achieving similar levels of success as their male counterparts.
Dougherty et al. (2013) conducted a second study to determine whether these findings could be replicated in a more male-dominated field and a male-dominated specialty position –software engineers at a major aerospace manufacturing firm. They collected questionnaires from males (n=292) and females (n=56) with similar educational backgrounds and experience in the firm. The questionnaires, like in their first study, reported the subject’s gender, mentor relationship, and career outcome (salary and career satisfaction). Again, they found that women with senior-male mentors fared about the same as men in the field while females with non-senior mentors fared far worse than males with non-senior-male mentors. This supports the previous suggestion that having a senior-male mentor could potentially offset sex differences in career outcomes in male-dominated fields, like the sciences, for women.
      While the study conducted by Dougherty et al. (2013) presents interesting data, there are many flaws that make the conclusions less reliable. First, the degree of the relationship the subject has with their mentor may differ within the sample and could have potentially skewed the results. Second, the number of female respondents was significantly lower than that of male respondents, which could have again led to skewed data. And in addition, we cannot determine that the mentor relationship was the cause of enhanced career outcome, as other factors, such as the protégé’s individual achievement and personality, may have independently led to their success. Furthermore, those with higher achievement and potential may have had increased access to mentors in general.
A similar study, conducted by Ragins and Cotton (1999), investigated career outcomes in relation to mentor-protégé relationships. Female (n=654) and male (n=500) subjects across fields were assessed using standard surveys used to determine history of mentoring relationships, including The Mentor Role Instrument (MRI), and standardized career outcome surveys, including Duncan’s Multiple Range test. Overall, they found that women and men across fields do not differ in the likelihood of having mentor. They also found that those with male mentors tended to have enhanced career outcomes than other groups and that mentor-protégé relationships were less effective for female protégés than male protégés overall. This difference could be due to the differences in the type of relationship that female protégés develop with their mentors.
The amount of support provided by the mentor may differ on the basis of sex, causing differences in career outcomes. In male-dominated fields, even when females obtain a male mentor, stereotyping may lead mentors to diminish the contributions of a female protégé, or prevent them from developing a closer relationship (Young et al., 2006). Furthermore, the mentor-protégé relationship may be more centered around male-relationship patterns than female-relationship patterns, causing them to be less beneficial for females. Klabflisch and Keyton (1995) propose that mentor-protégé relationships develop in a similar fashion as female friendship relationships, which may account for differences in levels of support and resulting outcomes. Females tend to develop their relationships by sharing personal accounts and interacting with others one-on-one, while males tend to interact in more of a group setting. These differences in interaction style, may affect the closeness and development of the mentor-protégé relationship.
Another explanation is that protégés may be more liked by their mentors based on interests, values and sex (Young et al. 2006). Ensher and Murphy (2007) found that those who considered themselves to be more similar to their mentors tended to have more positive and increased interactions with them. As a result, since females tend to have more cross-gender mentor-protégé relationships within the sciences due to the nature of the field, females may be unable to develop closer, more meaningful relationships with their mentors because they have different interests based on their gender. A study by Olian et al. (1988) however suggests that gender is not a significant factor, and that the field of interest and future career goals are more important factors in developing close and meaningful relationships (as cited by Young et al., 2006). In addition, Ragins and McFarlin (1990) found that there were no sex differences in terms of the amount of support female and male mentors gave to their protégés (as cited by Young et al. 2006). While the amount of support a person receives may be dependent on gender, determining the roots of differences in support and making sense of the contradictions in the field is rather difficult due to the plethora of other potential factors, including academic achievement, socioeconomic status, and race.
Inconsistencies within the literature

While there is a substantial body of literature that suggests men have more access to mentors than female peers, there are opposing studies that report men and women have similar access to mentors, and some that even suggest women have increased access (Young et al., 2006; Fuentes et al., 2014). Fuentes et al. (2014) examined student-faculty relationships and contact throughout college and across all disciplines. Students in the study took standardized self-report surveys prior to their freshman year, after their freshman year, and immediately after their senior year. They found that students who were higher achieving in high school, tended to have less contact with faculty during their first year. In addition, they found that students of color and those who were undecided about their major also had increased early contact. While these groups had increased initial contact with faculty, they tended to have less contact and experience with mentors after completing their senior year. In addition, the study found that females tended to have the most experience with mentor relationships by the end of their senior year, perhaps due to a tendency for female-typical interaction patterns, including participating in more one-on-one encounters. In addition, Erkit and Mokros (1984) surveyed 723 students from 6 liberal arts colleges and found that mentoring relationships may be more of a by-product of academic achievement rather than gender (as cited by Young et al., 2006). These findings show how many factors attribute to establishing mentor-protégé relationships and how results of studies differ based on who is looking and what factors they are examining. While such studies contradict previous findings, they are also not specific to male-dominated fields; as such there may still be differences within the sciences in regard to experience with mentor-protégé relationships.

Conclusion: Do differences in access to and degree of mentor-protégé relationships potentially play a role in increased attrition rates within the sciences?
The impact differences in access to, amount of support, and outcomes of mentor-protégé relationships on female retention within science related fields is hard to determine. Several studies suggest that women have decreased access to mentor-protégé relationships due to a variety of factors (Young et al., 2006, Milkman et al., 2014; Johnson et al., 1984; Ragins and Cotton, 1991), may create a disparity within the sciences that actively weeds women out of the field by causing them to be less competitive than their male peers. While mentor-protégé relationships have clear advantages in fostering interest and later success in an individual’s career, differences access to, benefits of, and other factors of these relationships are too highly debated within the literature to reach any definite conclusions. In addition, research that explicitly focuses on women within science related fields and during a more narrow time period (ie. undergraduate years) is largely missing. In order to determine the potential impact of differences in mentor-protégé relationships, more research needs to be conducted that explicitly focuses on the role mentor-protégé relationships plays in the sciences.    


References:
Bushardt, S. C., Fretwell, C., & Holdnak, B. J. (1991). The mentor/protégé relationship: A biological perspective. Human Relations, 44, 619-639.Ceci, S. J., & Williams, W. M. (2010). Sex differences in math-intensive fields. Current Directions in Psychological Science, 19(5), 275-179.
Dougherty, T. W., Dreher, G. F., Arunachalam, V., & Wilbanks, J. E. (2013). Mentor status, occupational context, and protégé career outcomes: Differential returns for males and females. Journal of Vocational Behavior, 83, 514-527.

Ensher, E. A., & Murphy, S. E. (1997). Effects of race, gender, perceived similarity, and contact on mentor relationships. Journal of Vocational Behavior, 50, 460-481.
Fuentes, M. V., Alvarado, A. R., Berdan, J., & DeAngelo, L. (2014). Mentorship matters: Does early faculty contact lead to quality faculty interaction? Research in Higher Education, 55, 288-307.

Halpern, D. F.,  Benbow, C. P., Geary, D. C., Gur, R. C.,Hyde, J. S., & Gernsbacher, M. A. (2007). The science of sex differences in science in mathematics. Psychological Science in the Public Interest, Supplement, 8(1), 1-51.

Jacobi, M. (1991). Mentoring and undergraduate success: A literature review. Review of Educational Research, 61(4), 505- 532).

Johnson, C. S. (1989). Mentoring programs. In M. L. Upcraft & J. Gardner (Eds.), The freshman year experience: Helping students survive and succeed in college (pp. 118-128). San Francisco: Jossey-Bass.
Kalbfleisch, P. J., & Keyton, J. (1995). Power and equality in mentoring relationships. In P. J. Kalbfleisch & M. J. Cody (Eds.), Gender, power, and communication in human relationships (pp. 189-212). Hillsdale, NJ: Lawrence Erlbaum.

Milkman, Katherine L. and Akinola, Modupe and Chugh, Dolly, What Happens Before? A Field Experiment Exploring How Pay and Representation Differentially Shape Bias on the Pathway into Organizations (April 23, 2014). Available at SSRN: http://ssrn.com/abstract=2063742 or http://dx.doi.org/10.2139/ssrn.2063742
NSF. (2013).Women, Minorities, and Persons with Disabilities in Science and Engineering. Retrieved from http://www.nsf.gov/statistics/wmpd/2013/tables.cfm
Pollack, E. (2013, October 5). Why are there still so few women in science? The New York Times. Retrieved from http://www.nytimes.com/2013/10/06/magazine/why-are-there-still-so-few-women-in-science.html


Ragins, B. R., & Cotton, J. L. (1991). Easier said than done: Gender differences in perceived barriers to gaining a mentor. Academy of Management Journal, 34(4), 939-951.

Ragins, B. R., & Cotton, J. L. (1999). Mentor functions and outcomes: A comparison of men and women in formal and informal mentoring relationships. Journal of Applied Psychology, 84(4), 529-550.

Young, A. M., Cady, S., & Foxon, M. J. (2006). Demystifying gender differences in mentoring: Theoretical perspectives and challenges for future research on gender and mentoring. Human Reseouce development Review, 5 (2), 148-175.



[1] There are differences in women’s participation in certain sectors of science however; women only obtain 18.2% of undergraduate computer science degrees, 18.4% of engineering degrees, and 43.1% of mathematics and statistics degrees (NSF, 2013).
[2] While they did not find any differences in extensive mentoring experience, the study only examined those within the organizations and those willing to respond to a survey on mentor-protégé relationships, which may have skewed the data.