9.3.1 Understand gender bias and how it can perpetuate gender inequality even when we do not realize it.
Gender inequality is often tricky to study because we don’t always see it for what it is. Gender stereotypes affect boys and girls, women and men, throughout their lives in ways that are not only easy to perceive, but also in ways more challenging to actually observe. Sometimes, ideas about natural gender differences prevent us from seeing how much we’re doing to make things harder for women and easier for men. And because of this, we sometimes fail to see gender inequality as inequality; it often masquerades as a natural gender difference. For instance, if we believe men are naturally more competent at, say, building and using tools, do you think this belief might affect whether we hire women for construction work? Remember that when sociologists use the term gender bias we are referring to the ways that we systematically treat men and women (and boys and girls) differently.
Proving gender bias is harder than you might think. How do we know that it’s gender bias at work and not something else? Here, we’ll consider two examples of gender bias relying on data from two sources: symphony orchestras and student evaluations of college professors. Both help to illustrate that gender inequality is often perpetuated in ways that are difficult to perceive. What if employers don’t think they are biased against women, but seem to disproportionately hire men anyway? What if students don’t think they are biased against women, but seem to rate the women teaching them less favorably than the men? Are they biased against women?
Two economists, Claudia Goldin and Cecilia Rouse (2000), provide one example of this issue in an interesting place—symphony orchestras (see FIGURE 9.2). How musicians tried out for symphony orchestras changed dramatically in the 1970s and 1980s in most major U.S. orchestras. Auditions were made “blind”—the identities of the musicians trying out were concealed from the judges. Generally, this meant that a screen was put up so that judges would not be able to see the person trying out. The screen was actually initially put up to guard against judges selecting only students from the most elite music schools. But screens had an unanticipated consequence as well—suddenly women became much more likely to be selected to join the orchestra based on their auditions. This is something social scientists sometimes refer to as a “natural experiment.” It would be unethical to conduct an experiment testing whether gender bias affected the occupational destinies of actual musicians. But because orchestras did this at different times and we can control for a variety of other factors (like the rise in women musicians more generally), blind auditions allow us to see how women fared in different orchestra auditions using different audition strategies.
These data illustrate shifts in the proportion of women in five orchestras between 1940 and 1996. These five orchestras are sometimes referred to as “The Big Five”: the New York Philharmonic, the Boston Symphony Orchestra, the Chicago Symphony Orchestra, the Philadelphia Orchestra, and the Cleveland Orchestra. In their research, they found that all major symphony orchestras that began employing blind auditions in the 1970s followed a similar pattern: they started hiring more women.
The following interactive is not accessible to keyboard and screen readers. What follows is an explanation of what appears on the screen. A line graph shows the changes in the proportion of women in U.S. symphony orchestras from 1940 to 1996.
The graph has “Proportion of Women in Orchestra” in percentage on the Y-axis and years from 1939 to 1995, in increments of 2, on the X-axis.
The data depicted in the graph are as follows:
Boston Symphony Orchestra: Starting with 1 percent in 1945, the value rises to 3 percent in 1955, to 4 percent in 1963, to 10 percent in 1976, to 14.5 percent in 1982, to 19 percent in 1987, and reaches 23 percent in 1993.
Chicago Symphony Orchestra: Starting with 6 percent in 1945, the value dips slowly to 3 percent in 1955, to 2 percent in 1958, follows a zig zag pattern to 9.5 percent in 1978, rises to 14 percent in 1986, to 21 percent in 1991, and dips slightly to reach 20 percent in 1995.
Cleveland Symphony Orchestra: Starting with 3 percent in 1950, the value rises to 6 percent in 1955, to 10 percent in 1965, to 14 percent in 1973, to 16 percent in 1983, to 20 percent in 1989, and dips to 14 percent in 1993.
Philadelphia Orchestra: Starting with 2.5 percent in 1939, the value rises to 7.5 percent in 1948, then dips to 5 percent from 1954 to 1963, rises to 9 percent in 1967, to 15 percent in 1985, to 19 percent in 1991, and dips slightly to reach 17.5 percent in 1994.
New York Philharmonic: Starting with 1 percent in 1939, the value remains at 1 percent till 1967, rises to 5 percent in 1973, to 10 percent in 1979, to 16 percent in 1983, to 25 percent in 1990, and reaches 34 percent in 1996.
In 1970, just about 10 percent of major orchestra members were women. By the mid-1990s, women jumped to roughly 35 percent. And the number of new hires for orchestras was actually declining over this period. So, the proportion of new hires who were women is even larger than it appears. Goldin and Rouse (2000) were able to show that blind auditions did cause musicians to receive more impartial auditions and for more women playing in major symphony orchestras today. This means that gender bias was present even though people might not have realized it was there. Perhaps beliefs about men’s inherent strength impacted how judges listened to men and women playing instruments that appear to rely on strength, like the strings or lifting and holding some of the larger instruments. Perhaps their bias was related to beliefs about the lung capacities of men and women for wind instruments. Regardless, when auditions were made blind, judges stopped hearing gender, and they just listened to and judged the music. Some orchestras have added carpeted walkways for musicians to use walking to their auditioning chair behind the screen to ensure that the sound their shoes make does not impact the judges. In others, musicians are prohibited from even clearing their throat during an audition in case the timbre of the musician’s voice can be assumed by the noise, potentially causing judges to guess at the gender. This is just one way gender bias in hiring has been discovered.
It is connected to major institutional practices that might feel “fair.” But when we subject them to sociological research and examination, they’re often less fair than they might feel. To consider a separate example, think of the evaluation forms you fill out at the end of every college course for your professors (or the ones you will fill out if this is your first semester). Students routinely evaluate their professors in college. And the “scores” professors receive from students in their classes are sometimes used in decisions regarding hiring, firing, and promotions and awards. But are student evaluations biased? Do you assess the women and men who teach you in college in different ways? Might you do this even if you are unaware? Research has shown that you do, and women are much more likely to receive lower student evaluations when compared with the men they work alongside. Women of color are particularly likely be evaluated more negatively, damaging their chances for tenure and promotion (see Lopez and Johnson 2014). Sociologists do not think this means that men are naturally more gifted teachers—rather, it is an indication that women and men are assessed differently, and this difference works to men’s collective advantage and to women’s collective disadvantage (Arbuckle and Williams 2003; Johnson, et al. 2008).
Women receive lower scores in experimental and real-world research on student evaluations. Is this because students perceive the occupation (“professor”) as having a gendered component that they might believe men are more naturally likely to possess? Certainly, professors are likely to be depicted as older, white men in movies. But students don’t just give women professors lower scores on average on student evaluations. They also highlight different qualities in women compared with men on their evaluations. Relying on data from the online site RateMyProfessor.com that allows students to anonymously and publicly post reviews of their professors, historian and data visualization expert Ben Schmidt produced a tool that shows how often students use specific words in their reviews of men and women professors by field of study (see FIGURE 9.3).
On RateMyProfessor.com, students are much more likely to use the word genius when they are referring to men teaching them (and music professors are more likely to receive the compliment than are professors of criminal justice). Men are also more likely to be reviewed as “funny” than are women. Women professors, on the other hand, are much more likely to be described by students as “disorganized” than are men on the site (and anthropologists are much more likely to receive the critique than are mathematicians).
Two bar graphs show the frequency with which the words “disorganized” and “genius,” respectively, are used per million words of text when describing men and women professors of various disciplines.
The data shown by the bar graphs are approximately as follows:
Uses of “disorganized” per million words of text for men and women, respectively, are as follows:
Education: 80, 159
Anthropology: 100, 157
Languages: 98, 138
Philosophy: 50, 108
Physics: 60, 102
Engineering: 68, 100
Science: 61, 100
History: 48, 98
Biology: 70, 98
Political Science: 60, 98
Sociology: 57, 92
Health Science: 57, 92
Psychology: 58, 90
Business: 44, 83
Humanities: 50, 82
Computer Science: 68, 81
English: 44, 79
Fine Arts: 39, 79
Communication: 50, 78
Music: 50, 70
Chemistry: 60, 70
Criminal Justice: 29, 61
Economics: 50, 58
Accounting: 39, 50
Mathematics: 40, 38
Uses of “genius” per million words of text for men and women, respectively, are as follows:
Music: 83, 32
Physics: 65, 36
Mathematics: 57, 23
Philosophy: 55, 25
Chemistry: 51, 20
Engineering: 49, 13
Fine Arts: 48, 16
Computer Science: 42, 13
Political Science: 42, 16
English: 39, 10
Humanities: 39, 10.5
Science: 39, 17
Economics: 38, 18
Anthropology: 37, 11
History: 30, 8
Biology: 37, 13
Psychology: 30, 6
Languages: 23, 5
Accounting: 22, 10
Communication: 21, 4
Business: 20, 6
Sociology: 20, 5.5
Health Science: 19, 3
Education: 18, 6
Criminal Justice: 16, 5
What we have learned about orchestras and student evaluations is equally true in many other realms of social life and across virtually any social institution you can imagine. We are all affected by gender bias, whether we realize it or not. And sociological research on gender bias is important because it is a powerful illustration of just how embedded gender inequality is in our society and how much it affects our lives. Next we briefly consider gender in a few major social institutions.