Categories and Sexuality.

Over the past several months, I have been intrigued by news stories and discussions that were on diverse topics but that all had in common the same underlying theme: categories--how to delineate them, how to think about them.

Consider an example related to race: California's Proposition 209, the law prohibiting state agencies from using gender or race in decisions related to education, hiring, or awarding contracts (Schmidt, 1997). In effect, it bans affirmative action. I had grown up with the idea that political conservatives evaluated and treated people differently according to their race, whereas political liberals treated people equally regardless of their race. But now, in arguments over affirmative action, political liberals (e.g., Democrats, the American Civil Liberties Union) are arguing that we should take race into account in decisions on admitting students to universities, hiring, and awarding government contracts. Conversely, political conservatives (e.g., Republicans) are arguing that we should treat people equally regardless of their race, which would take the form of abolishing affirmative action programs. In a sense, this argument is about categories--in this case, racial categories. At issue is how to treat people in different categories or, more basically, whether to emphasize or deemphasize categories.

Consider an example related to sexual orientation: I was at a Saturday morning meeting of a group called the Freedom Coalition, a Lawrence, Kansas, group that works for civil rights for people of all sexual orientations. After the meeting, several of us were discussing the categories lesbian, gay, bisexual, and queer, and the increasing visibility of self-labeled bisexuals in the queer-rights movement.

Some people regarded bisexuals as less threatening to the status quo than lesbians or gays. They said that the category bisexuals includes women who are attracted to men and men who are attracted to women, which makes them less threatening to the general public than lesbians, for example, because lesbians are not attracted to men and in that way are more threatening to men. So, they saw bisexuality as relatively nonthreatening to the status quo.

But this discussion also reminded me of a book I was reading, Bi Any Other Name: Bisexual People Speak Out, edited by Loraine Hutchins and Lani Kaahumanu (1991). Hutchins and Kaahumanu wrote of the revolutionary, disruptive possibilities of bisexuality:

   Heterosexuality needs homosexuality, to be reassured that it is different.
   It also needs the illusion of dichotomy between the orientations to
   maintain the idea of a fence, a fence that has a right (normal, good) and a
   wrong (abnormal, evil) side to be on, or fall from. To the extent that we
   collaborate in seeing homosexuality as an opposite polarity (not part of a
   diverse range of human sexuality), we perpetuate this unhealthy,
   unrealistic, hierarchical dichotomy. (p. xxii)

So, is the category bisexuality less or more threatening to the status quo than is homosexuality? However one answers this question, categories are again at issue.

Consider an example, constantly in the news, related to gender: When thinking about categories, I recalled when politicians took up the issue of the number of sexes, of all things. Biologist Anne Fausto-Sterling (1993) wrote an article, "The Five Sexes: Why Male and Female Are Not Enough," in which she conceptualized the categories of two totally distinct sexes as social or medical constructions. She described people born with both an ovary and a testis, whom she labeled as true hermaphrodites or herms. She also described people with ovaries and some aspects of male genitalia but no testes, whom she called female pseudohermaphrodites or ferms, and people with testes and some aspects of female genitalia but no ovaries, whom she called male pseudohermaphrodites or merms. So, with females, males, herms, ferms, and merms, Fausto-Sterling (1993) conceptualized five sexes. That idea seemed upsetting to right-wing politicians, who distorted Fausto-Sterling's (1993) position and took it up as a campaign issue. The Advocate quoted presidential candidate Pat Buchanan as saying, "They say there aren't two sexes, there are five genders.... I tell you this: God created man and woman--I don't care what Bella Abzug says" (Agenda, 1995, p. 11; see Fausto-Sterling, 2000, for more information). So we have another controversy about categories.

In this talk, I want to raise some issues and questions regarding categories, especially as they relate to sexuality. I'll discuss our tendency to reify categories after we create them and our tendency to exaggerate the differences between the categories that we create. I'll then discuss how confounds can make differences between categories difficult to interpret--a fact easily forgotten. Finally, I'll discuss problems associated with focusing on the differences between categories, as well as problems associated with ignoring the differences between categories.

CATEGORIES ARE SOCIALLY CONSTRUCTED

What is reality? I don't know. This is one of those questions we discussed in my Problems of Philosophy class when I was an undergraduate. It's been debated for thousands of years, and I'm not going to come up with the definitive answer tonight. But let me tell you how I find it useful to think about reality. I find it useful to believe that there is an objective world, with energy and molecules flying around in various patterns. However, I think it's important to keep in mind that we as humans have categorized and labeled these flying molecules and that there are many ways in which we could categorize and label the world.

Once we have made up categories or concepts, however, it is easy to reify them--that is, it is easy to treat them as real and universal and to forget that we made them up. It is easy to treat the category as having a real, true, invariant meaning and to forget that it is a word that we made up. In fact, there are no real, objective, universal categories. There is no "Golden Dictionary in the Sky" to tell us the true, real definitions of words. Let me give a few examples.

Gender

When I was a little kid, I knew what a woman was and what a man was, but I don't know now. In the 1950s, I knew that women were what my mom and grandma were and that men were what my dad and grandpa were. I knew that women had long hair and that men had short hair. I knew that men went to work and that women stayed home. I learned that boys had penises and that girls didn't.

Then in the 1960s, my mom got a job, but she was still a woman, and a lot of men let their hair grow long, but they were still men--so jobs and hair length do not necessarily distinguish women from men.

Then much later I learned that sometimes sex organs are ambiguous, appearing to be a large clitoris or a small penis--so that sex organs do not necessarily distinguish women from men. I learned that external genitals, internal sex organs, and chromosomes do not always match; someone can have XX chromosomes with genitals that look like a penis rather than a clitoris. I learned that some people seem clearly to be biologically one sex but have the gender identity of the other sex, or perhaps no gender identity at all. I learned that there are no personality traits that clearly distinguish women and men. I learned that even during the 1950s when I thought that women did not go out to work, most women of color had been going out to work.

So, even though I have a .5 appointment in Women's Studies, and even though I have taught Psychology of Women every semester since 1980, I don't even know what women are. My point here is that there is no real or universal definition.

Sexual Orientation

A similar line of thought applies to sexual orientation. In my Psychology of Women class, I show my students an excellent video called What Sex Am I? (Grant, Feury, Justice, & Yarrow, 1984) about transsexuals and transvestites. The video depicts two couples, one that, prior to surgery, appeared to be a heterosexual couple, and one that, prior to surgery, appeared to be a lesbian couple. In each couple, one of the partners was transsexual and underwent sex reassignment surgery; both couples stayed together and continued to be sexual with each other. But the previously heterosexual couple was now two females having sex, and the previously homosexual couple was now a female and a male having sex. Sometimes a student asks whether the members of the couple are heterosexual, homosexual, or what. I tell them that labeling someone as "a homosexual" is relatively new; that label was first used during the last decades of the 19th century (Bohan, 1996, p. 17; Weeks, 1989, p. 102). After watching the video, the students know the situations of the people in the video. What would it add to give them labels like heterosexual or homosexual, as if these were real categories, as if they somehow exist inside of people rather than being labels made up to describe people, as if everyone has to fit into one of these categories, and, if they do not, it reflects an aberration of the person rather than the arbitrariness of definitions?

At a recent Western Region SSSS conference, we heard a fascinating talk by Joann Loulan (1997), author of several books about lesbian sexuality. She discussed being in a sexual relationship with a man but still identifying as a lesbian, not as a bisexual. Similarly, Paula Rust (1995), in her book Bisexuality and the Challenge to Lesbian Politics, discussed lesbians who have sex with men.

What these examples illustrate is that homosexual and heterosexual are socially constructed categories. There are no objective definitions of these words; there is no "Golden Dictionary in the Sky" that contains the real definitions. These are words, categories, that we made up.

Rape

So far, I have talked about categories of people--women and men, homosexuals, heterosexuals, and bisexuals. Similar issues pertain to other categories, such as categories of acts. I will use one of my areas of research, rape, as an example here. Rape is a term we have all heard, and we probably have ideas or images about what it is, but in trying to define it, it becomes clear that there is no clear definition (Muehlenhard, Powch, Phelps, & Giusti, 1992). There are different types of definitions. There are legal definitions: There are different definitions in each of the 50 U.S. states, on federal land, and in different countries. There are researchers' definitions, and different researchers use different definitions. There are political definitions used by political activists who want to make a point about gender relations, power, and coercion in this society. There are implicit definitions, held by laypersons but not explicitly defined.

These definitions differ in numerous ways. They vary with respect to who is covered. Must victims be female and perpetrators be male, or do definitions allow for male victims and female perpetrators? Is an act excluded from the category rape if the people involved are married to each other? Can a husband rape his wife? Can a wife rape her husband? In old legal definitions, a man's forcing another man's wife to have sex was considered rape, but a man's forcing his own wife to have sex was not. Currently, in the U.S., legal definitions do cover marital rape, at least under certain circumstances. Many laypersons' implicit definitions of rape would exclude such acts, however.

Different definitions of rape vary with respect to what type of coercion counts as rape. Physical force only? How much physical force? What about explicit threats of physical force? What about implicit threats of physical force? And who gets to say what was being implied? What about verbal coercion? What about economic coercion?

Definitions of rape also vary with respect to what sexual behaviors are covered. Some definitions are drawn narrowly, covering only penile-vaginal intercourse. Other definitions also cover anal intercourse and fellatio. Some people, however, consider these definitions too narrow and phallocentric. They argue that what counts as rape is linked to what counts as sex and that in our culture we often assume that "real" sex requires a penis. The idea that behavior between two women could not constitute real rape may reflect the idea that behavior between two women could not constitute real sex. What behaviors count as real sex? (I've been President of SSSS for almost a year now, and I still don't know what sex is.)

But for rape, and for sex, I would argue that there is no true or universal definition. Language is made up by people; meanings of terms are socially negotiated and change over time. People with more power have more opportunity to influence these definitions than people with less power. Again, there is no "Golden Dictionary in the Sky" to tell us the real definition of rape or the real definition of sex. It makes no sense to ask what these words really mean. It does make sense to ask how most people define these words, how different people define these words differently, and how legislators and judges have defined these terms in legislation and case law. It also makes sense to ask about the political implications of various definitions--who benefits and who loses from various definitions. Using rape as an example, it does not make sense to ask, "What is rape, really?" It does make sense, however, to ask how different people define rape. It makes sense to ask about the consequences of adopting different definitions. So, if legislators define a sexual act as rape only if it is obtained using extreme physical force and only if it occurs between people who are not married to each other, who wins? Who loses? Who wins and loses if we conceptualize rape as gender neutral? I'll return to these issues later.

EXAGGERATED DIFFERENCES

How we think and talk about categories often has two effects that work together to produce a misleading picture: exaggerating differences between groups and ignoring differences within groups.

Gender

Take gender as an example. People often talk as if women and men are very different--as if women and men think differently, behave differently, and should be treated differently. This approach emphasizes gender differences, acknowledging no overlap between the groups. It also implies that all men are alike and all women are alike.

Here's an example from the popular media: Recently, the television show Dateline NBC conducted a poll related to sexual harassment in the workplace. One question asked whether it is a mistake for a man to keep a photo on his desk of his wife or girlfriend in a bikini. The poll revealed that 53% of the women said yes, and 47% said no; 48% of the men said yes, and 52% said no (NBC News, 1997). These results hardly reflect an overwhelming sex difference, but they were summarized on Dateline NBC as "Women said mistake, men said OK" (NBC News, 1997, p. 20).

This summary--"Women said mistake, men said OK"--exaggerated the difference between men and women, which was really very small: 53% compared with 48%. It also minimized the differences within each gender, implying that all women said it was a mistake and that all men said it was OK, thereby ignoring the 47% of the women who said that it was OK and the 48% of the men who said that it was a mistake.

The Role of Statistical Significance

One thing that perpetuates this problem is researchers' reliance on statistical significance. When investigating groups, we typically evaluate group differences as being either significant or nonsignificant.

If a study reveals that women and men differ significantly, all this means is that it is unlikely that researchers would have gotten a difference as large as they did "given the hypothesis that chance (i.e., random sampling variation) caused the results" (Becker, 1994, p. 217). In other words, the results are unlikely if the null hypothesis is true. A significant difference does not tell us how different the groups are. The size of this difference can be referred to as effect size.

Significance test statistics, such as F, t, or [chi square], can be conceptualized as the product of the effect size estimate times sample size (Rosenthal, 1994). This means that any nonzero effect size--however small--will reach statistical significance if the sample size is large enough. So, a significant difference between two groups can be statistically significant but trivial.

In an interesting article in American Psychologist, Sheldon Lachman (1993) called the term statistically significant difference "unfortunate" and "misleading" (p. 1093). He went on to write,

   many of the statistically significant differences reported in the research
   literature are of absolutely no significance. They are of no practical
   significance: they are of no theoretical significance.... Nevertheless, one
   of the major implications of the word significant is "meaningful." (p.
   1093)

He suggested replacing the term "statistically significant difference" with the term "probable nonchance difference" or "probable nonrandom difference" (p. 1093). That's all that a statistically significant difference means. Nevertheless, significant differences such as gender differences are often discussed as if they mean that the categories being compared, such as women and men, are totally different with no overlap between them.

Furthermore, a significant difference between two groups is often discussed as if there were total uniformity within each group. A significant gender difference is often summarized as "men do this; women do that," as if all men do this, and all women do that. Such generalizations are inaccurate; this is not what a statistically significant difference means.

I find such generalizations exasperating. I often hear or read, "Men do this; women do that," and I think, "But wait, I don't do that. I DON'T DO THAT!" To be accurate, they would at the very least have to add a footnote, "Men do this; women do that ... except for Charlene Muehlenhard"! Maybe you would want your name included in this footnote too!

Indulge me in a quick illustration from analysis of variance. Let's say you are analyzing a dependent variable, and you assume that the variable of interest is predicted by four independent variables (which, in the grand scheme of things, is still a pretty simplistic model). According to the assumptions undergirding analysis of variance, the model for any one person's score ([Y.sub.ijkmn]) would be the following:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where [Mu] is the population mean, and [[Alpha].sub.j], [[Beta].sub.k], [[Gamma].sub.m], and [[Delta].sub.n] are main effects (based on Myers, 1972, p. 123). There are 6 two-way interactions and 4 three-way interactions, a four-way interaction, and error. If alpha is the gender effect, then if there is a significant gender difference, all this gender difference means is that this one term, alpha, is probably not equal to zero (i.e., probably [[Alpha].sub.j], [is not equal to] 0). That is, in the equation above, if there is a significant gender difference, this means that one term, [[Alpha].sub.j], is probably not equal to 0, but there are still many other influences besides gender that can affect a person's score. This highlights the folly of saying "men do this; women do that." This simplistic statement ignores all the other effects on a person's behavior besides gender.

A simpler way to view this is to look at a diagram. Figure 1 illustrates a hypothetical distribution that may yield a significant gender difference. As you can see, there is a lot of overlap between women's and men's scores, and there is a lot of variation among women's scores and a lot of variation among men's scores. Saying "Women and men are just different; men do this; women do that" is almost never accurate and is almost always misleading. Whenever you encounter such statement, question it!

[Figure 1 ILLUSTRATION OMITTED]

CONFOUNDS

Even if there is a significant difference between two groups, it is not always obvious what this means. In doing research, the clearest way to isolate the effect of a variable is to assign people randomly to different levels of that variable--that is, to assign people at random to one of two or more groups and then to apply different experimental manipulations to each group. In this case, any differences between the groups should be attributable to the experimental manipulation. If instead you use existing groups, you can never really know all the ways in which the groups differ, and you can never really know what is responsible for these differences.

It could be a selection bias: Maybe group members self-selected into these different groups because they differed initially in some important way. It could be that the groups have different histories: Maybe the people in different groups were similar initially but were subsequently treated differently, so that this differential treatment, rather than group membership per se, has affected them in important ways. Selection bias and history are only two of the numerous possible explanations for a significant difference. Thus, unless individuals are randomly assigned to groups, it is difficult to interpret a significant difference.

Gender

The difficulty with doing research on many topics of interest to us is that random assignment is impossible. For example, researchers cannot randomly assign people to be women or men. Therefore, even if researchers find a significant gender difference, it is difficult to isolate the source of the difference. Perhaps if we could control for certain variables, the gender difference would disappear.

Consider this example: Women quit their jobs more often than do men. This might seem to suggest that if you are hiring someone for a job and you want them to stay on the job, it would be better to hire a man than a woman. However, it turns out that people with low-paying, dead-end jobs quit those jobs more often than do people with high-paying, high-level jobs, and gender is confounded with type of job--that is, women are more likely than men to have low-paying, dead-end jobs. When you control for income and level of job, the gender difference disappears (Deckard, 1983; Hyde, 1996). Thus, it is misleading to call this a gender difference--it would be more accurate to call it a crummy job effect--but we would never have known this if researchers had not thought to control for income and level of job.

Because of the difficulties associated with interpreting so-called gender differences, some researchers have questioned whether we should even study gender differences. For example, Roy Baumeister (1988) wrote the following:

   Normally, an experimental manipulation is considered scientifically suspect
   if there are two or three different interpretations of it. For sex there
   are literally dozens. A sex-difference finding is a question, not an
   answer. (p. 1093)

Baumeister went on to write,

   Experimental research ideally uses random assignment of subjects among
   conditions, but experimental subjects are not randomly assigned to sex. One
   might defend the study of sex differences on the basis of its resemblance
   to other individual-difference measures, but those are at least based on
   measures that focus on one single theoretical dimension, such as
   self-esteem or introversion. Sometimes these scales may be criticized as
   having certain ambiguities, but still they are far more precise than male
   versus female. In this case [male vs. female], differences may be
   confounded by social oppression, hormones, physical stature, past exposure
   to role models, personal ideas, flirtation with the experimenter, and so
   on.

   All of these variables are worth studying, but studying sex differences is
   a poor substitute for studying them, just as studying racial differences is
   a poor substitute for studying socioeconomic class.... [A] finding of sex
   difference explains very little. One needs to then determine what it is
   about the sexes that causes that difference. But why the detour? One might
   profitably start by focusing on the true causal variable, bypassing the
   initial demonstration of gender differences. (p. 1094)

So, we could think of gender as a proxy for other variables. Baumeister was suggesting that we study these other variables directly rather than studying gender as a proxy for these other variables.

Abandoning the study of sex differences is certainly a controversial idea. There are theoretical and political arguments against studying sex differences, but there are also theoretical and political arguments for studying sex differences, and there are pitfalls in focusing on similarities. The same can be said about studying differences related to other categories. This leads to my last point.

IMPLICATIONS OF FOCUSING ON SIMILARITIES AND DIFFERENCES

Neither emphasizing similarities between categories nor emphasizing differences between categories is necessarily linked to any political perspective. Emphasizing similarities between categories, or treating people in different categories similarly, may be linked to liberal or conservative perspectives. Conversely, emphasizing differences between categories, or treating people in different categories differently, may also be linked to liberal or conservative perspectives.

Race

My earlier example related to race illustrates this. Even though sometimes treating the races equally is the more liberal approach, currently it is conservatives who argue for equal treatment of the races by banning affirmative action. Neither a similarities perspective nor a differences perspective on race is necessarily linked to any political perspective (see Sleeper, 1997, for an interesting commentary on this paradox).

Gender

The same is true regarding gender differences. There are some feminists, sometimes called radical feminists or cultural feminists, who emphasize the differences between women and men (McFadden, 1984). They may conceptualize women as being closer to nature than men are. They may conceptualize women and men as speaking in different moral voices--some feminists regard women as focusing more on care and interpersonal relationships and men as focusing more on abstract concepts of justice (e.g., Gilligan, 1982).

But these difference-based approaches can be problematic (Hare-Mustin & Marecek, 1990; Muehlenhard, 1998; Tavris, 1992). A focus on differences often leads to an exaggeration of how large such differences are, which promotes stereotypes--often fairly traditional, even sexist stereotypes, such as the stereotype that women are better at relationships and men are more logical. Also, a focus on differences between the genders often implies similarity within each gender. Saying that women value interpersonal relationships and men value abstract justice ignores the many ways in which women differ from each other and men differ from each other. Other sources of difference--race, sexual orientation, background, and so forth, may be neglected.

In contrast, some feminists, sometimes called liberal feminists, emphasize similarities between women and men (McFadden, 1984). They emphasize similarities in women's and men's abilities, and they advocate equal treatment of women and men under the law, in the workplace, and so forth.

But sometimes treating women and men equally has negative consequences, especially if standards are designed to meet the needs of men, and in the name of equality these standards are imposed on women (Hare-Mustin & Marecek, 1990; Muehlenhard, 1998; Tavris, 1992). For example, if women's restrooms have the same number of toilets as men's restrooms, this typically results in long waiting lines for women. More seriously, employers might argue that they don't give paternity leave to men so they should not have to give maternity leave to women, or insurance companies may not pay for birth control pills for anyone. The effects of such "equal" treatment would be more negative for women than for men.

Paradoxically, I would recommend that when approaching a problem involving categories such as gender, we should consider both a similarities and a differences perspective. Both perspectives may have something to contribute (see Kimball, 1995, for an interesting discussion). I'll use my own area of research--sexual aggression--to show how both perspectives can be useful.

Most research on sexual aggression addresses only male perpetrators and only female victims, reflecting a gender-differences perspective. Taking a gender-similarities perspective--by which I mean considering the possibility that both women and men can be perpetrators and that both women and men can be victims--can have positive consequences (Muehlenhard, 1998). For example, asking questions about sexually aggressive women and sexually victimized men challenges sexist stereotypes--the stereotype that women are sexually passive, either resisting or acquiescing to male pursuit, and the stereotype that men always want sex and therefore cannot really be coerced. Furthermore, if we study only female victims of sexually coercive men, we discount the experiences of victims who do not fit this pattern. A gender-similarities approach acknowledges and validates the experiences of men sexually victimized by women, of women sexually victimized by other women, and of men sexually victimized by other men. Also, acknowledging only women sexually victimized by men could be regraded as acknowledging only heterosexuality--as implying that men's sexually victimizing men or women's sexually victimizing women doesn't count because it doesn't involve "real sex." A gender similarities approach that includes coercive same-sex sexual acts implicitly acknowledges that sex between two men or two women is important enough to count. So, a gender-similarities approach is useful in many ways.

A totally gender-neutral approach that does not take gender into account is also problematic, however (Muehlenhard, 1998). Considering gender differences is valuable because in many ways gender plays an important role in sexual aggression. With almost any definition of sexual victimization, more women than men are sexually victimized. For example, in the recent nationwide study summarized in the book, Sex in America, 22% of the women reported that they had been forced to do something sexually, compared with 2% of the men (Michael, Gagnon, Laumann, & Kolata, 1994). There are also large gender differences in fear of rape and the resulting loss of freedom because of this fear. Women report being much more tearful of rape than do men (Poirier, 1996). Women report restricting their behavior more than do men. For example, Gordon and Riger (1989) found that 25.3% of the women, but only 2.9% of the men, said they never walked in their neighborhood alone after dark. Clearly, rape does not have an equal impact on the lives of women and men.

There are also gender differences in the meaning and consequences of sexual aggression for women and men. One of my students, Arthur Satterfield, and I did a study asking women and men about situations in which someone did something sexual to them without their consent (Satterfield & Muehlenhard, 1996). We found several gender differences. Over twice as many women (58%) as men (26%) had been in such a situation. There were also gender differences in reports of emotions and thoughts following the incident: Compared with men, women reported more confusion, anger, self-blame, and depression. Compared with women, men reported more satisfaction and positive feelings and thoughts that the incident had been inconsequential or fun. So, overall, women reported more negative reactions than did men, highlighting the importance of a gender-differences perspective. However, on almost every scale, women's and men's responses showed considerable overlap; on almost every scale, women's and men's responses spanned the entire range of the scale, meaning that some men reported being very distressed and some women reported little or no distress. It would have been a mistake to assume a priori that these experiences would be distressing for all women and trivial for all men or that these experiences did not even happen to men. Thus, both a gender-differences perspective and a gender-similarities perspective have contributed to our understanding of the consequences of sexual coercion.

CONCLUSION

In conclusion, I want to leave you with four take-home messages. I hope that you will keep these points in mind during this meeting and when you return to your research, educational, and clinical work back home.

1. Categories are created by people; thus, there are no objective or universal definitions. Definitions change over time and place, and they have political implications, advantaging some people and disadvantaging others.

2. Beware of exaggerating the differences between categories, and beware of ignoring the differences within categories. A significant difference between groups does not mean that the groups do not overlap; they usually overlap considerably. There is almost always more variability within groups than between groups.

3. Categories are often proxies for other variables. Beware of confounds! When interpreting differences between categories, remember that other variables confounded with these categories could be driving the effect.

4. Neither a similarities perspective nor a differences perspective has totally positive or negative consequences. Consider both!

REFERENCES

Agenda. (1995, October 31). The Advocate, 11.

Baumeister, R. (1988). Should we stop studying sex differences altogether? American Psychologist, 43, 1092-1095.

Becker, B. J. (1994). Combining significance levels. In H. Cooper & L. V. Hedges (Eds.), The handbook of research synthesis (pp. 215-230). New York: Sage.

Bohan, J. S. (1996). Psychology and sexual orientation: Coming to terms. New York: Routledge.

Deckard, B. S. (1983). The women's movement: Political, socioeconomic, and psychological issues (3rd ed.). New York: Harper & Row.

Fausto-Sterling, A. (1993). The five sexes: Why male and female are not enough. The Sciences, 33(2), 20-25.

Fausto-Sterling, A. (2000). Sexing the body: Gender politics and the construction of sexuality. New York: Basic Books.

Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University Press.

Gordon, M. T., & Riger, S. (1989). The female fear. New York: Free Press.

Grant, L. (Director), & Feury, J., Justice, M., & Yarrow, M. B. (Producers). (1984). What sex am I? [video]. (Available from MPI Media Group, 16101 S. 108th Avenue, Orland Park, IL 60467)

Hare-Mustin, R. T., & Marecek, J. (Eds.). (1990). Making a difference: Psychology and the construction of gender. New Haven, CT: Yale University Press.

Hutchins, L., & Kaahumanu, L. (Eds.). (1991). Bi any other name: Bisexual people speak out. Boston: Alyson Publications.

Hyde, J. S. (1996). Half the human experience: The psychology of women (5th ed.). Lexington, MA: D. C. Heath.

Kimball, M. M. (1995). Feminist visions of gender similarities and differences. New York: Harrington Park Press.

Lachman, S. J. (1993). Statistically significant difference or probable nonchance difference. American Psychologist, 48, 1093.

Loulan, J. (1997, April). Exactly what does "lesbian" mean? What is sexual orientation? How does it differ from "desire"? Talk presented at the Western Region conference of the Society for the Scientific Study of Sexuality, Newport Beach, CA.

McFadden, M. (1984). Anatomy of difference: Toward a classification of feminist theory. Women's Studies International Forum, 7, 495-504.

Michael, R. T., Gagnon, J. H., Laumann, E. O., & Kolata, G. (1994). Sex in America: A definitive survey. Boston: Little, Brown.

Muehlenhard, C. L. (1998). The importance and danger of studying sexually aggressive women. In P. B. Anderson & C. Struckman-Johnson (Eds.), Sexually aggressive women: Current perspectives and controversies (pp. 19-48). New York: Guilford.

Muehlenhard, C. L., Powch, I. G., Phelps, J. L., & Giusti, L. M. (1992). Definitions of rape: Scientific and political implications. Journal of Social Issues, 48(1), 23-44.

Myers, J. L. (1972). Fundamentals of experimental design (2nd ed.). Boston: Allyn and Bacon.

NBC News. (1997, October 24). Dateline NBC [transcript]. Livingston, NJ: Burrelle's Information Services.

Poirier, J. (1996). The fear of rape: A comparison with other common fears of female and male college students. Unpublished master's thesis, University of Kansas, Lawrence.

Rosenthal, R. (1994). Parametric measures of effect size. In H. Cooper & L. V. Hedges (Eds.), The handbook of research synthesis (pp. 231-244). New York: Sage.

Rust, P. C. (1995). Bisexuality and the challenge to lesbian politics: Sex, loyalty, and revolution. New York: New York University Press.

Satterfield, A. T., & Muehlenhard, C. L. (1996, November). The role of gender in the meaning of sexual coercion: Women's and men's reactions to their own experiences. Paper presented at the annual meeting of the Society for the Scientific Study of Sexuality, Houston.

Schmidt, P. (1997, September 12). Supreme Court refuses to stay Proposition 209: California measure to bar affirmative action by state agencies will remain in place. The Chronicle of Higher Education, 44(3), p. A42.

Sleeper, J. (1997). Liberal racism. New York: Viking.

Tavris, C. (1992). The mismeasure of women. New York: Simon & Schuster.

Weeks, J. (1989). Sex, politics and society: The regulation of sexuality since 1800 (2nd ed.). London: Longman.

Manuscript accepted December 8, 1999

Charlene L. Muehlenhard University of Kansas

This article was Charlene Muehlenhard's presidential address to the Society for the Scientific Study of Sexuality, presented at the joint meeting of the Society for the Scientific Study of Sexuality and the American Association of Sex Educators, Counselors, and Therapists in Arlington, VA, November 13, 1997.

Address correspondence to Charlene Muehlenhard, Ph.D., Department of Psychology, 426 Fraser Hall, University of Kansas, Lawrence, KS 66045; E-mail: [e-mail].