Physically active women face an intriguing paradox: Western culture emphasizes an ideal, feminine body and demeanor, as opposed to an athletic body and demeanor.  Female athletes live in two cultures: the sport culture that is predominantly masculine and the expansive social culture in which a woman’s femininity is celebrated (Krane, Choi, Baird, Aimar, & Kauer, 2004). The 1972 passage of Title IX (Women’s Sports Foundation, 1998) has spurred women’s participation in professional, intercollegiate, and interscholastic sports to reach an exceptional high (Knight, 2001).  Title IX requires all federally funded programs, such as athletics, to provide equal treatment and opportunity for the participation in professional, intercollegiate,and interscholastic sports (Women’s Sports Foundation, 1998).

  Gender role orientation is the degree to which individuals perceive themselves as masculine or feminine (Lantz & Schrodeder, 1999).  The perceived incompatibility of female and an athlete roles, in addition to the social stigma classifying the athletic female as deviant (Miller & Levy, 1996), causes female athletes to experience a gender-role conflict.  The research cited in this section debates the existence of the gender-role conflict among female athletes in addition to addressing: gender perceptions of society, the feminine body image, and the perceived athletic ability of females as elements of the role conflict.
The role conflict perspective became one of the earliest conceptual frameworks used to identify the degree to which female athletes struggled with these inconsistent societal expectations. In society in general, and in the athletic environment, we tend to
expect athletes in certain sports to display a characteristic body size and shape.  For example we expect distance runners to be thin and Sumo wrestlers to be fat (Sherman & Thompson, 2001).  A series of studies was conducted in the late 1970’s and early 1980’s to identify the degree to which female athletes perceived and experienced role conflict (Allison, 1991; Sage & Loudermilk, 1979).  Despite the popular notion that female athletes experienced constant struggle with defining their femininity, most of the studies came to the same unexpected conclusion: role conflict appeared to be relatively low among female athletes.
An interesting pattern emerges from past research on the female athlete, not in the various research studies’ specific findings on role conflict, but the persistence of researchers to pursue the topic, despite the lack of empirical support for the concept.  Allison (1991, 55) suggests, “this persistence has been characterized by an apparent unwillingness to alter our research questions and refine the conceptualization of role conflict and/or develop different methodological approaches which may prove more useful.”  The reason for persistence deserves further analysis. 

The presentation of a feminine body is problematic for female athletes.  Ideally, athletic women have toned bodies, yet they must avoid excessive, muscular bodies, which may be perceived as masculine.  Krane, et al., (2004) note that successful athletes must be powerful and strong, yet obvious signs of this power are construed negatively as contradicting femininity.  Therefore, female athletes face a contradiction of the desire to be strong and successful but not have oversized muscles.  Women receive the message that thinness equals attractiveness.  In order to be attractive women are expected to be thin, but also muscular.
            Performative theories of gender propose that many women seek to portray the  socially desirable image, appropriate to their gender.  Ussher (1997) suggests that women actively choose from a variety of gender performance “scripts” depending on the situation and context.  Other theorists suggest that gender performance is not entirely voluntary because there are social repercussions for not performing one’s gender “correctly” (Bordo, 1993).  A woman’s “choice” to be feminine is not entirely a choice; if the only women privileged in society are those who conform to ideal femininity, and a woman wants social acceptance, then the only “choice” is to conform with the social ideal (Bordo, 1993). Women have been expected to be passive, dependent, and nurturant, while participation in athletics requires assertion, strength, and independence-characteristics judged as masculine (Hofereck & Hanick, 1985).  Female athletes who perform femininity correctly accrue power and privilege in society; while those female athletes perceived as masculine are labeled social deviants (Krane et al., 2004). 

Certain tasks and activities have traditionally been assigned to men and others to women, that is, they have been considered to be masculine or feminine activities (Koivula, 2001).  These categorizations, largely of social origin, are based on our expectations regarding gender. Our beliefs that gender categories are natural, unambiguous, bipolar, static, and individual also shape our gender role perceptions (Koivula, 2001).  These socially constructed assumptions about gender and gender categories are also formed by historically specific, and culturally specific representations of the social interactions between males and females. 
The characteristics of a specific sport determine the features that underlie the categorizations for a sport as either masculine or feminine.  Sports classified as feminine are those in which it is considered to be suitable for women to participate (Koivula, 2001).  The appropriateness of participation is measured to be whether or not the sport allows women to “remain true to the stereotyped expectations of femininity (such as being graceful and non aggressive) and that allot for beauty and aesthetic pleasure” (Koivula, 2001).  Beauty is an important element in classifying a sport as feminine.  In addition to being a feminizing agent for sport, beauty serves as a way for female athletes to assert their femininity in the midst of proving their athletic capabilities. 
The predominant view in our society defines the roles of female and athlete as incompatible.  Not only has sport long been considered a male domain, it is also male defined by the masculine sex-typed traits and behaviors that are deemed appropriate.  Girls and women who participate in sport have the opportunity to develop and maintain stereotypically masculine qualities such as competitiveness, independence, and assertiveness.  In fact, sport demands that females address certain issues and behaviors that are not a focus of female gender appropriateness.  Thus, the female athlete must step out of her stereotyped gender role, if only temporarily, to experience success in sport (Desertrain & Weiss, 1988).  This claim outlines the gender-schema theory (Bem, 1981).  According to the gender schema theory, “…the phenomenon of sex typing derives in
part, from gender-based schematic processing, from a generalized readiness to process information on the basis of the sex-linked associations that constitute the gender schema (Bem, 1981, p.355).  Thus, sex-typed individuals process incoming information in terms of society’s definitions of masculinity and femininity, whatever these definitions may be (Bem, 1981). 
The present study re-examines the possible existence of a gender role conflict perceived to be problematic for female athletes.  The variable of physical appearance is examined as a factor.  Attractiveness is thought to increase the likelihood of a role conflict for female athletes as they are often forced to differentiate between their roles as a female and as an athlete (Knight, 2001; Koivula, 2001; Krane et al., 2001 & Krane et al., 2004).  Research indicates we expect athletes to look a certain way based on the sport they play (Sherman & Thompson, 2001).  I hypothesize that if the participants perceive the featured female’s physical appearance as deviant from either the feminine gender role expectations (Hofereck & Hanick,1985) or the appearance expectations of their sport (Sherman & Thompson, 2001); participants will rate the female higher in masculine ratings than in feminine ratings.
back to top

Participants

Participants were 32 students (10 males and 22 females) attending a small, Catholic, liberal arts college in New England ranging in age from 18-21.  The participants volunteer for the study to fulfill course extra credit requirements from several introduction to psychology courses, a sophomore basic statistics class, and sections of freshman biology.

Prior to the commencement of the study a consent form, to be signed and returned, was given to participants. After giving consent, each participant was given a packet to complete consisting of a short seven-question demographics questionnaire, four copies of the thirty question short form of the Bem Sex Role Inventory (1978), and three fictional profiles of females. The Bem Sex Role Inventory (1978) an instrument used to measure gender role perceptions. The BSRI contains ten masculine adjectives (e.g. independent, assertive), ten feminine adjectives (e.g. affectionate, tender), and ten neutral items (e.g. conscientious, reliable).  Using a one to seven point scale, participants indicated to what extent each word described them.  A choice of “one” indicates that the participant never or almost believes the adjective describes him/herself.  A choice of  “seven” indicates that the subject always or almost always believes the adjective accurately describes them individually (Bem, 1981).  Masculine and feminine scores are derived by calculating the mean response to each respective category.  Bem (1974) reported high internal consistency and test-retest reliability of the BSRI.  Coefficient alphas computed for masculinity and femininity revealed high reliability (Masculinity alpha = .86; Femininity alpha = .82).  The BSRI test-retest reliability within a sample of 28 males and 28 females demonstrated high reliability over a four week period (Masculinity r = .90; Femininity r = .90) (Holt, 1998).  To explore how physical attractiveness affected people’s gender role perceptions the three fictional female profiles were either accompanied by a headshot photograph of a fit looking female, an overweight female, or no picture was included.  Each packed contained three of the same type of picture.  In addition, each profile placed the featured female into one of three sport conditions; rugby/basketball (masculine sports), tennis/golf (feminine sports), or playing no sport at all.

The experiments were run in a classroom setting.  Participants were first given an informed consent form to sign explaining that they would be asked to complete four brief paper and pencil tests including a self report inventory.  The consent formed also assured participants of the confidentiality of any information gathered.  Participants were given a packet, which randomly assigned them to one of three conditions.  The packets contained either a fictional description of a college aged female accompanied by a photograph of a fit looking female, an overweight female, or there was no picture provided at all. The first page of each packet was a short seven-item questionnaire created to assess generalbackground information as well as perceived athleticism of the participant.  Next each participant was asked to complete the thirty question short form of the BEM Sex Role Inventory (1978), an instrument used to measure gender role perceptions. The BSRI is widely used in psychology because it measures masculine and feminine gender roles separately.  Next participants were asked to read a series of three descriptions portraying a fictional female athlete.    The descriptions were accompanied by a headshot photograph of a fit looking female, an overweight female, or no picture was provided at all. Each packet contained three of the same type of description.  For example, if the first description was accompanied by a photograph of an overweight female the next two photographs would also be of overweight females.  In each description the fictional female was reported to be a junior year business major and in two of the three conditions as having a minor and an internship.  Also each description included fictional account of the featured female’s volunteer experience. Variation in the descriptions was found in the sports they participated in.  One female was described as a varsity rugby captain and
basketball starting point guard, the second described as tennis singles player and golf team member, and the third had no sport affiliation.  The order of the descriptions was varied to ensure random assignment.  After reading a description, participants were asked to complete the thirty question Bem Sex Role Inventory.  The directions indicated the inventory be completed for the female featured in the description.  Thus, participants were asked to use the same seven-point scale, as they had previously, when completing the BSRI for themselves, to indicate the accuracy with which the adjective accurately described the female featured in the description.
At the completion of the packet participants were thanked, provided with a debriefing statement, and a credit slip to account for their participation in this study. 
back to top

A 3 (physical appearance condition: fit, overweight, no picture) X 3 (sport participation: rugby/basketball, tennis/golf, no sport) analysis of variance (ANOVA) was conducted for two dependent variables, the masculine or feminine characteristics means, collected from the short form of the Bem Sex Role Inventory.
Participant ratings of the fictional female description indicated there was no significance for the main effect of physical appearance condition in terms of overall masculine characteristic ratings. (F (2, 87)= 1.237, p = .295). The participants rated
significantly higher masculine characteristic ratings overall for the main effect of sport participation (F (2, 87) = 19.839, p<.01).  No interaction was found for masculine characteristic ratings overall between the physical appearance condition and sport participation (F (4, 87) = .307, p= .873).
Participant ratings’ for the fictional female descriptions of feminine characteristics showed no significance in the main effect of the physical appearance condition (F (2, 87) = .593, p = .555). No significance was determined for feminine characteristic ratings for the main effect of sport participation (F (2, 87) = .593, p = .170). In addition no significance was found for the interaction between the physical appearance condition and sport participation when participants were rating feminine characteristics (F (4, 87) = .921, p = .456).
Next, a Post Hoc Test was performed to look at the significance of the masculine characteristic ratings for each sport condition.  Significance was found for the mean difference between all sport conditions.  Rugby/basketball condition showed a higher rating of masculine characteristics than the golf/tennis condition (p = .011).  Participants rated the rugby/basketball condition as having greater significance of masculine characteristics than the no sport (p = .000), even higher ratings were given to the no sport condition than the golf/tennis condition.  The golf/tennis condition showed significance with a higher rating of masculine characteristics than in the no sport condition (p = .005).  The no sport condition showed low rating of masculine characteristics as significance for the rugby/basketball (p = .000) and golf/tennis (p = .005) conditions were high.

The present study re-examines the possible existence of a gender role conflict among female athletes.  The variable of physical appearance was examined as a perceived factor in creating a role conflict for the female athlete.  The effects of the female athlete’s chosen sport and its perception as more masculine or more feminine was also looked at as a possible factor in the gender role conflict.  The data indicate that physical appearance did not yield more masculine ratings (p = .295), as hypothesized, of the featured female regardless of the sport she played.
The evidence supports the findings of previous research for low gender role conflict among athletes (Sage & Loudermilk, 1979; Anthrop & Allison, 1983; Allison &
Butler, 1984; Boutilier & SanGiovanni, 1983).  Findings from this study supports Sage & Loudermilk’s (1979) third argument suggesting that perhaps the penalties for participation in sport are not as great as they had once been.  The data collected by Anthrop & Allison (1983) indicates that female athletes perceived and experienced little conflict with regard to self-image and/or physical attractiveness; these findings are supported by the present study. 

Despite the insignificant findings evidence stated in research suggests that female athletes face expectations for their physical appearance that male athletes do not have to consider. Krane et. al., (2004) note that successful athletes must be powerful and strong, yet obvious signs of this power are construed negatively as contradicting femininity   butthe sport of participation yielded more masculine ratings.. Ideally, athletic women have toned bodies, yet avoiding excessive muscular bodies that may be perceived as masculine.  In addition, society teaches that in order to be attractive women are expected to be thin, but also muscular (Koivula, 2001).
Another aspect physical appearance dichotomy for male and female athletes to be examined in further research is the differences in male and female uniforms.  As suggested by Krane et al. (2004) female athletes indicated they felt uncomfortable and sexualized in their revealing uniforms. The uniforms attracted unwanted attention from males and drew attention away from their sport (Krane et al., 2004).  An interesting research question may be centered around why some female sports require women to wear skirts and whether or not the skirt is most practical for the athletic event. A suggestion for further research is to examine women’s sport coverage in the media.  One should examine the quantity as well as the quality of the media source’s comments about the female athlete for any support of the gender role conflict.
The results of this study also indicate that participants gave higher masculine ratings to the sport they perceived as the more masculine.  Rugby/basketball showed the greatest significance when compared to golf/tennis (p = .011) or no sport (p = .000).  The golf/tennis condition proved significant over the no sport condition (p = .005). The high masculine ratings could be evidence for Methany’s (1965) appropriatness ratings of sports. Methany (1965) used three categories of sports for identifying their perceived appropriateness for female participation: “not appropriate,” “may be appropriate,” and “wholly appropriate” (p. 51).  She outlined the characteristics of sports that belong in each category and gave examples (Csizma, Wittig, and Schurr, 1988).  “Not appropriate” sports involve subduing an opponent by bodily contact, applying force to heavy objects, projecting the body into or through space over long distances, and face-to-face competition in which some bodily contact may occur.  According to these criteria, the rugby/basketball condition is considered “not appropriate” for females, as reflected in the high masculine ratings.  The “may be appropriate” sports involve moderate distances, moderate weight objects, and the display of strength in controlling bodily movement. “Wholly appropriate” sports involve projecting the body in aesthetically pleasing patterns, using manufactured devices to facilitate bodily movement, applying force through a light implement, overcoming resistance of a light object, and competing where a physical barrier is present.  These last two criteria of appropriateness are also reflected in the results of the present study with golf/tennis significantly higher in masculine ratings than the no sport condition.

The findings also support the gender schema theory, outlined by Bem (1981) According to the gender schema theory, “…the phenomenon of sex typing derives in part, from gender-based schematic processing, from a generalized readiness to process information on the basis of the sex-linked associations that constitute the gender schema (Bem, 1981, p.355).  The theory indicates that sex-typed individuals process incoming information in terms of society’s definitions of masculinity and femininity.  Therefore, additional research is necessary to examine the origin of the significant findings. 
While, the findings of this study do not indicate the existence of a gender role conflict among female athletes, evidence from the study suggests that sport appearance expectations (e.g. a small gymnast, a fat Sumo wrestler) may influence participants’ perceptions of the female athlete as to adherence to gender role expectations.
Further inquiry into the participant’s background, the basis of their gender role perceptions, may be an interesting extension upon this research. Perhaps looking at socializing agents in childhood influence a person’s gender role perceptions throughout life.  Exposure to sports would also be an interesting component to examine as perhaps more exposure to men’s and women’s athletic events would influence perceptions of gender equality in sports.
Because this study was conducted with participants from only one campus, the results should be considered tentative.  Future research on gender role orientation and athletic identity should be conducted at larger and same size universities to investigate generalizability.
Though my study did not provide empirical evidence for the existence of a gender role conflict among female athletes. Each study including my own has used some type of paper and pencil test to evaluate participants perceptions of the gender role conflict, it is possible that these paper and pencil tests are not asking the questions in the correct way perhaps an interview based research style or an open ended questionnaire would produce evidence for the gender role conflict as in Krane, Waldron, Michalenok & Stiles-Shipley (2001).
back to top

The arduous task of completing the Psychology Department’s Senior Thesis has been an emotional roller coaster ride I will never forget.  I laughed when it was 3 a.m. and I was still pouring through research articles, I grew excited when I thought I found significance, and became disappointed when in some cases I did not, I shed a few tears as the sleep deprivation took over and still there was more intro. to write, but nothing will compare to the great sense of accomplishment I have upon handing my thesis in.

I would like to thank those of you who saw my thesis to its completion with me.

Mom and Dad, thank you, for your consoling phone calls, words of wisdom, and unconditional love and support. A phone call home always seemed to make things better.

Tanner, thank you for your good nature and brotherly love that made me smile when the thesis became overwhelming. It is so nice to have you just a walk away.

Patrick, thank you for your love, patience, advice, and understanding. You knew when to give me the extra hug. You patiently read countless pages of data to me and lovingly listened to all my concerns and complaints regarding the thesis. 

Meg & Bianca, my psychology soulmates, without you two to share in practically every moment of this thesis process I do not know how I would have made it through.  Thank you for being such good friends.

To my roommates, thank you for putting up with my late hours, understanding periodic craziness, and supporting me all the way through!

To my Experimental 2 class, congratulations to each one of you for making it through the dreaded senior thesis.  I feel as though along the way I came to know each of you better and our class became a support system for thesis woes. And to all Senior Psychology majors, congratulations on completing your senior thesis!

Last, but not least, I would like to thank my professors.  In my four years here at Saint Anselm College, each of you has had an intrical part in my intellectual development, the fruits of which I hope is portrayed in my senior thesis.  In particular, I would like to thank Professor Finn for being my mentor through this grueling process.  Your time and effort you gave to me is greatly appreciated. 
back to top

back to top

INFORMED CONSENT AND RIGHTS OF RESEARCH PARTICIPANTS IN THE DEPARTMENT OF PSYCHOLOGY AT SAINT ANSELM COLLEGE

Sample Items from the directions for the short form of the Bem Sex Role Inventory (BSRI, 1978)

On the next page, you will find listed a number of personality characteristics.  We would like you to use those characteristics to describe yourself, that is, we would like you to indicate, on a scale from 1 to 7, how true of you each of these characteristics are.  Please do not leave any characteristic unmarked.

Example: sly

Write a 1 if it is never or almost never true that the individual is sly.
Write a 2 if it is usually not true that you believe the individual is sly.
Write a 3 if it is sometimes but infrequently true that the individual is sly.
Write a 4 if it is occasionally true that the individual is sly.
Write a 5 if it is often true that the individual is sly.
Write a 6 if it is usually true that the individual is sly.
Write a 7 if it is always or almost always true that the individual is sly.

Thus, if you feel it is sometimes but infrequently true that you are “sly,” never or almost never true that you are “malicious,” always or almost always true that you are “irresponsible,” and often true that you are “carefree,” then you would rate these characteristics as follows:

Appendix C

Bem Sex Role Inventory

Saint Anselm College Psychology Department

Sex Roles: A Journal of Research

Women's Sports Foundation