Intergender Distancing Behavior: A Field Experiment

This study investigated intergender distancing behavior in a naturalistic interpersonal situation. During four evenings, the 1,942 people who passed through the lobby of a College of Continuing Education building were unobtrusively observed as they chose to approach or not approach male or female confederates standing next to identical or different signs, handing out demographic questionnaires. Actual participants in this study were the 526 persons (286 women and 240 men) who did approach confederate stimulus persons. During two evening sessions young men and women (in their ear ly twenties) were stimulus persons and during two other evening sessions older men and women (in their mid-forties) were stimulus persons. Passers-by who took questionnaires could choose to return or not return them to a collection box. It was predicted that men would approach female stimulus persons less often than they would approach similarly aged male confederates handing out material; that this difference in men's gender-based approach would be greater when the stimulus persons were apparently handing out different material than when they were handing out the same material; and that this difference in men 's gender-based approach would be greater when the stimulus persons were in their forties than in their twenties. It was also predicted that women would not differ significantly in their approach of young or older male and female stimulus persons under any of the conditions. Findings did not support these hypotheses. Contrary to prediction , men did not base their approach of young or older stimulus persons on gender in any of the conditions . Women consistently approached young and older women significantly more often than they approached similarly aged men. An interpretation of these findings which proposes that social norms and other situational variables are possible predictors of intergender distancing is

more often from one gender than from the other, they may be said to demonstrate sexist discrimination .
prejudice against the female sex [and], 2. any arbitrary stereotyping of males and females on the basis of their gender."  define sexism as a belief that the separate and inferior status of women is natural and right.
In both popular culture and professional literature the term sexism is traditionally used to describe negative reactions to women but not to men.   or on the cognitive component (e .g ., Martin, 1984), i .e., on prejudice or stereotypes. The body of literature focusing on the behavioral component of sexism may be said to include examinations of sexual aggression (e .g ., Malamuth, 1986;; sexual harassment (e .g., Tangri, Burt, & Johnson, 1982); incestuous assault (e.g., ; wife battering (e.g ., Giles-Sims, 1983); and institutional discrimination (e.g., Clayton, Baird, & Levinson, 1984).
Although distancing behaviors have long been investigated by social psychologists interested in status, power, and intergroup relations (e .g ., Allport, 1954 ;, few researchers have examined interpersonal avoidance and distancing behaviors as examples of sexist discrimination in face-to-face situations . More general research pertaining to distancing behaviors has found, for example, that people who want to seem friendly choose smaller distances than those who do not want to seem friendly (Patterson & Sechrest, 1970); friends prefer to stand closer together than do strangers (Ashton, Shaw & Worsham, 1980); and people who are sexually attracted to each other stand close together (Allgeier & Byrne, 1973) . In a review of the literature , Evans and Howard (1973) concluded that "the preponderance of data suggest that persons who are friendly with each other or wish to communicate a positive affect will tend to interact at smaller distances than those who are not friendly" (pp . 336 f.).
Some social-psychological studies of racism have examined distancing as manifestations of racist behavior. One such study (Ward, Zanna, & Cooper , 1974) found that White male participants interviewing Black and White confederates sat further away from Blacks than from Whites, · made more verbal errors when speaking to Blacks than to Whites, and terminated the interviews of Blacks sooner than the interviews of Whites. The investigators interpreted these behaviors as racist discrimination. Crosby, Bromley, and Saxe ( 1980) reviewed several unobtrusive studies of racial discrimination and found that verbal self-reports of prejudice did not always predict observed discrimination.
In most of the studies they reviewed, participants exhibited racism in overt behavior more than they did in self-report measures . This pattern of findings suggests that unobtrusive observations of discrimination behavior are more reliable measures o f racism than self-reported altitude s or beliefs.
Investigations of interpersonal distancing behaviors between men and women have reported conflicting · tesults.  found that higher status persons in general (including men vis a vis women) are likely to intrude on the personal space of subordinates but that subordinates do not intrude on the personal space of persons in superior positions .  review of the literature on nonverbal behaviors , however , concluded that there is insufficient evidence to support Henley's generalization.
In a laboratory study ,  found that men tended to distance themselves more from a previously unacquainted partner of the other gender than they did from a partner of the same gender whereas women behaved similarly toward partners of both genders . While working on a neutral task , men were observed to make negative statements about and turn away from their partners significantly more often if their partner was a woman than if their partner was a man . Men also accepted advice significantly less often from female partners than from male partners . Women who participated in Lott's study did not exhibit any gender-related distancing behaviors, and self-report measures of attitudes did not suggest that men or women would respond differently to same-and other-gender partners.
In a second study, Lott (in press) examined the behaviors of prime-time television characters and found that a sample of male TV characters were reported by trained student observers to distance themselves from female characters significantly more than from male characters. Female TV characters were found not to exhibit such differential behavior.
In a third study, Lott, Lott, & Fernald (1989) examined the behavioral intentions of men and women using a Photo Choice had been predicted, men approached women more often than they did men .
Although this finding was contrary to the hypothesis, men's differential response to women and men was greatest, as had been predicted, in the condition in which they could choose between two different signs, and least in the condition where a man or woman confederate was handing out surveys alone.
No effects were found attributable to the signs or to left-right positions of the confederates.
In order to neutralize the appearance of the confederates in the above study, they dressed in dark-colored clothing and coats, wore no make-up or jewelry , and were instructed to keep their facial expressions bland. Despite these efforts , the men passers-by may have approached the women confederates more often than the men confederates because of sexual cues. Both passers-by and confederates were young persons in their twenties , and the study was conducted on a college campus . (1) men will approach female confederates (stimulus persons) handing out questionnaires less often than they will approach similarly aged male confederates (stimulus persons) handing out material; (2) this difference in men's approach to male and female stimulus persons will be greater when the stimulus persons are apparently handing out different material than when they are handing out the same material; (3) this difference in men's approach to male and female stimulus persons will be greater when the · stimulus persons are in their forties than in their twenties; and (4)

Results
This study utilized a design in which the dependent variable was number of approaches and the independent variables were gender of the persons doing the approaching, gender of the stimulus persons , age of the stimulus persons, and masked or unmasked choice conditions. To analyze differences in frequencies, Chi-Square is appropriate .
The first predictions tested were that under sexually neutral conditions, men are more likely to approach unacquainted men than similarly aged unacquainted women whereas women do not respond differently on this dimension to women and men. Other predictions were that this differential m men 's behavior is greater under condition s in which choice between approaching a man or a woman can be masked as choice between other stimuli , and that men will be more likely to distance from women and approach men if the choice is between older rather than between younger persons.
To test these hypotheses, separate 1 X2 Chi-Squares were obtained indep . endently for men and women participants to analyze differences m the number of men and women who approached young male and female stimulus persons in each of the two conditions (unmasked choice and masked choice ) and the numbers who approached older male and female stimulus persons m both the unmasked choice and masked choice conditions. Table 3 presents the   frequencies  of men's approach  behaviors  and the related Chi-squares  and   Table 4 presents this information for the women participants.

Analysis of Men's Approach Behaviors
The prediction that men will approach male stimulus persons more often than similarly aged females was not upheld . As can be seen in Table 3  The data a lso failed to support the prediction that men 's differential approach to women than Lo men . would be greater in the masked choice than unmasked choice condition . For both young and older stimulus persons , no significant differences were found in men's approaches toward males and females when the stimulus persons appeared · to be handing out the , same material or when stimulus persons were handing out different material.
Again , contrary to prediction , the age of stimulus persons did not influence men's behavior. An overall Chi -square of men's approach behaviors toward male and  and women are presented graphically in Figure 1.

Analysis of Women's Approach Behavior s
Women's approach behavior was predicted not to be influenced by the gender of either the young or older stimulus persons in either the unmasked or masked choice conditions. As can be seen in   postcards were returned, 28 by women and 23 by men. Table 5 shows the number of postcard questionnaires returned as simple frequencies and as percentage of postcards taken under the different conditions . Both men and women returned the same percent (9 .6%) of cards taken from all stimulus persons . Both women and men returned proportiona tely more cards taken from women than from men; women returned 11.3 % and men returned 11 .4 % of cards taken from women whereas women returned 7.9 % and men returned 7 .6% of cards taken from men .
Both men and women returned proportionately more cards taken from older stimulus persons than from young stimulus persons ; men returned 12.4 % and women returned 11.5% of cards taken from older stimulus persons whereas men returned 6.5% and women returned 7.7% of cards taken from young stimulus persons.

Discussion
In this study of intergender distancing behaviors , it was expected that men would distance themselves from women but that women would not distance themselves from men during face -to-face interpersonal interactions between strangers m a naturalistic setting. These expectations were based upon an analysis of sexism  in which sexist discrimination in faceto-face situations is defined by behaviors that achieve distance from women.
Previous empirical studies in a variety of settings have supported Lott 's model Lott , in press;Lott, Lott, & Fernald , 1989).
In the present study contradictory findings were obtained . Women consistently distanced themselves from s imilarly aged young or older men and they did so even when their choice could not be concealed as a choice between two different questionnaires. Men, on the other hand, did not base their approach behaviors on the gender of the stimulus persons; they did not approach women less (or more) than men.  Bagozzi, 197 8;Bandura, 1986 as the man who had exposed himself. The article advised women to be alert to the possibility of sexual assault, to employ the buddy system when using the rest rooms, and to allow the new security guards-to walk them to their cars . It seems likely that a climate of fear existed for women in this research setting.
Such a climate may have motivated women to distance themselves from men.
In neither the pilot study nor in  laboratory study were there any such signs, articles, or security guards present.
As has been previously noted, the term sexism typically refers to men's