Canadian attitudes toward female topless behaviour: a national survey.
Key words: Female toplessness Sexual attitudes National Canadian survey
In 1996, the Ontario court of appeals ruled that women going topless did not imply indecent or sexual behaviour as defined by the Canadian criminal code (R. v. Jacobs, 1996). Similar rulings were made in Saskatchewan in 1998 and in British Columbia in 2000. This chain of events began in 1991 when University of Guelph student Gwen Jacobs, walked in downtown Guelph without a top, thus exposing her breasts. She was arrested and charged with committing an indecent act (DeLonghi, 1992). Jacobs stated that she went topless because it was a hot day and she believed that women should have the same right to go topless as men. She said her behaviour was about empowerment and not just about taking off her top (Canadian Press, 1992). As a result of the Ontario Court of Appeal decision, female toplessness is not considered an indecent act, yet few women go topless in Canada (Murdock, 1996).
The present study used national survey data to document Canadian attitudes toward female topless behaviour in different contexts and to identify demographic variables associated with such attitudes. The literature review that follows provides the background and rationale for the study.
RESEARCH ON FEMALE TOPLESS BEHAVIOUR
Cultural norms regarding body exposure vary widely. In their analysis of anthropological research from 190 different societies, Ford and Beach (1951) found that, in many cultures, women were not required to cover their breasts. They also found no relationship between the importance of breasts in sexual behaviour and requirements for breast concealment.
While culture certainly influences the perception of breasts as sexual, other research has examined whether context may play a role in the definition of female toplessness as sexual within a particular culture. For example, researchers have examined motivations and behavioural scripts for going topless at Mardi Gras (Forsyth, 1992; Redmon, 2002; Shrum & Kilburn, 1996). During the Mardi Gras parade, women reveal their breasts in exchange for beads and other trinkets, a behaviour Shrum and Kilburn (1996) described as a ceremonial exchange of goods. They explain the behaviour using exchange theory and gender roles, with women wanting to receive goods (beads) and men wanting to see women's breasts.
There has been little Canadian research on attitudes toward female topless behaviour. In 1992, the Gallup organization conducted a national poll which asked Canadian adults whether they believed women should be legally allowed to bare their breasts in public (Gallup Canada, 1992). Most Canadians surveyed (62%) were opposed to this, with women and older Canadians being more likely to believe it should not be legal for women to go topless. This survey only referred to toplessness in general rather than in specific situations such as at public beaches. It may be that female topless behaviour would be more acceptable in some situations than others.
Only one study has examined situational context and motivations relating to female topless behaviour among university students (Herold, Corbesi, & Collins, 1994; 1995). In Australia, a country with demographic and social similarities to Canada, it is generally acceptable for women to go topless at public beaches, but not in other contexts. Herold et al. (1994; 1995) found that almost all of the university students (88%) studied believed that women have the fight to go topless when at the beach. Also, almost all of the participants agreed that it was inappropriate for women to be topless outside of the beach context. Members of both genders were similar in their attitudes.
The Australian students who were accepting of toplessness at the beach thought that it gave women a sense of personal freedom and meant greater liberation for women. Most of the students did not define toplessness as sexual in nature, yet a majority believed that topless women were more likely to arouse sexual feelings in men (Herold et al., 1994; 1995). There were significant differences in the attitudes of women who reported not going topless and women who reported going topless. Women who had not gone topless were more likely than those who did to think that a woman's decision to go topless was sexual in nature and reflected decaying morals.
The findings of the Australian study suggest that researchers would be wise to consider context when examining attitudes toward female topless behaviour. Studies of casual sex behaviour on vacation also indicate that context can influence attitudes toward sexual behaviour. For example, students on Spring Break to Florida had more favourable attitudes toward engaging in casual sex while on vacation compared with their attitudes about casual sex while not on vacation (Mewhinney, Herold, & Maticka-Tyndale, 1995; Maticka-Tyndale, Herold, & Mewhinney, 1998).
Although the Australian study provided useful insights into the motivations for topless behaviour, this study is somewhat limited because of the restricted sample which was exclusively composed of university students. We argue that a more comprehensive understanding of attitudes toward female topless behaviour would require the use of broader research samples such as those associated with national surveys. In this vein, the current study utilized a national survey to examine attitudes toward female topless behaviour.
NATIONAL SEXUALITY SURVEYS IN CANADA
In their analysis of sexuality research in Canada, Barrett, King, Levy, Maticka-Tyndale, and McKay (1997) concluded there was a lack of large-scale national survey data on which to draw conclusions about the sexual attitudes and behaviours of Canadian adults and that more comprehensive surveys of the Canadian population were needed. This situation contrasts with that in the United States where several large national surveys have included sexuality questions (Christopher & Sprecher, 2000).
National surveys in Canada that included sexuality questions typically have focused on specific sexual health issues and have been largely conducted by government agencies. For example, the Health Promotion Survey conducted in 1990 by Statistics Canada included only a few questions asking about number of sexual partners and age at first intercourse (Statistics Canada, 1991). Similarly, a few surveys conducted by Health Canada, such as the National Population Health Survey (NPHS) and the 1994 and 1995 Canada Health Monitors (CHM) surveys, have also included some questions about sexual behaviour (Maticka-Tyndale, McKay, & Barrett, 2000). However, these questions focused on health-related issues such as condom use and not on the social or relational aspects of sexuality (Barrett et al., 1997). There would appear to be few contexts in which to gather nationally representative data on Canadian attitudes toward sexuality related policy issues. Public atittudes toward topless behaviour in public represent one such issue.
The one Canadian national study that did examine this issue (Gallup Canada, 1992) did not take into account the degree of correlation among predictor variables in order to determine which of the demographic variables had the most important and significant effects on respondents' attitudes toward toplessness.
In 1998, the Compas survey organization, on behalf of the Sun Newspaper group, conducted a large-scale proportionally representative national survey of the Canadian adult population (Compas Polling, 1998). The Compas survey included more questions on sexuality than had any previous Canadian national survey of adults. The current study used data from that survey to analyze which selected demographic factors best predict Canadian attitudes toward female topless behaviour in different settings.
The first objective of this study was to determine the attitudes of Canadian adults toward female topless behaviour in three different contexts; (1) at public beaches, (2) in public parks, and (3) on city streets. The second objective was to determine the predictive value of several demographic variables (gender, age, education, religiosity, marital status, and region of Canada) in explaining attitudes toward topless behaviour in these three contexts. Based on previous surveys findings, we predicted that the following groups of adults would have less favourable attitudes toward female topless behaviour: (a) women, (b) those who are less educated, (c) older adults, (d) married individuals, (e) those who attend religious services often and (f) those living in provinces other than Quebec.
This study involved the secondary analysis of a national Canadian telephone survey conducted by the Compas polling organization on behalf of the Sun Newspaper Group in 1998. The survey instrument was developed by one of the authors (ESH) in consultation with the survey firm. Preliminary percentage data were published in the Canadian Sun newspapers between October 25 and November 1, 1998.
The Compas survey of the Canadian population included 1479 adults (50% males and 50% females) over the age of 18. Using the 1996 census as a basis of comparison, the sample was proportionally representative of the Canadian population with regard to the demographic variables of gender, age, region, education, and religious affiliation. Over-sampling in Alberta, and of unmarried individuals under the age of 30, was done for the purposes of drawing meaningful conclusions about those particular groups. The data were weighted to reflect the proportions of the Canadian population. Response rate and refusal rate were not reported but were considered to be similar to other surveys conducted in the same manner by national survey organizations.
The predictor variables analyzed in this study included gender, age cohort, education level, marital status, religiosity and region of Canada. Frequency data for each of the predictor variables are presented in Table 1. Gender and region of Canada were recorded by the interviewer at the conclusion of the interview. Region of Canada was determined by area code and the categories were: Atlantic, Quebec, Ontario, Prairies and British Columbia.
Participants' ages were categorized as shown in Table 1. About half of respondents were under the age of 50. The 93 respondents aged 18-19 were not included because the sample size was much smaller than the other age categories. Education was measured by asking respondents, "What was your last year of formal education?" Response categories were (1) post-graduate, (2) university graduate, (3) college or technical school, (4) some university or college, (5) high school graduate, (6) some high school, and (7) elementary school graduate. To ensure that there were sufficiently large numbers in each category for the analysis, education levels were collapsed into the following categories: (1) post-graduate/university graduate, (2) college or some university, (3) high school graduate, and (4) less than high school education (Table 1).
Marital status was determined using the response categories (1) never or not yet married, (2) never married, now living common law, (3) divorced/ separated, now living common law, (4) divorced/ separated, not living common law, (5) married to first spouse, (6) married to second or later spouse, and (7) widowed. To ensure that there were sufficiently large numbers in each category for the analysis, marital status for this analysis was collapsed into the following categories: (1) single (never or not yet married), (2) common law (never married living common law, divorced/separated living common law), and (3) married or ever married (divorced/separated, not living common law, married to first spouse, married to second or later spouse, widowed). Statistical analyses were also done with marital status collapsed in other ways. Using logistic regression, it was determined that marital status was not a significant predictor variable regardless of the method of categorizing the variable.
Religiosity was determined by asking respondents, "Approximately how often do you attend religious services?" Response categories were (1) never, (2) a few times a year, (3) about once a month, (4) two to three times a month, and (5) once a week or more. For the analysis, categories three and four were collapsed into "one to a few times a month" because sample sizes for those categories were too small for comparison.
Attitudes to topless behaviour were measured by the question, "Should it be against the law for women to go topless, (a) on city streets, (b) in public parks, (c) at public beaches, and (d) on private property reserved for nudists?" The responses were coded as (1) yes, (2) no or (3) don't know. The category of "nudist property" was excluded from the analysis because it does not refer to a public place. As noted below, only respondents who gave yes or no replies were included in the statistical analysis. Those who said don't know or who did not offer a reply to a particular question (7.7% to 10.5% of respondents depending on the question) were not included in the statistical analyses.
The data were collected in 1998 by the Compas survey organization (Compas Polling, 1998). Households were selected through computer-assisted telephone technology and only listed telephone numbers were included for selection. Surveys were conducted in either English or French. A bilingual translator was hired to translate the survey and a francophone employed with Compas polling verified the final translation.
Interviewers at Compas introduced themselves by name and indicated to the participants that they were calling on behalf of Compas, the national public opinion research firm. Individuals were asked if they were over the age of 18 and if they were willing to participate in a confidential survey about modern life.
Any questions about the study were referred to the director of Compas polling and a phone number was provided. Individuals were also informed that they could refuse to answer any question. At the conclusion of the survey, participants were thanked and reminded that the survey was confidential and the results would be posted on the company website. Gender, interview language and residing province were recorded at the conclusion of the survey.
Responses to whether or not it should be illegal for women to be topless on city streets, in public parks, or at public beaches, are presented in Table 2. Visual inspection of findings suggests a hierarchy in which female toplessness was most dissapproved (i.e., should be illegal) on city streets (72.4%), less so in public parks (62.1%) and less still at public beaches (48.1%). In other words, 27.6% would not want female toplessness to be illegal on city streets, 37.9% in public parks, and 51.9% at public beaches (Table 2).
To confirm these observations, comparisons were made among the three topless situational questions to determine whether the responses to the questions were significantly different. Using McNemar's [chi square] statistic for nominal non-independent samples, it was found that responses to all three contexts were significantly different from one another. Specifically, respondents were indeed more likely to think it should be illegal for women to go topless on city streets than in public parks, McNemar [chi square] = 205.14 (1),p < .001, and also more likely to think that it should be illegal to go topless in public parks than at public beaches, McNemar [chi square] (1, N = 1421) = 16.59,p <. 001. This hierarchy also applied to males and females although gender differences in this respect, described below, are apparent.
In addition, Spearman correlations were calculated for the three topless contexts (see Table 3). Although the topless contexts shared a significant amount of variance (ranging from 27% to 41%), there was more unshared than shared variance. Accordingly, there were separate analyses for each context. The possibility of combining the three items to form a scale that could be used as an additional criterion was considered but deemed uninformative. As noted in Table 3, Spearman correlations revealed that endorsement of topless behaviour in all three contexts examined was highly correlated (r's ranging from .83 to .88). (The word "endorsement" is used here to reflect the opinion that it should not be illegal.) Therefore, using a combined measure would have resulted in redundant information while not capturing the subtleties associated with the use of the separate criterion variables. Spearman correlations were also calculated for the predictor variables and are presented in Table 4.
All of the analyses for this study were conducted using Binary Logistic Regression, which is the most appropriate method since all of the predictors and the outcome variables were categorical. All predictor variables were appropriately dummy coded, entered into a single block and used to predict respondents' endorsement of topless behaviour across the three contexts examined in separate analyses.
To provide an appropriate comparison point in the logistic regression analysis, the comparison vectors for each predictor variable were coded as the category with the highest percentage of respondents saying "yes" to each topless question. Subsequent analyses included cross-tabulations to determine where any significant differences existed between levels of the categorical predictor variables. The results are presented separately for each of the topless situations, namely, on city streets, in public parks, and at public beaches.
TOPLESSNESS ON CITY STREETS
A test of the full model regarding going topless on city streets, with all six predictors against a constant-only model was statistically reliable, [chi square] (17, N = 1357) = 124.76, p < .001. This indicated that the predictors, as a set, reliably distinguished between those who thought it should and those who thought it should not be legal for women to go topless on city streets. The variance accounted for in acceptance of topless behaviour was modest but significant (13% using the Nagelkerke [R.sup.2] estimate). Table 5 shows regression coefficients, standard errors, Wald statistics and odds ratios for each of the six predictors. According to the Wald criterion, gender (Wald  = 20.25, p < .001), age (Wald  = 9.44, p < .05), religiosity (Wald  = 51.74, p <. 001) and region (Wald  = 15.90, p < .01), reliably predicted acceptance of female topless behaviour on the street.
Significantly more women than men thought it should be illegal for women to go topless on city streets, [chi square] (1, N = 1459) = 37.20,p < .001, V = .16 (see also Table 2). Among respondents of both sexes, those aged 40-49 (67.5%) were the least likely to think it should be illegal and differed significantly from those aged 60 and over (73.2%) and those aged 20-29 (79.9%), [chi square] (4, N = 1366) = 11.77, p < .05, V = .09. Those aged 50-59 (69.4%) and those aged 30-39 (76.6%) did not differ significantly from those aged 40-49.
The significant difference associated with the religious attendance variable showed that those who attended services once a week or more were significantly more likely to think it should be illegal for women to go topless on city streets than those who attended a few times a year (85.1 vs. 72.3%) or those who never attended (57.7%), [chi square] (3, N = 1453) = 68.95, p < .001, V = .22. Those who attended religious services once a month (83.3% disapproval) did not significantly differ in their attitudes from those who attended once a week or more.
Canadians who live in the Prairies were most likely to think it should be illegal for women to go topless on the city streets (78.2%). They differed significantly from people who live in Ontario (70.8%) and those who live in British Columbia (62.2%), who were most accepting of topless behaviour, [chi square] (4, N = 1459) = 21.97, p < .001, V = .12. Those living in Quebec (74.6%) and in the Atlantic provinces (72.5%) did not differ significantly in their attitudes from people living in the Prairies.
TOPLESSNESS IN PUBLIC PARKS
A test of the full model for going topless in public parks, with all six predictors against a constant-only model was statistically reliable, [chi square] (17, N = 1338) = 153.11, p < .001, which indicates that the predictors significantly distinguished between those who thought it should and those who thought it should not be legal for women to go topless in public parks. The amount of variance in acceptance of topless behaviour accounted for was modest but significant (15% using the Nagelkerke [R.sup.2] estimate; see Table 6 for full results). According to the Wald criterion, gender (Wald  = 49.43, p < .001), religiosity (Wald  = 38.62, p <. 001) and education (Wald  = 14.34, p < .01) reliably predicted acceptance of female topless behaviour in public parks.
Significantly more women than men thought it should be illegal for women to go topless in public parks, [chi square] (1, N =1441) = 71.65, p < .001, V = .22 (see also Table 2). Those who attended services once a week or more (78.5%) were significantly more likely to think it should be illegal than those who attended a few times a year (61.0%) or those who never attended (46.7%), [chi square] (3, N = 1435) = 68.01, p < .001, V = .22. Those who attended religious services once a month (70.7%) did not significantly differ in their attitudes compared to those who attended once a week or more.
Those with high school education (66.7%) were more likely to think it should be illegal for women to go topless in public parks than those with university education (54.3%), [chi square] (3, N = 1439) = 12.77, p < .01, V = .09. Those with less than high school education (65.2%) and some college (62.7%) did not differ significantly from those with high school education.
TOPLESSNESS AT PUBLIC BEACHES
Here again, our full model reliably distinguished between those who thought it should and those who thought it should not be legal for women to go topless at public beaches, [chi square] (17, N = 1351) = 193.29, p < .001. As before, the amount of variance in acceptance of topless behaviour accounted for was modest but significant (18% using the Nagelkerke [R.sup.2] estimate). As shown in Table 7, gender (Wald  = 36.29, p < .001), religiosity (Wald  = 62.51, p < .001) education (Wald  = 30.07, p < .001) and region (Wald  = 17.42, p < .01) reliably predicted acceptance of female topless behaviour.
Once again, women more so than men thought it should be illegal for women to go topless at public beaches, [chi square] (1, N = 1454) = 58.76, p < .001, V = .20. (see also Table 6). Those who attended religious services once a week or more (68.1%) were most likely to think it should be illegal when compared with those who never attended (31.8%) or those who attended a few times a year (44.7%), [chi square] (3, N = 1448) = 88.65,p < .001, V = .25. Those who attended one to a few times a month (59.7%) did not differ significantly from those who attended once a week or more.
Those with high school education (55.5%) were most likely to think it should be illegal for women to go topless at public beaches when compared to those with university education (37.5%) and those with some college (47.0%), [chi square] (3, N = 1452) = 25.88, p < .001, V = .13. Those with less than high school education (54.3%) did not differ significantly from those with high school education.
People in the Prairies (55.7%) and Atlantic Provinces (55.7%) were most likely to think it should be illegal for women to go topless at public beaches when compared with those living in Quebec (44.2%), Ontario (46.5%) and British Columbia (39.0%), [chi square] (3, N = 1454) = 21.82, p < .001, V = .12.
The objective of this study was to analyze Canadian attitudes toward the legality of female topless behaviour. Only one previous Canadian national survey had examined this issue (Gallup Canada, 1992) by asking Canadians if they thought it should be legal for women to go topless in public. Research conducted in Australia suggested that context is very important in determining people's attitudes toward women going topless in that toplessness at public beaches was perceived to be far more acceptable than outside the beach context (Herold et al., 1994; 1995).
The findings reported here are consistent with the observation that while about half of respondents in Canada thought female toplessness should be illegal in all three public contexts, there was greater acceptability of this behaviour at public beaches than in public parks or on city streets. The findings clearly demonstrate the importance of including a particular context when asking opinions about topless behaviour.
The second study objective was to determine the role of demographic variables in predicting attitudes toward the legality of female topless behaviour. The demographic variables considered in our study were gender, age, education, marital status, religiosity and region of Canada. In all three contexts considered, the logistic regression model was statistically reliable with variance accounted for by the predictors ranging from 13% to 18% depending on which context was examined. The findings indicate that demographic variables are able to account for a modest but significant proportion of the variance in attitudes toward the legality of female topless behaviour.
Of the six demographic variables only two, namely gender and religiosity, were significant predictors in all three situational contexts. With regard to gender, the findings of this study are consistent with those of previous researchers who have reported that men typically have more liberal attitudes toward sexuality than do women (Baumeister & Tice, 2001; Bibby, 2001; Jenish, 1994; Laumann et al., 1994; Oliver & Hyde, 1993; Smith, 1992). More specifically, for all three topless contexts, females were decidedly more likely than males to respond that topless behaviour should be illegal. Interestingly, even most males were disapproving of female toplessness on the street or in parks. Toplessness was only acceptable for the majority of males in the beach context. This finding is consistent with the 1992 Canadian Gallup Poll survey which reported that males (42%) were more accepting of topless behaviour than were females (22%) (Gallup Canada, 1992).
Milhausen and Herold (1999) suggested that women are more judgmental of women's sexual behaviour than are men. This may explain why women were found to be less accepting of female topless behaviour being legal. Women may feel that if some women go topless then this would put pressure on all women to go topless. There may also be a jealousy element on the part of some women. Additionally, some women in the Australia study were concerned that topless women were more likely to sexually arouse men and that this would lead to an increase in sexual assaults (Herold et al., 1994; 1995). A similar sentiment could be influencing the views of women in our study.
Religiosity was also a consistent predictor of attitudes toward female topless behaviour. For all three topless situations, those who attended religious services once a week or more were significantly more likely to believe it should be illegal for women to go topless in public than those who never or rarely attended. Those who attended about once a month were not found to differ significantly from those who attended more often. However, in examining the percentage distribution of responses, there is a clear trend with opposition to the legality of toplessness increasing as attendance at religious services increases. This trend is consistent with previous research findings that those who attended religious services more often held more conservative sexual attitudes than those who attended services less often or not at all (Laumann et al., 1994; Lefkowitz et al., 2004). Going topless is more likely to be seen as a moral issue by those who are more religious than by those who are non-religious. This group is more likely to adhere to traditional religious teachings about sexuality.
Age was a significant predictor of attitudes toward female topless behaviour only for the street context. A curvilinear relationship was found between age and acceptance of topless behaviour such that those aged 60 and over and those under the age of 40 were least accepting of female topless behaviour and those ages 40 to 59 were most accepting. The finding that the oldest age group was the least accepting of topless behaviour is consistent with previous findings that older people hold more conservative sexual attitudes (Laumann et al., 1994). The middle age group represents the "baby boomers" that came of age in the permissive 1960s and this may explain why they have more tolerant attitudes toward this issue. It should be noted that as this study analyzed age cohorts, differences might be due to cohort effects. It is not clear whether people generally become more conservative as they age or if aging effects are primarily due to the differential socialization of age cohorts, such as with the baby boomers (Rathus, Nevid, Fichner-Rathus, & Herold, 2004).
Education was a significant predictor of attitudes toward female topless behaviour for the park and beach contexts but not for the street context. For each of the former situations, those who had high school education were more likely to believe that female topless behaviour should be illegal than were those with more education. Those with less than high school education did not differ from those with high school education. The results are consistent with previous findings that those with more education hold more liberal attitudes toward sexuality (Laumann et al., 1994), perhaps because being in university facilitates a more liberal perspective regarding sexuality. This may be because in university, one is exposed to a wide diversity of ideas.
There were significant regional differences in attitudes regarding toplessness for both the street and the beach contexts. In particular, those who lived in the Prairie provinces were more likely than people living in British Columbia and Ontario to believe that female topless behaviour should not be legal either on the beach or on the street contexts. For the beach situation, people in the Atlantic region were more opposed to topless behaviour than those in British Columbia. This last finding is consistent with previous national surveys which indicated that people in the Maritimes have more conservative sexual attitudes than Canadians living in other provinces (Clark, 1999; Milne, 1999).
However, unlike previous studies which reported that people in Quebec have the most sexually liberal attitudes (Barrett et al., 1997; Clark, 1999; Milne, 1999) the results of our study reveal that people in British Columbia were the most accepting of topless behaviour being legal for both the beach and street contexts. These results are consistent with the 1992 Gallup Poll finding that people in British Columbia were the most accepting of female topless behaviour and those living in Atlantic provinces were the least accepting. People living in British Columbia have been typically found in the Maclean's national surveys (Clark, 1999; Milne, 1999) to be the second most liberal province after Quebec in their sexual attitudes. For example, people in Quebec and British Columbia have been found to be the most accepting of legalized prostitution (Jenish, 1994). Determining whether our finding that people in British Columbia are now more accepting of topless behaviour than are those in Quebec are generalizable, and what factors contribute to this pattern of results, are beyond the scope of our study and require more in-depth questions.
Previous studies have found that married people are more conservative in their sexual attitudes than those who are single or living in common law relationships (Laumann et al., 1994). In this study, the chi-square analyses also indicated significant differences between those who were married and those who were single in that for the three contexts, married people were more likely to believe it should be illegal for women to go topless. However, in the logistic regression analyses marital status was not significantly predictive of attitudes in any of the three contexts. This indicates that marital status may only be predictive when it is analyzed on its own. When other demographic variables are included in the analysis, marital status loses its significance because of its shared variance with some of these other variables.
Overall, Canadian attitudes toward the legality of female topless behaviour are affected by the context in which the behaviour occurs. Specifically, more Canadians think it should be legal for women to go topless at public beaches than in other public contexts such as in public parks or on city streets. However, the majority of Canadian adults are still opposed to this latter behaviour being legal despite the Gwen Jacob's case, in which female toplessness in this most restrictive case was effectively rendered legal by the courts' determination that it is not an indecent act.
Like all studies, ours contains some limitations that we could not rectify. For instance, the data we used has many single-item measures rather than scales with established psychometric properties; a common limitation of national opinion surveys. Another limitation is that the items related to topless behaviour only focused on the issue of legality and thus many other issues concerning toplessness were not included, such as the attitudes and motivations of women who choose to go topless in public. While the use of secondary data often limits the availability of research options, this nationally representative survey did provide a rare opportunity to examine an important normative issue regarding Canadian attitudes toward toplessness.
We assume our study was also restricted by a relatively high non-response rate. Michael, Gagnon, Laumann, and Kolata (1994) estimated that between 25 and 35% of individuals refuse to complete telephone sexual surveys. Also, non-responders were mostly male, older (O'Hara, 2001), and those with less education. The survey organization purposely asked to speak to both males and females to ensure gender balance and the sample also had a wide range of education and ages. The response rate was similar to other national telephone surveys, and we were pleased to note that the demographics of the sample were very similar to that of the Canadian population, and that the non-response rates were quite low among those who responded to the survey. Given that typical non-response rates for sexuality surveys are much higher (Michael et al., 1994), the fact that most respondents who agreed to participate did answer the relevant questions can be perceived as a strength of this study.
Despite these limitations, we believe that our study has two main contributions. First, we showed that context plays a role in determining the acceptability of female topless behaviour in Canada. It is clear that Canadians are more accepting of topless behaviour on the beach than in other contexts. This finding provides additional support to the body of research examining the effects of context on the definition of topless behaviour as sexual or non-sexual (Forsyth, 1992; Herold et al., 1994; 1995, Maticka-Tyndale et al., 1998; Mewhinney et al., 1995; Redmon, 2002; Shrum & Kilburn, 1996). In our study many Canadians expressed disapproval of female topless behaviour for each of the three contexts. This finding may provide some indication as to why the great majority of women in Canada do not go topless in public. It would appear that many Canadians continue to view public female breast exposure as sexual, regardless of context, and hence unacceptable. Even women who favour legal acceptance of toplessness in some contexts (e.g., at the beach) might well be dissuaded given the anticipated level of disapproval and the apparent lack of public consensus on this topic.
A second major contribution of this study was the use of logistic regression to analyze which demographic variables from a national sample were the most predictive of attitudes toward topless behaviour. In most national surveys of sexual attitudes (e.g., Laumann et al., 1994) demographic variables are studied in isolation from one another, which limits the ability to consider the relative importance of each predictor within the context of their correlations with other demographic variables. This issue is well illustrated by the marital status variable, which had been identified in past studies as a significant predictor of various attitudes and behaviours related to sexuality and yet did not significantly predict our respondents' perceived support for toplessness across the three contexts considered in this study. Our findings show that marital status is well correlated with age and more modestly, but still significantly, correlated with both gender and religiosity. The removal of the variance shared by these predictors, as is the case in logistic regression, provides a more accurate indication of their unique contribution to the prediction of people's views about toplessness across our three different contexts. Such information should help sexuality researchers determine with some degree of accuracy what factors indeed predict attitudes toward toplessness in Canada. We suspect that such attitudes are a potentially informative indicator of Canadians' social and sexual norms.
Barrett, M., King, A., Levy, J., Maticka-Tyndale, E., & McKay, A. (1997). Canada. In R. Francoeur (Ed.), The International Encyclopedia of Sexuality (pp. 221-343). New York: Continuum Publishing Company.
Baumeister, R.F., & Tice, D.M. (2001). The Social Dimension of Sex. Toronto, ON: Allyn and Bacon.
Bibby, R. W. (2001). Canada's Teens: Today, Yesterday and Tomorrow. Toronto, ON: Stoddart Publishing.
Canadian Press. (1992). Bare-breast protests draw cheers, leers. The Toronto Star, July 20, A9.
Christopher, F.S., & Sprecher, S. (2000). Sexuality in marriage, dating, and other relationships: A decade review. Journal of Marriage and the Family, 62, 999-1017.
Clark, A. (1999). Bastion of liberalism: Quebec's credo: live and let live. Maclean's, January 4, p. 44.
Compas Polling (1998). [Modern life survey of the Canadian adult population]. Unpublished raw data.
DeLonghi, G. (1992). Court hears Jacob's indecency appeal. The Guelph Tribune, June 24, p. 3.
Ford. C.S. & Beach, F.A. (1951). Patterns of Sexual Behavior. New York: Harper Row.
Forsyth, C.J. (1992). Parade strippers: A note on being naked in public. Defiant Behavior: An Interdisciplinary Journal, 13, 391-403.
Gallup Canada (1992). More men than women back topless right. Toronto Star.
Herold, E., Corbesi, B., & Collins, J. (1994). Psychosocial aspects of female topless behavior on Australian beaches. The Journal of Sex Research, 31, 133-142.
Herold, E., Corbesi, B., & Collins, J. (1995). Attitudes toward female topless beach behaviour: A study of male Australian university students. The Canadian Journal of Human Sexuality, 4, 177-182.
Jenish, D. (1994). Canada under the covers. Maclean's, January 3, pp.20-21.
Laumann, E.O., Gagnon, J.H., Michael, R.T., & Michaels, S. (1994). The Social Organization of Sexuality: Sexual Practices in the Unites States. Chicago: The University of Chicago Press.
Lefkowitz, E.S., Gillen, M.M., Shearer, C.L., & Boone, T.L. (2004). Religiosity, sexual behaviors, and sexual attitudes during emerging adulthood. The Journal of Sex Research, 41, 150-159.
Maticka-Tyndale, E., Barrett, M., & McKay, A. (2000). Adolescent sexual and reproductive health in Canada: A review of national data sources and their limitations. The Canadian Journal of Human Sexuality, 9, 41-65.
Maticka-Tyndale, E., Herold, E.S., & Mewhinney, D. (1998). Casual sex on spring break: Intentions and behaviors of Canadian students. The Journal of Sex Research, 35, 254-264.
Mewhinney, D.M., Herold, E.S., & Maticka-Tyndale, E. (1995). Sexual scripts and risk-taking of Canadian university students on spring break in Daytona Beach, Florida. The Canadian Journal of Human Sexuality, 4, 273-288.
Michael, R.T., Gagnon, J.H., Laumann, E.O., & Kolata, G. (1994). Sex in America: A Definitive Survey. Boston, MA: Warner Books.
Milhausen, R.R., & Herold, E.S. (1999). Does the sexual double standard still exist? Perceptions of university women. The Journal of Sex Research, 36, 361-368.
Milne, C. (1998/1999). Sex and the single teen. Maclean's, pp. 48-49. More men than women back topless right. (August 24, 1992). Toronto Star, p. 14.
Murdock, R. (1996). Our bodies, our breast--The person just got political ... Again. Herizons, 11, 27-29.
O'Hara, J. (2001). Public pillow talk. Maclean's, January 1, pp. 38-40.
Oliver, M.B., & Hyde, J.S. (1993). Gender differences in sexuality: A meta-analysis. Psychological Bulletin, 114, 29-51.
Rathus, S.A., Nevid, J.S., Fichner-Rathus, L., & Herold, E.S. (2004). Human Sexuality in a World of Diversity. Toronto, ON: Pearson.
Redmon, D. (2002). Testing informal social control theory. Examining lewd behavior during Mardi Gras. Deviant Behavior: An Interdisciplinary Journal, 23, 363-384.
R. v. Jacob (1996). Court of Appeal for Ontario. Canadian Legal Information Council. Available online: http:// www.canlii.org/on/cas/onca/1996/1996onca10568.html
Shrum, W., & Kilburn, J. (1996). Ritual disrobement at Mardi Gras: Ceremonial exchange and moral order. Social Forces, 75, 423-458.
Smith, T.W. (1992). Discrepancies between men and women in reporting number of sexual partners: A summary from four countries. Social Biology, 39, 203-211.
Statistics Canada (1991). Health Promotion Survey, 1990. [On-Line]. Available: http://www.statcan.ca/english/ IPS/Data/82M0007GPE.htm.
Statistics Canada (2002). 2001 Census: Families and household profile. [On-line]. Available: http:// www.2.Statcan.ca/english/census01/products/ analytic/companion/fam/Canada.cfm.
Dayna S. Fischtein Edward S. Herold
Department of Family Relations and Applied Nutrition
University of Guelph
Department of Psychology
University of Guelph
Correspondence concerning this paper should be addressed to Dayna Fischtein, Department of Family Relations and Applied Nutrition, University of Guelph, Guelph, ON, Canada, N1G 2W1
Table 1 Distribution of Sample by Demographic Variables Variable Percentage N Gender Male 50.0 739 Female 50.0 740 Age 60 and over 26.9 372 50-59 23.8 330 40-49 22.0 305 30-39 13.9 193 20-29 13.3 184 Education Post-Graduate/University Graduate 24.8 366 College/Some University 41.2 608 High School Graduate 22.8 337 Less Than High School 11.2 166 Marital Status Single 31.1 459 Common Law 9.8 145 Ever Married 59.1 873 Religiosity (Frequency of Attendance) Never 24.3 358 A Few Times a Year 46.0 678 One to a Few Times a Month 13.5 199 Once a Week or More 16.2 238 Region Atlantic 4.7 70 Quebec 28.5 421 Ontario 21.7 321 Prairies 26.8 397 BC 18.3 270 Note: Values exclude those who did not respond to the questions. Table 2 Distribution of Responses to Toplessness Questions * Illegal for Women to go Topless on City Streets (%, n) Gender Yes No Male 65.6 (442) 34.4 (232) Female 78.1 (540) 20.9 (143) Total 72.4 (982) 27.6 (375) Illegal for Women to go Topless in Public Parks (%, n) Gender Yes No Male 50.5 (339) 49.5 (322) Female 72.7 (492) 27.3 (185) Total 62.1 (831) 37.9 (507) Illegal for Women to go Topless at Public Beaches (%, n) Gender Yes No Male 38.5 (258) 61.5 (412) Female 57.6 (392) 42.4 (289) Total 48.1 (650) 51.9 (701) * Values are reported for respondents only. Those who said, "don't know" and or did not respond (7.7% to 10.5% depending on the question) were excluded from these calculations. Table 3 Correlation Matrix for Topless Questions and Constructed Topless Scale Topless Context 1 2 3 4 1. Street -- .62 * .52 * .83 * 2. Park -- .64 * .88 * 3. Beach -- .85 * 4. Topless Scale -- Note: * p < .01; topless scale responses ranged from 0 (not legal in any context) to 3 (legal in any context). Table 4 Correlation Matrix for Predictor Variables Predictor Variable 1 2 3 4 5 6 1. Region -- .00 -.03 .04 -.04 -.03 2. Age -- .15 * -.03 .16 * .57 * 3. Education -- -.01 -.01 .04 4. Gender -- -.10 * -.13 * 5. Religiosity -- .15 * 6. Marital Status -- Note: * p < .001. Table 5 Results of Logistic Regression Analysis for Toplessness on City Streets Predictors B S.E Wald Exp (B) Gender .59 .13 20.25 *** 1.80 Age 60 and over .06 .27 .06 1.07 50-59 -.28 .25 1.24 .76 40-49 -.48 .24 3.96 * .62 30-39 -.13 .27 .23 .88 20-29 9.44 * Education University Graduate -.20 .23 .72 .82 College/Some University .00 .23 .00 1.00 High School Graduate .13 .25 .25 1.13 Less Than High School 3.23 Marital Status Single -.33 .18 3.21 .72 Common Law -.06 .22 .08 .94 Ever Married 3.31 Religious Attendance Never -1.42 .23 37.62 *** .24 A Few Times a Year -.77 .22 12.35 *** .46 One to a Few Times a Month -.22 .28 .64 .80 Once a Week or More 51.74 *** Region Atlantic .17 .33 .28 1.19 Quebec .54 .19 8.36 ** 1.71 Ontario .23 ??19 1.47 1.26 Prairies .68 .19 12.52 *** 1.98 B.C. 15.90 ** Nagelkerke [R.sup.2] = .13 *** Note: *** p < .001; ** p <.01; * p < .05 Table 6 Sex Difference in Legal Acceptance of Female Toplessness in Different Public Contexts % Legal Acceptance Contexts Male Female City streets 34.4 20.9 Public parks 49.5 27.3 Public beaches 61.5 42.4 * Said it should not be illegal. For n values, see Table 2. Table 7 Results of Logistic Regression Analysis for Toplessness in Public Parks Predictors B S.E Wald Exp (B) Gender .85 .12 49.43 *** 2.35 Age 60 and over -.02 .25 .01 .98 50-59 -.12 .23 .30 .88 40-49 -.42 .22 3.67 .66 30-39 .11 .25 .21 1.12 20-29 8.90 Education University Graduate -.56 .22 6.23 ** .57 College/Some University -.15 .22 .45 .87 High School Graduate .03 .24 .02 1.03 Less Than High School 14.34 ** Marital Status Single -.27 .17 2.55 .76 Common Law -.32 .21 2.38 .73 Ever Married .16 Religious Attendance Never -1.20 .21 33.07 *** .30 A Few Times a Year -.67 .19 12.39 *** .51 One to a Few Times a Month -.33 .24 1.99 .72 Once a Week or More 38.62 *** Region Atlantic .49 .32 2.28 1.63 Quebec .37 .18 4.41 * 1.45 Ontario .20 .19 1.10 1.22 Prairies .48 .18 6.82 ** 1.61 B.C. 8.43 Nagelkerke [R.sub.2] = .15 ** Note: *** p <.001; ** p <.01; * P<.05 Table 8 Results of Logistic Regression Analysis for Toplessness at Public Beaches Predictors B S.E Wald Exp (B) Gender .71 .12 36.29 *** 2.04 Age 60 and over -.21 .24 .79 .81 50-59 -.18 .22 .68 .84 40-49 -.51 .21 5.65 * .60 30-39 -.02 .23 .01 .98 20-29 8.59 Education University Graduate -.75 .22 12.03 *** .47 College/Some University -.25 .21 1.44 .78 High School Graduate .16 .22 .54 1.18 Less Than High School 30.07 *** Marital Status Single -.16 .17 .92 .85 Common Law -.15 .21 .55 .86 Ever Married 1.13 Religious Attendance Never -1.47 .20 53.62 *** .23 A Few Times a Year -.95 .18 28.56 *** .39 One to a Few Times a Month -.40 .22 3.35 .67 Once a Week or More 62.51 *** Region Atlantic .60 .31 3.88 * 1.83 Quebec .15 .18 .67 1.16 Ontario .25 .19 1.71 1.28 Prairies .65 .18 13.09 *** 1.92 BC 17.42 ** Nagelkerke [R.sub.2] = .18 *** Note: *** p <.001; ** p <.01; * p<.05
|Printer friendly Cite/link Email Feedback|
|Author:||Fischtein, Dayna S.; Herold, Edward S.; Desmarais, Serge|
|Publication:||The Canadian Journal of Human Sexuality|
|Date:||Sep 22, 2005|
|Previous Article:||Talking to Your Kids About Sex: from Toddlers to Preteens: A Go Parents! Guide.|
|Next Article:||Young women's sexual adjustment: the role of sexual self-schema, sexual self-efficacy, sexual aversion and body attitudes.|