Protective versus risk factors for self-objectification in different age and gender cohorts

The harmful effects of objectification and self-objectification have been widely investigated, but few studies have examined factors that may predict self-objectification. This research intends to assess the protective versus risk role of sociodemographic and physical characteristics (age, BMI), psychosocial variables (self-esteem; self-oriented perfectionism and socially prescribed perfectionism), and social factors (influence of family and friends; internalization of media standards) on self-objectification in men and women. The selfobjectification was assessed with two subscales of the Objectified Body Consciousness Scale: Body Shame and Body Surveillance. Participants were 812 Italian adults of different age cohorts (age range 21–60 years; 50.7% females) recruited via a quota sampling method. Two regression models separately for males and females were performed. Results showed that mass media influence was the strongest predictor for body surveillance and body shame in both men and women, whereas gender-related patterns emerged for physical, psychological, and relational variables with age as moderator.

Self-objectification can lead to harmful outcomes, such as depression, disordered eating, sexual dysfunction and have a negative impact on interpersonal relations (see Moradi & Huang, 2008). Self-objectification has been extensively studied in women (see Grabe, Ward, & Hyde, 2008), nevertheless researchers have begun to explore men's experience as well (Johnson, McCreary, & Mills, 2007;Rollero, 2013). Men seem to show lower levels of self-objectification than women do; nevertheless, young male adults pay increasing attention to their physical appearance (Rollero, 2013;Weltzin et al. 2005). This concern with one's looks and attractiveness may reflect the intensifying objectification of the male physique in Western societies and the consequent interest men have about maintaining a healthy body image (Daniel, Bridges, & Martens, 2014;Johnson et al. 2007). What few studies involving men have shown is that selfobjectification is usually lower and that patterns differ from those seen in women (Calogero, 2009;Rollero & Tartaglia, 2016).
Deep links between self-objectification and its psychological consequences have been demonstrated, but less is known about the factors that may predict self-objectification. In the present study, we sought to extend current knowledge about self-objectification by examining factors that may contribute to or protect against this process. Identifying the factors that may promote or prevent selfobjectification can yield insight into how to interrupt its development and harmful effects. In this study we focus on the potential role of sociodemographic and physical characteristics (i.e., age, body-mass index), psychosocial variables (i.e., self-esteem, personal and socially prescribed perfectionism), and social influence (i.e., influence exerted by parents, friends, sexual partner and mass media) in the development of self-objectification.

Antecedents of self-objectification: risk and protective factors
Two biological characteristics seem to be most at stake when talking about self-objectification: age and body-mass index (BMI). An elevated body mass has been closely linked to increased body dissatisfaction and self-objectification among women but its effect on men is less clear cut (Slevec & Tiggemann, 2011). Regarding age, research has highlighted that the concern for body image is a pervasive lifetime issue especially for women (Borges, Gaspar de Matos, & Diniz, 2013). Despite this, participants enrolled in such studies are usually adolescents or young adults and mostly female, to the exclusion of middle-aged PSIHOLOGIJA, 2018 OnlineFirst, 1-13 women and men (Ålgars et al., 2009). The limited number of studies that have involved middle age women have provided intriguing but inconsistent arguments. Some scholars argue that, as women age, they become less objectified by society, are less pressured to be attractive, and therefore may show lower levels of selfobjectification (Tiggemann & Lynch, 2001). Accordingly, in a cross-sectional study involving women aged between 18 and 64 years, Greenleaf (2005) found that younger women reported higher levels of self-objectification. Other scholars, in contrast, note that the growing demand for skin creams and cosmetic surgery may indicate that older adults are becoming increasingly interested in retaining an attractive appearance (Ring, 2000). Moreover, there is some support for a positive relationship between fear of aging and body dissatisfaction in middle-aged women (Midlarsky & Nitzburg, 2008). McKinley's longitudinal research (2006) showed that body surveillance and body shame -i.e. the two main dimensions of self-objectification -are stable correlates of body-esteem for young women, while the importance of body surveillance is less relevant for young and middle-aged men. Moreover, as compared with young men and middle-aged women, the correlations between body surveillance and body shame are stronger for young women. Nevertheless, since to date very few studies have included samples of middle-aged women and men, there is a clear need for more research in this age group (Ålgars et al. 2009).
With reference to psychological dimensions, some characteristics have been identified that foster or protect against self-objectification. Among them, self-esteem plays a key role. Although there may be a reciprocal relation between self-esteem and body satisfaction, research has shown that the first can be considered a predictor of both body surveillance and body shame (Green & Pritchard, 2003;Tylka & Sabik, 2010). In their study integrating self-esteem within the objectification framework, Tylka and Sabik (2010) found that selfesteem negatively predicts both body surveillance and body shame. Given that women with high self-esteem are more satisfied with their personal qualities and appearance, they are more likely to accept their body as it is. Consistently, other studies have found that women with high levels of self-esteem are more satisfied with their personal and physical characteristics than women who report lower self-esteem (Green & Pritchard, 2003;Le Page, Crowther, Harrington, & Engler, 2008).
Another personal quality that may play a role in shaping personal body perception is perfectionism, both self-oriented and socially prescribed. The former refers to critical self-examination and adherence to self-imposed standards, while the latter regards holding to socially prescribed high standards (Hewitt, Flett, Besser, Sherry, & McGee, 2003). According to Midlarsky and Nitzburg (2008), perfectionism may play a role in predicting body dissatisfaction and disordered eating. Perfectionism could also have a more complex role in the process of internalizing the discontent with one's body image (i.e., by interacting with body shape and size, effects of aging on appearance, and influence of sociocultural pressures). Specific study on the relation between perfectionism and self-objectification is desirable.
Concerning social influence, pressure from significant others (e.g., family, partners, peers) has been found to influence body dissatisfaction and disordered eating (Midlarksy & Nitzburg, 2008), but little is known about the extent to which it may influence self-objectification. Katz-Wise and colleagues (2013) examined the relationship between mothers' and adolescents' self-objectification, highlighting that mothers' body shame positively predicts adolescents' body surveillance. To our knowledge, the influence of significant others on selfobjectification has not been sufficiently explored. In order to develop the knowledge on this topic, it may be useful to examine pressures (i.e., romantic partner, friends, and parental modelling) that may affect self-objectification in men and women.
The most widely studied and socioculturally relevant factor in the process of self-objectification is the influence of mass media. A complex interplay between exposure to objectified media models and self-objectification has been found for both men and women (Vandenbosch & Eggermont, 2014). Studies have highlighted that processes of internalisation of (objectified) media standards lead to self-objectification and an objectified perception of personal worth (Parent & Moradi, 2011;Rollero, 2015).

The current study
The study sought to extend current knowledge about self-objectification by evaluating several biological, psychological, and sociocultural factors that may foster or prevent it. Moreover, since previous work has involved mainly female adolescents or college students, a particular feature of this study is that men and women from different age cohorts made up the study sample.
In line with previous research self-objectification was operationalised through the construct of objectified body consciousness (McKinley, 2011) and the two main components of this construct were measured: body surveillance and body shame (see Calogero, 2009;Greenleaf, 2005;McKinley, 2006;Parent & Moradi, 2011;Tylka & Sabik, 2010). We hypothesised that a normal BMI and self-esteem may act as protective factors for body surveillance and shame, while perfectionism, influence of significant others, and internalisation of media standards would represent risk factors for increased body surveillance and shame. Furthermore, since our main aim was testing whether potential protective and risk factors play a similar or different role in men and women, we tested these hypotheses on each gender separately. Finally, since age may represent a core characteristic, we considered it as a potential moderator.
No difference between the sexes was found for age, education or employment. The mean BMI was higher among men than women (25.88 ± 4.04 vs. 22.74 ± 3.79, t = 11.38, p <.001). Women reported a normal BMI more often than men, whereas men were more frequently overweight or obese (χ 2 = 37.27, p <.001).

Procedure
The Ethics Committee of the University of Torino approved the study protocol. Participants were recruited via a quota sampling method (by age and gender) with student assistance. Specifically, they were selected from four subgroups based on age (21-30 years, 31-40 years, 41-50 years, and 51-60 years) and in each subgroup men and women were equally present. Participants were recruited via a quota sampling method. More precisely, students contacted potential participants through recreation centers and personal contacts, selecting subjects who were available to take part in study. Participants were invited to participate in a study about psychosocial issues and were informed that participation in the study was voluntary and that their responses were anonymous. Particular attention was paid to recruit subjects from each age and gender group until the above specified quotas were met. Recruitment and data collection lasted about six months.

Materials
Data were gathered from a structured, self-report, pencil-and-paper questionnaire which took about 20 minutes to complete. We used validated scales -when existing -and translated and back-translated scales for the other measures. The following variables were assessed: Self-objectification: Body shame. The Body Shame subscale of the Objectified Body Consciousness Scale (OBCS, McKinley & Hyde, 1996) is an eight-item scale that measures self-objectification and feelings of shame when one's body does not conform to cultural standards (e.g., "When I can't control my weight, I feel like something must be wrong with me"). Responses were marked on a 7-point scale ranging from "strongly disagree" to "strongly agree" (in the current study Cronbach's α = .76).
Self-objectification: Body surveillance. The Body Surveillance subscale of the OBCS (McKinley & Hyde, 1996) measures the frequency with which participants monitor their physical appearance; it consists of eight items (e.g., "I rarely think about how I look," reverse coded) rated on a 7-point scale ranging from "strongly disagree" to "strongly agree" (in the current study α = .61). Rosenberg's (1965) Self-Esteem Scale (e.g., "I feel that I am a person of worth, at least on an equal plane with others") were rated on 4-point scale ranging from "strongly disagree" to "strongly agree" (in the current study α = .76).

Self-Esteem. The items on
Perfectionism. The Self-Oriented Perfectionism (in the current study α = .70) and the Socially Prescribed Perfectionism (in the current study α = .65) subscales of the Multidimensional Perfectionism Scale (Hewitt & Flett, 1991) are composed of five items each and were rated on 7-point scale ranging from "strongly disagree" to "strongly" agree (e.g., "It makes me uneasy to see an error in my work"; The people around me expect me to succeed at everything I do"). These two subscales were used separately.
Influence of family and friends. The Family and Friends Scale (Myers & Crowther, 2007) is composed of a total of 20 items. Participants were asked whether a specific person (in the current study mother α = .73, father α = .81, partner α = .84, friends α = .80) encouraged them to be worried about their appearance (e.g., "My mother/father/friends/ partner encourages/encouraged me to be concerned with my appearance in general"). Items were formulated with both the present and the past tense in order to allow every participant to answer (even subjects whose parents were deceased). The responses were rated on a 4-point scale ranging from "completely untrue" to "completely true".
Internalization of media standards. The Internalization-General subscale of the Sociocultural Attitudes Towards Appearance Questionnaire-3 (SATAQ-3, Thompson, van den Berg, Roehrig, Guarda, & Heinberg, 2004) assesses internalization of ideals of beauty promoted by the media (e.g., "I compare my body to the bodies of TV personalities and movie stars") and is composed of 10 items. The responses were rated on a 5-point scale ranging from "completely untrue" to "completely true" (in the current study α = .95).
Sociodemographic characteristics. Data were collected on participant age, gender, marital status, educational level, and body weight and height to calculate the BMI (kg/m²).

Data analysis
Correlations between measures used in the study were carried out. Then, t-tests were performed to test gender differences on the studied variables (age, educational level, BMI, body surveillance, body shame, self-esteem, self-oriented and socially prescribed perfectionism, influence exerted by mother, father, friends, and partner, and internalisation of media standards). Finally, to test our hypotheses, we performed two multiple regression models (stepwise method) replicated on each of the two dimension of OBCS. The stepwise technique was chosen for two main reasons. First, our focus was on determining the optimal set of predictors by limiting the number of predictors without significantly reducing the R 2 coefficient. Indeed, to our knowledge, this was the first study aimed at assessing the role of several biological, psychological, and sociocultural factors on men and women from different age cohorts. Second, other techniques, such as hierarchical regression, do not allow managing the problem of multicollinearity, which could have been relevant in our study due to the variables considered.
In each regression model, the dimension of OBCS was regressed onto: age, educational level, BMI, self-esteem, self-oriented perfectionism, socially prescribed perfectionism, influence exerted by mother, father, friends, and partner, and internalisation of media standards. Since age was considered as a moderator, we included the interaction between age and each protective versus risk factor in the regression models. The models were tested within the male and the female sample separately. Statistical analyses were carried out using SPSS software (Version 22.0). Table 1. Body surveillance was positively related to body shame and internalisation of media standards, whereas body shame showed positive correlations with both self-oriented and socially prescribed perfectionism and with all the sources of influence. A negative correlation was found between body shame and self-esteem. This last variable was negatively related also to socially prescribed perfectionism, friends' influence and internalisation of media standards. All the sources of influence were related each other. Finally, educational level was positively related to self-esteem, but also to body surveillance and internalisation of media PSIHOLOGIJA, 2018 OnlineFirst, 1-13 standards, whereas normal weight individuals scored lower on body shame and on significant others' influence. Concerning gender differences, the body shame score was higher for women than for men, while men had higher scores for self-esteem. Marked differences between men and women were also noted in relation to the influence of others: men had higher scores on the influence of father and partner, whereas women had higher scores on the influence exerted by friends (Table 2). Internalisation of media standards predicted surveillance in both men and women. Although this predictor was positively associated with body surveillance for both men and women, the beta value was much higher for the latter; moreover, internalisation of media standards interacted with age only among women, indicating that younger women were especially vulnerable to media influence (Table 3). The proportion of variance of body surveillance explained by predictors is analogous for both genders. Higher educational level and self-esteem acted as protective factors against body shame among women, whereas socially prescribed perfectionism, the influence of mother and friends, and internalisation of media standards fostered body shame (Table 4). Among men, the influence of the mass media played an analogous role, whereas a normal BMI correlated with less body shame. Age moderated the relationship between self-esteem and shame: lower self-esteem was associated with increased shame especially among the younger subjects. Age also interacted with the influence exerted by mother and partner: older men were more susceptible to pressure exerted by such relevant sources of influence. The proportion of explained variance of body shame is much higher than the variance of body surveillance explained by the same model. In the case of the male sample this is particularly significant, as predictors explained the 51% of variance of body shame.

Discussion
This study sought to extend existing knowledge about objectification by examining several factors that may foster or protect against self-objectification. A particular area of focus was participant gender and age, as these two characteristics have been neglected in previous studies investigating selfobjectification. Our findings about body surveillance showed that the influence PSIHOLOGIJA, 2018 OnlineFirst, 1-13 of the mass media is the most powerful risk factor for self-objectification among both men and women, and this was particularly relevant for the younger women. Results concerning body shame revealed a larger set of influencing variables: educational level and self-esteem appear to be protective for women, while a normal BMI is protective for men. Socially prescribed perfectionism and the influence exerted by the mother, friends, and mass media were found to be risk factors for this dimension of self-objectification among women. Internalisation of media standards was noted also among men, while the negative effect of lower self-esteem and influence of others (i.e., mother and partner) was observed only among the younger and the older subjects, respectively.
Taken together, our findings suggest that while self-objectification may be fostered or discouraged by physical, psychological, and social factors, relevant specificities also emerged in relation to the participants' characteristics and to the dimension of self-objectification considered. More specifically, the experience of body surveillance seems to be clearly related only to the internalisation of media standards. The harmful consequences of internalisation have been demonstrated (e.g., Rollero, 2015;Vandenbosch & Eggermont, 2014). We found, however, that not only is self-objectification common to both men and women, but that it appears to be especially harmful for younger women. Two opposing claims have been advanced to explain this difference: some scholars argue that younger women may be more sensitive to media standards as they may gain their selfworth from their appearance, whereas middle-aged and older women may focus more on their accomplishments as a source of self-esteem (Greenleaf, 2005). Others have noted that the fear of aging in middle-aged women may also be seen as a result of media pressures to retain an attractive appearance (Midlarsky & Nitzburg, 2008). Our findings seem to be in line with the first explanation.
Concerning body shame, all the protective and risk factors we examined except for self-oriented perfectionism seem to play a role. Among personal characteristics, the literature has documented that an elevated body mass is linked to greater body shame in women, but the effect of BMI on men still needs to be explored (Slevec & Tiggemann, 2011). Our results do not confirm such a relationship in women but certainly in men. Regarding the role of age, our findings suggest that while age did not influence the level of body shame in women, among men it interacted with self-esteem and the influence of mother and partner. In other words, the experience of shame is analogous for all female cohorts: this is in line with the original theorisation of Fredrickson and Roberts (1997) who defined objectification processes as culturally pervasive experiences especially targeted to women. In men, shame was increased in the younger subjects with low self-esteem and in the older individuals subjected to pressures by significant others. Thus, self-esteem and relational influence are not relevant per se but only in specific situations.
Conversely, self-esteem represented a powerful protective factor for all women, an observation shared by Tylka and Sabik (2010), whereas socially prescribed perfectionism was a risk element. This last finding strengthens previous work on the damaging effects of perfectionism as being positively associated with body dissatisfaction and disordered eating (Slevec & Tiggemann, 2011). Since perfectionism implies striving for compliance with the standards imposed by society, and such standards refer also to physical appearance, women who tend to be more perfectionist feel more ashamed if they fail to meet standards of appearance. Indeed, the images of models promote unrealistic canons for most women and can be particularly harmful for those who strongly feel the need to achieve societal standards, i.e., socially prescribed perfectionists.
Social pressures on women also come from their relationship with their mother and friends (see also Opsenica-Kostić & Stefanović-Stanojević, 2010). Since we did not control for friends' gender, this issue deserves further attention to investigate whether women are influenced especially by other women.
Finally, a positive role is played by women's educational level: to our knowledge, this is the first study to consider the importance of education and the results are optimistic. Indeed, highly educated women are probably more conscious of their worth and competences and are more able to discriminate between societal pressures and their own personal standards and ideals.
Our study has several limitations, the most evident being its cross-sectional research design. In addition, the innovative nature of this research implies some limits and questions that require investigation in further studies. First, we did not consider other constructs related to self-objectification and more genderspecific constructs such as drive for thinness and drive for muscularity. As Parent and Moradi (2011) underlined, the broader sociocultural meaning and consequences of muscularity-related and thinness-related objectification are not equivalent. Second, ideological components may also need to be considered, such as the endorsement of sexist attitudes, in line with the conception of selfobjectification as a dominant cultural lens through which individuals come to view themselves, and through which they perpetuate their own disadvantaged/ advantaged state. Finally, given the relevance of the cultural dimension for these processes (Gervais, Bernard, & Riemer, 2015), a future area of focus would be to investigate the cross-cultural generalisability of these results.