Development and Psychometric Validation of the Women-Sexually Objectifying Media Exposure Scale (W-SOMES) and Men-Sexually Objectifying Media Exposure Scale (M-SOMES)

DOI: https://doi.org/10.21203/rs.3.rs-2746976/v1

Abstract

In this digital era, individuals are regularly exposed to sexual objectification in a variety of media types, which may negatively affect body image. However, existing measures do not fully capture exposure to and direct experiences of sexual objectification in different media. The purpose of the current programme of research was to develop and evaluate the Women-Sexually Objectifying Media Exposure Scale (W-SOMES) and Men-Sexually Objectifying Media Exposure Scale (M-SOMES), to measure exposure to and experiences of sexual objectification in the media. In Study 1, drawing from existing literature and two online surveys (women = 80, men = 76, age representative samples), items for the W-SOMES and M-SOMES were developed. Optimal items solely reflected exposure to sexual objectification in the media. In Study 2, exploratory factor analysis (women = 340, men = 100) suggested an underlying structure of three correlated factors for the W-SOMES and a single factor for the M-SOMES. In Study 3, confirmatory factor analysis (women = 331, men = 328) supported a higher-order model for the W-SOMES (15 items across three subscales: Importance of Physical Appearance, Sexualised Body Representation and Body Evaluation) and a single factor model for the M-SOMES (4 items measuring Sexualised Body Representation). The W-SOMES and M-SOMES displayed satisfactory internal consistency, test-retest reliability, and differentiation by known groups but did not adequately display convergent and discriminant validity. The M-SOMES also displayed satisfactory incremental validity. Future research should further examine the psychometric properties of the W-SOMES and M-SOMES for measuring exposure to sexual-objectification in the media.

Introduction

Sexual objectification is defined as the experience of being treated as a body existing for the use and pleasure of others (Fredrickson & Roberts, 1997). According to objectification theory (Fredrickson & Roberts, 1997), experiences of sexual objectification may lead to self-objectification (the internalisation of the third person’s perspective on one’s body: Fredrickson & Roberts, 1997) and persistent body surveillance (McKinley & Hyde, 1996). The adverse outcomes of experiencing sexual objectification for women and men are supported by a large body of research: For instance, greater sexual objectification experience and exposure is associated with great risk of internalisation of sociocultural ideals of appearance (Vandenbosch & Eggermont, 2012; Vandenbosch & Eggermont, 2013), self-objectification (Aubrey, 2006a ; Vandenbosch et al., 2015;), body surveillance (Aubrey, 2007; Vandenbosch & Eggermont, 2013) and other psychological well-being and health-related effects including increases in body dissatisfaction (Barlett et al., 2008; Strahan et al., 2008). 

Women experience sexual objectification mainly through interpersonal interaction (Kozee et al., 2007), but are also exposed to and directly experience sexual objectification via the media. Since the 1980s, women’s bodies have been frequently shown in fragments (i.e., only a body part) in magazine advertisements (Winship, 1987). In contemporary media, men are also depicted as objects of sexual desire and sexually objectified (Rohlinger, 2002). The media sexually objectifies women and men in multiple different ways: by showing women being commented on (Montemurro, 2003), degraded (Martino et al., 2006), and gazed at (Aubrey & Frisby, 2011), and depicting young, toned, muscular, and bare-chested men being sexually evaluated by young, attractive women (Gill, 2009). In visual media, women wear revealing clothing in television programs (Eaton, 1997), advertising (Lindner, 2004) and music videos (Aubrey & Frisby, 2011), and men’s bodies are shown half-naked or engaging in sexually suggestive acts (Hatton & Trautner, 2011), and their body parts highlighted when displaying products (Rohlinger, 2002). Additionally, media sexually objectifies individuals through thematic messages and stories (Fouts & Vaughan, 2002). Examples include magazine articles referring to the importance of women and men’s bodies for sexual desirability and advising readers on how to enhance their physical attractiveness and fitness to increase their sexual success (Aubrey & Hahn, 2016; Ricciardelli et al., 2010). Moreover, given the increased popularity of social media and online games (Pew Research Center, 2015), individuals living in the current information era are also exposed to and experience sexual objectification online. For example, women’s bodies and sexuality are commented on by online users and both women and men are targeted by inappropriate sexually explicit messages and images (Barak, 2005; Gordon-Messer et al., 2013; Guizzo et al., 2021).  It is clear that women and men are exposed to and experience sexual objectification in a diverse range of ways in equally diverse forms of media. 

Research indicates that individuals’ body image is negatively affected by habitual exposure to sexually objectifying media content (Aubrey, 2006a; Aubrey, 2006b; Aubrey, 2007; Barlett et al., 2008; Dakanalis et al., 2012; Vandenbosch & Eggermont, 2013; Vandenbosch et al., 2015), and experiences of interpersonal sexual objectification online (Lin et al., 2022; Luo et al., 2019). Daily exposure to sexually objectifying fashion magazines, music videos and social networks is associated with greater internalization of sociocultural ideals of appearance, self-objectification, and body surveillance in adolescent girls (Vandenbosch & Eggermont, 2012) and self-objectification in women (Fardouly et al., 2015). Total exposure to everyday sexually objectifying television programs and magazines is also associated with greater body surveillance, body shame, appearance anxiety and sexual self-perception in young women (Aubrey, 2007). Similarly, experiences of interpersonal sexual objectification online, are negatively related to broad conceptualisations of beauty and intuitive eating in Chinese girls (Luo et al., 2019), and self-objectification in Chinese women (Lin et al., 2022). Consistent with patterns in women, frequent exposure to sexually objectifying media content negatively affects men’s body image and is associated with decreases in body-esteem, self-esteem and body satisfaction (Barlett et al., 2008). Greater exposure to sexualising pornographic websites and television programmes is associated with greater internalisation of muscular ideals, self-objectification and body surveillance in adolescent boys (Vandenbosch & Eggermont, 2013) and men (Dakanalis et al., 2012). Finally, a 2-year longitudinal study demonstrated that daily exposure to sexually objectifying TV programs predicted an increase in self-objectification one year later in undergraduate women (Aubrey, 2006a), and self-objectification and body surveillance in undergraduate men (Aubrey, 2006a). 

Measuring Sexually Objectifying Media Exposure

Given the negative effects of sexually objectifying media on individuals’ body image, it is important to be able to accurately measure exposure to and experiences of sexual objectification in both women and men, across different forms of media. However, there are critical issues with current measures of exposure to and experiences of media-based sexual objectification. 

One commonly used measure in media research (e.g., Barzoki et al., 2017; Fardouly et al., 2015; Wright & Tokunaga, 2015) assesses individuals’ exposure to sexual objectification in the media by measuring participants’ frequency of media consumption (e.g., “How long do you spend watching television on a typical day?”). However, individuals’ exposure to sexual objectification may differ as a function of the typical content within the media that they consume, and thus differentially affect self-objectification. For example, exposure to non-appearance focused TV programs (e.g., information-based shows) is negatively associated with self-objectification (Andrew et al., 2016). 

Other measures of exposure to sexual objectification in the media involve time-consuming rating procedures (e.g., Aubrey, 2006a; Aubrey, 2006b; Brown et al., 2007; Dakanalis et al., 2012), where participants first report the frequency of exposure to popular media content (e.g., “How often do you watch Dismissed/The Big Bang Theory?”), and trained judges then rate the content according to how sexually objectifying they consider it to be. The final score is calculated as a composite of the two scores. However, rapid changes in media content and the labour demands of the rating procedure (Den Hamer et al., 2014) increase the difficulty of using this measure, and this method cannot specify which forms of media content negatively affect one’s body image (Aubrey, 2006b). More importantly, this measure can only capture sexually objectifying content in a limited number of media types (e.g., magazine advertisements, TV programmes: Aubrey, 2006a; music videos: Vandenbosch & Eggermont, 2012) and is less well-suited to capturing exposure to and or experiences of sexual objectification in social media and online content. 

In the objectification literature, the Interpersonal Sexual Objectification Scale (ISOS; Kozee et al., 2007) was originally designed to capture women’s experiences of ‘in-person’ interpersonal sexual objectification by focusing on perceived body evaluation and unwanted explicit sexual advances. As women can also directly experience interpersonal sexual objectification in online environments (Gordon-Messer et al., 2013; Guizzo et al., 2021), the ISOS can also be amended to capture this aspect of interpersonal sexual objectification. For example, the Online Interpersonal Sexual Objectification Scale (Luo et al.,2019) was developed to assess Chinese women’s experiences of being evaluated or sexually harassed in the online environment, by integrating items from the ISOS with features of the Chinese online environment. As media content also depicts other women’s bodies being evaluated or harassed (Gill, 2009; Montemurro, 2003), items on the ISOS can also be selected and amended to capture women’s exposure to media portrayals of interpersonal sexual objectification. Indeed, Holland et al. (2017) used a checklist adapted from Kozee et al.’s (2007) ISOS to measure the prevalence of body evaluation and unwanted explicit sexual advances depicted in the media. However, scales based on the ISOS only capture part of people’s exposure to and experiences of sexual objectification, neglecting other forms of sexual objectification that aren’t based around interpersonal interactions, including exposure to sexualised depictions of women and men (Rohlinger, 2002; Stankiewicz & Rosselli, 2008) and sexually objectifying thematic messages (Fouts & Vaughan, 2002; Ricciardelli et al., 2010). Moreover, as the ISOS was initially designed for women, it is unclear whether amended forms of the ISOS adequately capture men’s exposure to and experiences of sexually objectifying media.  

The Current Research 

To date, no existing standardised measures assess exposure to and experiences of sexual objectification across all forms of contemporary media encountered by women and men. This limits our ability to understand the effect of sexually objectifying media exposure and experience on individuals’ body image. As women and men are likely to experience or be exposed to qualitatively different forms of sexual objectification in media (Hatton & Trautner, 2011),  the current research aimed to develop and test the psychometric properties of two separate scales designed to measure exposure to and experiences of sexual objectification in the media in women and in men, which we later called the Women-Sexually Objectifying Media Exposure Scale (W-SOMES) and the Men-Sexually Objectifying Media Exposure Scale (M-SOMES). Given potential differences in sexual objectification experiences as a function of sexual orientation and transgender/cisgender identity (Tebbe et al., 2021), we also anticipated that separate scales would be needed as a function of sexual orientation and transgender/cisgender identity. Given resources were limited, we made the pragmatic choice to sample and develop scales for cisgender heterosexual women and cisgender heterosexual men only. Age representative samples were recruited to maximally capture exposure to and experiences of sexually objectifying media across the lifespan. 

We therefore report three studies from this programme of research. Studies 1a and 1b aimed to generate the initial pool of potential items for the W-SOMES and M-SOMES and obtain participants’ feedback on items, through a literature review and two online surveys. Study 2 aimed to assess the initial factor structure for the W-SOMES and M-SOMES via exploratory factor analysis (EFA). Study 3 aimed to verify the factor structure for the W-SOMES and the M-SOMES by conducting confirmatory factor analysis (CFA) and also examine validity (convergent, discriminant validity, differentiation by known groups, and incremental validity), and reliability (internal consistency and 2-week test-retest reliability). All studies were pre-registered (Study 1a: https://osf.io/q65hn/?view_only=b9ce8f7a77bb4019939e8ae54e09b399; Study 1b: https://osf.io/nv7by/?view_only=ed4e8755e36346718301bf25d286f92e; Study 2: https://osf.io/eqk4f/?view_only=13fa12b49d004f24a9764680e4f424de; Study 3: online ) and all study data and online resources can be found at the Open Science Framework. The hypotheses concerning incremental validity of the W-SOMES subscales in body image examined in Study 3 were not pre-registered. Finally, some of the pre-registered hypotheses have been renumbered or split into separate hypotheses to improve clarity. Any other deviations from the pre-registration are made clear in the manuscript. 

Study 1

As recommended by Boateng et al. (2018), Study 1 involved three scale development steps, including domain generation, item generation, and examination of content validity. In Study 1a, a literature review was conducted to generate relevant domains and items. Participants then rated how often they were exposed to and experienced (i.e., frequency) and how sexually objectifying (i.e., extremity) they considered each item to be, and suggested additional items for inclusion. In Study 1b, the same participants plus an additional sample of participants rated the frequency and extremity of new items. Data from the two studies were then combined for analysis. Ethical approval for both studies was obtained from the [REDACTED] Ethics Subcommittee at a University in the North of England. 

Method 

Design and Participants

Study 1a. Questionnaires were administered via the online survey platform, Qualtrics (Provo, UT). To constrain effects of cultural norms on self-objectification (Crawford et al, 2009), only people who had lived in the UK most of their life were eligible to participate in the online surveys. Data was collected in September 2021.

Data from 80 cisgender heterosexual women (Mage = 36.72, SDage = 11.90), and 76 cisgender heterosexual men (Mage = 36.54, SDage = 11.51) were included in analysis for Study 1a. Most women (66.30%) and men (64.50%) identified as White (see Online Resource 1 for a full breakdown of Study 1a and 1b participant samples by gender and ethnicity: https://osf.io/7pg2n?view_only=dd0709e61d8a45d8a9e84564a5df8f1e). A total of 270 participants originally completed the Study 1a survey. Participants were excluded from all analyses if they were missing more than 20% of survey items (n = 99, 36.67%) or failed attention checks (n = 15, 5.56%). Participants were recruited from online research recruitment platform Prolific (www.prolific.co) and received £2.50 (~$2.96) for their participation. A sample size of 15-30 individuals is suggested to identify patterns across participant-generated data (Terry & Braun, 2011).

Study 1b. Date was collected in November 2021. Data from 81 cisgender heterosexual women (Mage = 37.21, SDage = 11.67) and 79 cisgender heterosexual men (Mage = 37.54, SDage = 11.71) were included in Study 1b. This included 56 women (70% completion rate) and 51 men (67.10% completion rate) from the same participant sample used in Study 1a, and an additional 25 women and 28 men (see Online Resource 1). Participants were excluded from all analyses if they were missing more than 20% of survey items (n = 13, 7.34%) or failed attention checks (n = 4, 2.26%). Participants received £1.88 (~$2.23) for their participation.

Item Development 

Study 1a. Initial items were generated through a review of relevant literature, including Objectification Theory (Fredrickson & Roberts, 1997), sexualising media reviews (APA, 2007; Ward, 2016), content analysis on sexually objectifying media (e.g., Aubrey 2010; Aubrey & Frisby, 2011; Aubrey & Hahn, 2016; Flynn et al., 2016; Gestos et al., 2018; Hatton & Trautner, 2011), and existing measures of sexual objectification (e.g., the Interpersonal Sexual Objectification Scale: Kozee et al., 2007; the Sexual Minority Women’s Sexual Objectification Experience Scale: Tebbe et al., 2021). This stage aimed to generate as many items as possible (Boateng et al., 2018), and did not focus on how sexually objectifying media experience varies across demographic characteristics (e.g., gender, sexual orientation). A total of 47 items were generated that captured five broad domains: Sexualised Body Representation (exposure to media content featuring individuals’ bodies and body parts in sexualised ways, 17 items; Fredrickson & Roberts, 1997), Body Evaluation (exposure to media content featuring comments and behaviours that sexualise and evaluate individuals’ bodies and appearance; 12 items; Kozee et al., 2007), Online Interpersonal Sexual Objectification Experience (experiences of one’s own body and physical appearance being commented on or evaluated online, or being the target of sexual advances by other users; 9 items; Kozee et al., 2007); Importance of Physical Appearance (exposure to media content underscoring the importance of being physically attractive in individuals’ lives, 7 items; Fredrickson & Roberts, 1997) and Explicit Sexual Advances (exposure to the media content depicting sexually aggressive behaviours that are unwelcome and unreciprocated by the receiver, 2 items; Kozee et al., 2007). The research team (consisting of a psychology doctoral student, and two supervisors with expertise in social and health psychology) discussed the degree to which each item was likely sexually objectifying, removed repetitious items, and revised items for clarity. 

In Study 1a, participants were first asked to report their demographic information, including gender, whether they identified as cisgender or transgender, sexual orientation, whether they have lived in the UK for most of their life, age and ethnicity. Participants were provided with definitions of sexual objectification based on the literature, i.e., “ being treated like a sex object, and like your body exists for the use and pleasure of others”, and “when you are sexually objectified, your worth is defined by the pleasure and benefit your body gives to others” (paraphrased from Fredrickson & Roberts, 1997, p. 174 - 175), and “your thoughts, feelings and behaviours do not matter, and what your body looks like indicates what kind of person you are” (paraphrased from Bartky, 1990, p. 20). Participants were then asked to describe any sexually objectifying interactions they have directly experienced on different media platforms (e.g., “Please describe any instances of sexually objectifying interaction you have experienced in video or computer games in the box below”) and any instances of sexual objectification targeting other people of the same gender that they have been exposed to on different media platforms (e.g., “Please describe any instances of sexual objectification that you have observed in TV programs or movies in the box below”), in order to generate new items for evaluation in Study 1b. 

Study 1a participants were then asked to rate how often they have experienced or observed (i.e., item’s frequency), and how sexually objectifying they considered (i.e., item’s extremity) each of the 47 draft items. Frequency and extremity for each item were rated on a scale ranging from 1 (Very infrequently/Not Sexually Objectifying at all) to 100 (Very frequently/Extremely Sexually Objectifying). Item frequency and extremity rating were used as selection criteria for further item reduction. Two attention checks (e.g., select “Very infrequently” or “Not Sexually Objectifying at all” in response to two questions, respectively) were embedded to check the quality of the online survey responses. Participants who failed at least one of the attention checks were excluded from all analyses.

Study 1b. The qualitative data collected in Study 1a were analysed using NVivo 12. Rather than the pre-registered thematic analysis, we instead employed a content analysis to identity key instances of sexual objectification mentioned by participants, and calculate the frequency of each instance across the samples. The research team reviewed 25 items with the highest frequency and discussed the degree to which each instance was likely sexually objectifying. A total of 13 new items were then generated from this content, capturing Sexualised Body Representation (5 items), Explicit Sexual Advances (3 items), Online Interpersonal Sexual Objectification Experience (2 Items), Importance of Physical Appearance (2 items) and Body Evaluation (1 item). Following the procedures detailed above for Study 1a, Study 1b participants were provided with definitions of sexual objectification and then rated how often they have experienced or observed (i.e., item’s frequency) and how sexually objectifying they considered (i.e., item’s extremity) each of the 13 new items. One attention check (i.e., select the “Very infrequently” in response to one question) was embedded, and participants who failed the attention check were excluded from all analyses.

Item Reduction 

Data for Study 1a and Study 1b were aggregated, and then analysed separately for male and female participants. Specifically, the mean score and standard deviation of frequency and extremity of each item were first calculated for both women and men (see Online Resource 2: https://osf.io/6fys8?view_only=dd0709e61d8a45d8a9e84564a5df8f1e). Following the sexually objectifying media exposure rating procedure (Aubrey, 2006a), the mean score for extremity was multiplied by the mean score for frequency, such that the final cross-product score reflected both participants’ frequency of exposure to/experiences of the item and the extent to which the item was considered sexually objectifying. 

For women and men, the extremity, frequency, and cross-products of all items were normally distributed. The pre-registration specified that items with low extremity and frequency would be removed from the item pool, but we did not specify exact criteria a priori. In practice, to remove items with low extremity, we applied different criteria to the W-SOMES and M-SOMES item pools, given their different characteristics. In women, all W-SOMES items’ mean extremity scores were higher than 60 (Mextremity = 75.66, SDextremity= 5.87), suggesting that women considered all items as moderately sexually objectifying. To reduce the number of items in the W-SOMES, we therefore removed 10 items with extremity scores lower than one standard deviation below the mean extremity score of the total pool of W-SOMES items. To reduce the number of items in the M-SOMES, we removed 48 items with extremity scores below 60 (indicating participants did not consider the item as at least moderately sexually objectifying). In both scales, we removed items with standard deviations for extremity greater than the mean standard deviation for extremity of the relevant item pool (12 items removed from the W-SOMES and 7 items from the M-SOMES), as this indicated disagreement in the extent to which participants considered the item to be sexually objectifying. We also removed items with cross-product scores lower than the mean cross-product score of the relevant item pool, thus removing items that had lower extremity and were less frequent (24 items removed from the W-SOMES and 32 items from the M-SOMES). Finally, two additional items each were removed from the W-SOMES and M-SOMES due to conceptual redundancy. This resulted in a final set of 34 items for the W-SOMES (see Online Resource 3: https://osf.io/nvq7d?view_only=dd0709e61d8a45d8a9e84564a5df8f1e), representing Sexualised Body Representation (14 items), Body Evaluation (12 items), Sexualised Body Representation (14 items) and Importance of Physical Appearance (7 items), and Explicit Sexual Advances (1 item) and a final set of 10 items for the M-SOMES (see Online Resource 4: https://osf.io/ye5rh?view_only=dd0709e61d8a45d8a9e84564a5df8f1e)  representing Sexualised Body Representation (7 items) and Importance of Physical Appearance (3 items). 

Discussion 

Study 1 developed the initial item pools for the W-SOMES (34 items) and M-SOMES (10 items). The draft W-SOMES included items representing Sexualised Body Representation, Body Evaluation, Importance of Physical Appearance and Explicit Sexual Advances. The draft M-SOMES included items representing Sexualised Body Representation and Importance of Physical Appearance. The Domains of Sexualised Body Representation and Importance of Physical Appearance in the W-SOMES and the M-SOMES indicate that media tends to sexually objectify women and men by visually or verbally featuring them in sexualised way (Aubrey & Frisby, 2011) and thematically highlighting the value of physical appearance in people’s lives (Aubrey & Hahn, 2016; Ricciardelli et al., 2010). The unique domains of Body Evaluation and Explicit Sexual Advances in the W-SOMES versus the M-SOMES indicates that media disproportionately sexually objectifies women within interpersonal encounters, specifically those involving the evaluation of or action towards women’s bodies. This finding is consistent with previous media content analysis (Aubrey & Frisby, 2011; Montemurro, 2003), suggesting women are more likely to be the target of gaze and experience gender-based harassment than men. 

At odds with previous research (e.g., Gordon-Messer et al., 2013; Guizzo et al., 2021), neither the W-SOMES nor M-SOMES included items representing Online Interpersonal Sexual Experience. Most of these items were removed because they were not frequently experienced by the current samples. As such, the draft W-SOMES and M-SOMES solely represent exposure to sexually objectifying media, rather than both exposure and experience. Given that data collection was conducted in the context of increased digital interaction due to the COVID-19 Pandemic (Barr & Copeland-Stewart, 2022; Cellini et al., 2020), we might expect evidence of increased experiences of sexual objectification online. In contrast, the current study suggests that relative to high levels of exposure to sexually objectifying content across different media types, direct experiences of sexual objectification online are less common. This may be explained by the selection of age-representative samples that capture older as well as younger age groups, while older generation do not use online platforms for interacting with other people as much as younger generation (Pew Research Center, 2021).

Study 2

Study 2 reports an exploratory factor analysis (EFA) that aimed to explore the factor structure of the draft 34-item W-SOMES and the draft 10-item M-SOMES and identify optimal items for the final scales. As the number of items corresponding to Explicit Sexual Advances in the W-SOMES was below three (Hair et al., 2014), it was expected that the W-SOMES would capture three domains: Sexualised Body Representation, Body Evaluation, and Importance of Physical Appearance; and that the M-SOMES would capture two domains: Sexualised Body Representation and Importance of Physical Appearance. 

Method

Design and Participants

A cross-sectional survey was conducted online via Qualtrics. A total of 350 cisgender heterosexual women and 105 cisgender heterosexual men completed Study 2. Participants were recruited from Prolific, and those who completed the survey received £1.25 (~$1.49) for their participation. Sample sizes followed the guidance that a minimum of 10 participants per item is recommended for conducting EFA (Guadagnoli & Velicer, 1988). 

Participants were excluded from all analyses if they were missing more than 20% of survey items (nwomen= 7, 1.37%; nmen = 5, 4.76%) or failed attention checks (nwomen = 2, 0.57%; nmen = 0). Only the first submission of the data was retained if participants completed the survey more than once (nwomen = 1, 0.29%; nmen = 0). Accordingly, data from 340 cisgender heterosexual women (Mage = 38.32, SDage = 11.52), and 100 cisgender heterosexual men (Mage = 38.33, SDage = 11.85) were analysed. Most women (89.7%) and men (85%) identified as White (see Online Resource 5 for detailed demographics: https://osf.io/rcg3s?view_only=ab90c6705d92486fa582b5c754555014). 

Measures

W-SOMES and M-SOMES. Women completed the 34-item W-SOMES, and men completed the 10-item M-SOMES from Study 1. Based on Kozee et al.’s (2007) Interpersonal Sexual Objectification Scale, participants reported how often they noticed each sexually objectifying act occurring in various forms of media within the past year on a five-point Likert- scale (i.e., 1 = never, 2 = rarely, 3 = occasionally, 4 = frequently, 5 = almost always). A mean score for all items was calculated for each participant, such that higher mean scores indicate more frequent exposure to sexually objectifying media. 

Attention Check.  To check the quality of the online survey responses, two attention checks (i.e., select the “Never” or “Rarely” option to two questions, respectively) were embedded in the W-SOMES, and one in the M-SOMES. Participants who failed at least one of the attention checks were excluded from all analyses. 

Procedure 

Data was collected in January 2021. Ethical approval was obtained from the [REDACTED] Ethics Subcommittee at a University in the North of England. In the study advertisement, participants were recruited to participate in a study about sexualised representations of male/female bodies in the media. 

Participants were asked to report their demographic information, including gender, cisgender/transgender identity, sexual orientation, whether they have lived in the UK most of their life, age and ethnicity. Women were then asked to complete the 34-item W-SOMES, and men were asked to complete the 10-item M-SOMES. Items were presented in random order for each participant. 

Results 

W-SOMES

Data Screening. For the final 340 participants, scores on all 34 items were normally distributed (Kline, 2011). The Kaiser-Mayer Olkin measure of sampling adequacy was .96, and Bartlett’s Test of Sphericity was statistically significant, χ2 (561) = 8298.18, < .001, indicating data were appropriate for conducting factor analysis (Tabachnick et al., 2007). 

EFA. EFA was conducted using IBM SPSS Statistics 26. The principal axis factoring method was used for factor extraction as this method can provide a factor solution uncontaminated by error variance and unique variance (Tabachnick et al., 2007). Oblique rotation was employed because factors were expected to be correlated, and Promax was chosen as it resulted in a simpler structure and fewer cross-loading variables (i.e., values with loadings of .30 or higher on more than one factor) than direct oblimin in the current dataset (Brown, 2009). Parallel analysis (Horn, 1965) was used in conjunction with Kaiser’s eigenvalue criterion to determine the factors to retain. The Kaiser’s eigenvalue criterion suggested retaining up to five factors. Based on 500 random data sets and 95th percentile, the parallel analysis (using principal axis/common factor analysis) found the first five factors of the actual data had eigenvalues (15.48, 2.57, 1.81, 1.49, 1.23) greater than the eigenvalues generated from the random data (0.84, 0.74, 0.66, 0.60, 0.55). Therefore, five factors were extracted, accounting for 66.42% of the variance. However, the parallel analysis using principal axis factoring trends could indicate more factors than finally extracted (Buja & Eyuboglu, 1992). 

The pattern matrix loading was examined to make item retention decisions. For a sample size of 300, a pattern factor loading value (λ) of .298 is considered significant (Stevens, 2002). Items were retained if they had λ over .40 on the primary factor and values less than .30 on other factors to maximise confidence in the factors derived from the solution. Of the initial 34 items, 31 items met the criteria (with 8 items on Factor 1, 7 items on Factor 2, 8 items on Factor 3, 6 items on Factor 4, and 2 items on Factor 5). The 5-factor solution revealed by the first round of EFA did not display a clear and interpretable pattern. The 31 items were further screened for conceptual and statistical redundancy (i.e., those items with an inter-item correlation above .70). In the item clusters with high inter-item correlations, only the item with the highest factor loading value on its factor was retained. A cluster of 8 items on Factor 1 had similar conceptual meaning and inter-item correlations greater than .67. The item with the highest factor loading value on Factor 1 (i.e., Item 12) was retained, resulting in the deletion of 7 items. Similarly, items on Factor 2 (n = 1), Factor 3 (n = 2) and Factor 5 (= 3) were deleted due to high inter-item correlations with other items and lower factor loading value. Overall, 21 items were retained (with 2 items on Factor 1, 6 items on Factors 2, 3, 4 respectively, and 1 item on Factor 5). 

The 21 items were entered into the second EFA, and 4 factors were extracted, accounting for 63.90% of the variance. Item communalities and factor loadings were then examined to seek a more parsimonious solution that accounts for more variance. Items were deleted if their communalities (h2) were below .40 (i.e., Item 23) or cross-loaded into two factors (i.e., Item 25).

The remaining 19 items were entered into the third EFA, and 3 factors were extracted, accounting for 61.30% variance. Item 6 (λ = .49, h= 42) was deleted as the lowest factor loading value on its factor, with relatively low communality.

The remaining 18 items were entered into the fourth EFA, and 3 factors were extracted, accounting for 62.35% variance. Item 5 (λ = .51, h2 = .42) was deleted as the lowest factor loading value on its factor, with relatively low communality.

The remaining 17 items were entered into the fifth EFA, and 3 factors were extracted, accounting for 63.52% variance. The Item 31 (λ = .60) and item 7 (λ = .60) with the lowest factor loading values on Factor 1 and Factor 2 were deleted, as achieving an equal number of items for each factor was considered to make calculation and interpretation of the scale score simpler for users. 

Given sufficient items loaded on each factor, and clear interpretable patterns (Williams et al., 2010), the retained 15 items were entered into the final EFA, and 3 factors for extraction were specified. The three factors solution accounted for 64.94% variance; Factor 1 accounted for 46.47% of variance; Factor 2 accounted for 10.79% of variance, and Factor 3 accounted for 7.78% of variance. Table 1 displays the W-SOMES, along with the factor loadings for each item. Based on the items associated with each factor, Factor 1 was named ‘Importance of Physical Appearance’, Factor 2 ‘Sexualised Body Representation’, and Factor 3 ‘Body Evaluation’. 

The W-SOMES Total and subscales were moderately to strongly correlated, with r ranging from .54 to .86 (see Table 2). The internal consistency of the W-SOMES Total and three subscales were good, with all Cronbach’s alpha (α) values exceeding .8, and the intercorrelations of EFA factors ranged from .59 to .69. 

M-SOMES 

               Data Screening. For the final 100 participants, data of all 10 items were normally distributed (Kline, 2011). The Kaiser-Mayer Olkin measure of sampling adequacy was .88, and Bartlett’s Test of Sphericity was statistically significant, χ2 (45) = 421.49, < .001, indicating data were appropriate for conducting factor analysis (Tabachnick et al., 2007). 

               EFA. The principal axis factoring method with Promax was used for factor extraction and rotation. The Kaiser’s eigenvalue criterion suggested retaining up to two factors. Based on 500 random data sets and 95% percentile, the parallel analysis (using principal axis/common factor analysis) found the first two factors of the actual data had eigenvalues (4.94, 1.06) greater than the eigenvalues generated from the random data (0.82, 0.58). Therefore, two factors were extracted, accounting for 60.01% of variance. Factor 1 represented Sexualised Body Representation, and Factor 2 represented the Importance of Physical Appearance. 

For a sample of 100, the pattern factor loading (λ) above .512 is considered significant (Stevens, 2002). Item communality (h .40) was also examined for item deletion. Of the initial 10 items, item 8 (λ = .43) and item 9 (λ = .38) were deleted because of low factor loading values. This resulted in 8 items retained (with 7 items in Factor 1 and 1 item in Factor 2). 

The 8 items were entered into the second EFA. One factor was extracted, accounting for 54.62% variance. Item 10 (λ = .49, h= .25) was deleted due to low factor loading and low item communality. This resulted in 7 items retained on a single factor.  

The 7 items were entered into the third EFA. One factor was extracted, accounting for 59.63% variance. Item 1(h= .38) was deleted due to low communality, resulting in 6 items on a single factor. 

The remaining 6 items were entered into the final EFA, and one factor for extraction was specified. The single factor structure accounted for 61.73% variance. All items represented men’s exposure to media content featuring men’s bodies in a sexualised way, i.e., Sexualised Body Representation. Table 3 displays the M-SOMES, along with factor loadings for each item. The internal consistency of M-SOMES was excellent, with Cronbach’s α of .87 and inter-item correlation ranging from .42 to .66. 

Discussion

Study 2 explored the factor structure of the W-SOMES and the M-SOMES. A 15-item three-factor structure was obtained for the W-SOMES, comprising exposure to media content that highlights the importance of physical appearance on women’s lives (Factor 1); represents women’s bodies in sexualised ways (Factor 2); and depicts women’s bodies being sexually evaluated by others (Factor 3). At odds with the assumed two-factor structure, a 6-item single-factor structure was obtained for the M-SOMES, with all items representing exposure to media content that represents men’s body in sexualised ways. Given the small number of items (n = 3) representing the importance of physical appearance in the initial M-SOMES item pool, it is possible that men’s exposure to media content that underscores the importance of physical appearance, does not represent a meaningful component of men’s overall experiences of sexually objectifying media. 

Study 3

In Study 3, a confirmatory factor analysis (CFA) was conducted to verify the three-factor structure of the W-SOMES and the single-factor structure of the M-SOMES obtained with EFA in Study 2. The validity (including convergent, discriminant validity, and differentiation by known groups, and incremental validity), and reliability (including internal consistency and 2-week test-retest reliability) of the final version of the W-SOMES and M-SOMES were also examined. 

Confirmatory Factor Analysis

Following the results of the EFA, Hypothesis 1 predicted that a three-factor structure of the W-SOMES and a single-factor structure of the M-SOMES would provide an acceptable fit to data from samples of cisgender heterosexual women and men, respectively. For the W-SOMES, the fit of a higher-order model with three first-order factors loaded into one second-order factor and a unidimensional model were also explored. 

Convergent Validity

The internalisation of sociocultural standards of appearance (the extent to which individuals consider the societal norms of size and appearance to be appropriate standards for their own size and appearance; Thompson & Stice, 2001), body shame (feelings of shame when one’s body does not conform to internalised appearance standards; McKinley & Hyde, 1996), self-objectification (the internalisation of a third person’s perspective on one’s body; Fredrickson & Roberts, 1997) and body surveillance (viewing the body as an outside observer; McKinley & Hyde, 1996) are associated with more frequent exposure to sexually objectifying media content in women (Aubrey 2006a; Aubrey, 2007; Vandenbosch & Eggermont, 2012 ) and men (Aubrey 2006a; Vandenbosch & Eggermont, 2013). As previously discussed, the Interpersonal Sexual Objectification Scale (ISOS; Kozee et al., 2007) can also be amended to capture individuals’ exposure to media portrayals of interpersonal sexual objectification. Accordingly, Hypothesis 2, 3, 4, 5 and 6 predicted satisfactory convergent validity of the W-SOMES Total and subscales and the M-SOMES with internalisation of sociocultural standards of appearance, body shame, self-objectification, body surveillance, and exposure to media portrayals of interpersonal sexual objectification, operationalised as correlations exceeding r = .3, as recommended by Cohen (1992). 

Discriminant Validity

Both the W-SOMES Body Evaluation subscale and exposure to media portrayals of interpersonal sexual objectification content focus on a similar form of sexually objectifying media content, based around interpersonal evaluation and interaction. In contrast, the other two W-SOMES subscales capture forms of sexual objectifying media content that are based around sexualised portrayals of the body and the value of physical appearance. Accordingly, Hypothesis 7 predicted discriminant validity of the W-SOMES subscales, with exposure to media portrayals of interpersonal sexual objectification being more strongly correlated with the W-SOMES Body Evaluation subscale than other two W-SOMES subscales. This was operationalised as significant differences in correlations between each subscale and exposure to media portrayals of interpersonal sexual objectification, using Fisher’s z.

While the W-SOMES Total, M-SOMES and the measure of exposure to antisocial media content (Den Hamer et al., 2017) both focus on individuals’ exposure to media, they differ in their focus on sexually objectifying content versus more broadly antisocial content. Similarly, while the W-SOMES Total, M-SOMES and the measure of interpersonal sexual objectification experience focus on sexual objectification, they differ in their focus on observed sexual objectification via the media versus directly experienced sexual objectification in person. Accordingly, Hypotheses 8 and 9 predicted discriminant validity of the W-SOMES Total and M-SOMES from exposure to antisocial media content and interpersonal sexual objectification experience, operationalised as correlations below = .2, as recommended by Cohen (1992).

Differentiation by Known Groups

Sexual objectification can be found in almost every medium (Ward, 2016), and it is likely that individuals who spend more time on media (i.e., heavy media users) will experience more sexual objectification than those spending less time on media (i.e., light media users). Accordingly, Hypothesis 10 predicted significantly greater W-SOMES Total and subscales and M-SOMES scores in higher media users than lower media users.

Incremental Validity 

The W-SOMES was expected to capture more facets of media-based sexual-objectification than media portrayals of interpersonal sexual objectification alone, and the M-SOMES was expected to capture men’s true sexually objectifying media exposure. Accordingly, Hypotheses 11, 12 and 13 predicted the incremental validity of the W-SOMES and M-SOMES, such that scores on each scale will predict internalisation of sociocultural standards of appearance, self-objectification, and body surveillance, above and beyond individuals’ exposure to media portrayals of interpersonal sexual objectification, after controlling for in-person experiences of interpersonal sexual objectification. 

Internal Consistency 

               Hypothesis 14 predicted acceptable internal consistency of the W-SOMES Total and subscales and the M-SOMES, operationalised as Cronbach’s α exceeding .70, as recommended by Cronbach (1951). 

Test-retest Reliability 

               Hypothesis 15 predicted acceptable test-retest reliability of the W-SOMES Total and subscales and the M-SOMES over a 2-week test-retest interval, operationalised as the Intraclass Correlation Coefficient (ICC) exceeding .6, as recommended by Cicchetti (1994). 

Method

Design and Participants

Questionnaires were administered via the online survey platform, Qualtrics, at two-time points, two weeks apart. A total of 697 participants completed the Time 1 survey. Participants were excluded from all analyses if they were missing more than 20% of survey items or 5 items on a single scale (n = 33, 5.01%), or failed attention checks (n = 0). Only the first submission of the data was retained if participants completed the survey more than once (n = 5, 0.72%). Accordingly, data from 331 cisgender heterosexual women (Mage = 38.24, SDage = 11.24), and 328 cisgender heterosexual men (Mage = 38.09, SDage = 11.75) were included in analyses at Time 1. Most women (93.4%) and men (89.6%) identified as White (see Online Resource 6 for detailed demographics: https://osf.io/sbjru?view_only=574f2f730bd9466eb4643ce82363b94a). Participants were recruited from Prolific. 

A total of 641 participants completed Time 2. Participants were excluded from all analyses if they were missing more than 20% of survey items or 5 items on a single scale (n = 16, 2.50%), or failed attention checks (n = 10, 1.56%). Only the first submission of the data was retained if participants completed the survey more than once (n = 23, 3.59%). Time 2 data from 5 participants (0.84%) could not be matched with Time 1 data and was excluded, yielding a final sample of 587 participants for conducting test-retest reliability analysis. Accordingly, data from 288 cisgender heterosexual women (Mage = 39.04, SDage = 11.16), and 299 cisgender heterosexual men (Mage = 38.66, SDage = 11.83) were included in analyses involving Time 2 data (87.01% and 91.16% completion rate respectively). Most women and men identified as White (94.1% and 91.3%, respectively). 

Participants who completed both surveys received £3.31 (~$4) for their participation, and those who completed the Time 1 survey only received £0.44 (~$0.53).   

A medium effect size (d = .5) was used to calculate the required sample size for comparing W-SOMES/M-SOMES scores between heavy media users and light media users. G*Power (version 3.1; Heinrich Heine University Dusseldorf, Germany) showed a minimum requirement of 64 participants per group to detect = .5, with 80% power and α = .05, in an independent t-test. As we operationalised heavy and light media user as the upper and lower 20% of the population (Henry J. Kaiser Family Foundation, 2010), an overall minimum sample of 320 per gender group was required. This sample size met the minimum required sample of 200 for CFA (Barrett, 2007) and relevant correlation and regression analyses. 

Measures 

Unless otherwise stated, final scores for each measure were calculated as a mean of all items, such that higher mean scores indicate greater levels of the measured construct. 

W-SOMES and M-SOMES. The scales were administered as described in Study 2. 

The Internalisation of Sociocultural Standards of Appearance. The 9-item Internalisation-General subscale of the Sociocultural Attitudes Towards Appearance Scale-3 (SATAQ3- IG; Thompson et al., 2004) asks participants to rate their levels of agreement with each statement 5-pointscale ranging from “totally disagree” (1) to “totally agree” (5). Cronbach’s α in the current samples of women (α = .96) and men (α = .95) were excellent.

Body Shame. The 8-item Objectified Body Consciousness Body Shame Scale (OBC-Body Shame; McKinley & Hyde, 1996) asks participants to rate their level of agreement with each item using a 7-point Likert-type scale ranging from “strongly disagree” (1) to “strongly agree” (7). Cronbach’s α in the current samples of women (α = .88and men (α = .86) were good.

Self-Objectification. The 14-item Self-Objectification Beliefs and Behaviors Scale (SOBBS; Lindner & Tantleff-Dunn, 2017) asks participants to rate their level of agreement with each using a 5-point Likert-type scale ranging from “strongly disagree” (1) to “strongly agree” (5). The SOBBS has satisfactory psychometric properties for measuring self-objectification in both cisgender heterosexual women and men (Author citation, paper under review). Cronbach’s α in the current samples of women (α = .92) and men (α = .92) were excellent. 

Body Surveillance. The 9-item Objectified Body Consciousness Body Surveillance Scale (OBC-Body Surveillance; McKinley & Hyde, 1996) asks participants to rate their level of agreement with each item using a 7-point Likert-type scale ranging from “strongly disagree” (1) to “strongly agree” (7). Cronbach’s α in the current samples of women (α = .87and men (α = .89) were excellent.

Interpersonal Sexual Objectification Experience. The 15-item Interpersonal Sexual Objectification Scale (ISOS; Kozee et al., 2007) measures interpersonal sexual objectification directed at oneself in-person. Participants reported the frequency of each experience within the past year using 5-point Likert scales ranging from “never” (1) to “almost always” (5). For men, one item was modified to better capture gendered sexual objectification experiences (the term “breast” was changed to “chest” in the item “How often have you noticed someone staring at your breasts when you are talking to them?”; Davidson et al., 2013). Cronbach’s α in the current samples of women (α = .96) and men (α = .89) were excellent.

Exposure to Media Portrayals of Interpersonal Sexual Objectification. Holland et al. (2017) used a Sexually Objectifying Events Checklist (adapted from the ISOS; Kozee et al., 2017) in their ecological momentary assessment to track witnessed sexual objectification of others. We included their checklist but adapted the response options to include the 5-point Likert scale from the ISOS, and had participants complete the measure of the exposure to media portrayals of interpersonal sexual objectification within the past year. Cronbach’s α in the current samples of women (α = .90and men (α =.90) were excellent.

Antisocial Media Content Exposure. The 12-item Antisocial subscale of the Content-based Media Exposure Scale 2 (C-ME2-AS; Den Hamer et al., 2017) measures an individual’s exposure to antisocial media content. Participants reported the frequency of each item using a 5-point Likert scale ranging from “never” (1) to “very often” (5). Cronbach’s α in the current samples of women (α = .93and men (α = 92) were excellent.

Daily Media Usage. Daily Media Usage were measured following Rosen et al. (2013). Participants reported the amount of time they spent on a typical day using different media and technology on a scale ranging from “not at all” to “more than 8h”. Responses were coded into hours of use, including not at all (0), 1-30 mins (.25), 31 min to 1 h (.75), 1-2h (1.5), 3h (3), 4-5h (4.5), 6-8h (7), more than 8h (9). A higher total amount of hours indicates greater levels of daily media usage. 

Attention Check.  To check the quality of the online survey responses, two attention checks (i.e., select the “Rarely” or “Not at all” option to two questions, respectively) were embedded at Time 1, and one at Time 2. Participants who failed at least one of the attention checks were excluded from all analyses.

Procedures 

Data collection started in March 2022 and ended in April 2022. Ethical approval was obtained from the [REDACTED] Ethics Subcommittee at a University in the North of England. 

At Time 1, participants were asked to report their demographic information, including gender, cisgender/transgender identity, sexual orientation, whether they have lived in the UK most of their life, age, ethnicity, education and annual household income. Participants then completed the W-SOMES or M-SOMES first, followed by other questionnaires outlined in the measures section, in random order for each participant. Items in each questionnaire were also presented in random order for each participant. At Time 2, two-weeks later, participants completed the W-SOMES or M-SOMES. A self-generated ID entered in both surveys was used to match participants’ data across time points.

Results

Data Screening

 Data from each W-SOMES item and M-SOMES item were normally distributed, meeting the assumptions of univariate normality for CFA. Mardia’s (1970) normalised multivariate kurtosis estimate of 44.14 and 8.53 indicates that neither the W-SOMES nor M-SOMES data showed multivariate normality; thus, a Maximum likelihood (ML) estimation was used (Muthen & Kaplan, 1992). Data from other measured outcome variables were mostly normally distributed, with values of skewness ranging from -2 to 2 and values of kurtosis ranging from - 1 to 3.5. 

Confirmatory Factor Analysis 

IBM SPSS AMOS 27 was used to conduct the CFA.  The model fit was evaluated using the Chi-square test of exact fit (χ2), Comparative Fit Index (CFI), Tucker Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardised Root Mean Residual (SRMR). A CFI and TLI exceeding .95 (Hu & Bentler, 1999) and RMSEA and SRMR below .08 (Browne & Cudeck, 1992; Hu & Bentler, 1999) indicate an acceptable fit. The Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were used to compare non-nested models, with lower AIC and BIC indicating parsimonious fit. 

W-SOMES. The fit of the three-factor correlational model based on the EFA was first tested. This three-factor correlational model yielded a close fit:  χ2 (87, = 331) = 137.63, <.001, CFI = .981, TLI = .977, RMSEA = .042 (90% CI: .028, .055), SRMR = .032, AIC = 203.630, BIC = 329.100. Though the χ2 test was significant, χ2 statistics are affected by sample size, such that large samples may yield statistically significant results even with well-fitting models (Hair et al., 2014). The item content, factor loadings, and modification indices were screened for potential model improvement. Although the largest modification index observed in the model suggested covariance of the error terms for Item 4 and the Item 27, no change was made, as correlating the error terms could affect the understanding of the phenomenon in question (Hair et al., 2014). Factor intercorrelations in the model were as follows: .81 (Importance of Physical Appearance with Sexualised Body Representation), .89 (Importance of Physical Appearance with Body Evaluation), and .86 (Sexualised Body Representation with Body Evaluation).

A higher-order model was then explored in which the three first-order factors were loaded onto a second-order factor (i.e., general sexually objectifying media exposure). The fit for the higher-order model was identical to that of the previous three-factor correlational model: χ2 (87, = 331) = 137.63, <.001, CFI = .981, TLI = .977, RMSEA = .042 (90% CI: .028, .055), SRMR = .032, AIC = 203.630, BIC = 329.100. All three first-order factors loaded significantly on the second-order factor. Data in the model explained a large amount of variance in the specific factors: 83% of Importance of Physical Appearance, 79% of Sexualised Body Representation, and 94% of Body Evaluation. Factor loadings for the three-factor correlational model and the higher-order model for the W-SOMES were identical (see Figure 1).  

The unidimensional model was finally explored with all 15 W-SOMES items loading on a general sexually objectifying media exposure factor. This unidimensional model yielded poor fit: χ2 (90, = 331) = 219.756, <. 001, CFI = .925, TLI = .913, RMSEA = .082 (90% CI: .072, .093), SRMR = .047, AIC =351.756, BIC =465.820.  

In line with Hypothesis 1 therefore, the three-factor correlational model of the W-SOMES was supported and demonstrated an acceptable fit to data. The 15-item higher-order model with three first-order factors and one second-order model was retained. This higher-order model supported the use of the 15 W-SOMES items to compute three subscale scores, and the use of three subscale scores to measure women’s exposure to sexually objectifying media content. 

M-SOMES. The fit of the 6-item single-factor model was first tested. This model yielded a relatively poor fit:  χ(9, = 328) = 51.75, <.001, CFI = .962, TLI = .937, RMSEA = .121 (90% CI: .090, .153), SRMR = .032, AIC =75.75, BIC = 121.27. The item content, item factor loadings and modification indices were screened and identified two areas for potential model improvement. Item 2 was first deleted as it had the lowest loading on the factor and shared a lower proportion of common variance with the latent factor. Both Item 3 and Item 6 had the same loadings on the factor. As Item 6 was also included in the W-SOMES, and there may be benefits in having some overlap between the W-SOMES and the M-SOMES, Item 3 was deleted, and Item 6 was retained. The 4-item single factor model yielded an acceptable fit:  χ2 (2, = 328) = 6.430, <.001, CFI = .993, TLI = .980, RMSEA = .082 (90% CI: .015, .157), SRMR = .016, AIC =22.430, BIC = 52.774. The factor loading of the 4-item single factor model for M-SOMES was reported (see Figure 2).

In line with the Hypothesis 1 therefore, the single factor model of the M-SOMES was supported; however, 2 items were deleted for model improvement. This 4-item single factor model supported the use of 4 M-SOMES items to measure men’s exposure to sexually objectifying media content. 

Validity and Reliability

W-SOMES. 

Convergent Validity. In line with Hypothesis 6, scores on the W-SOMES Total and subscales were moderately-to-strongly and positively correlated with exposure to media portrayals of interpersonal sexual objectification (see Table 4). However, at odds with Hypotheses 2, 3, 4 and 5, scores on the W-SOMES Total and subscales were weakly correlated with internalisation of sociocultural ideals of appearance, body shame, self-objectification and body surveillance.

               Discriminant Validity. Fisher’s tests were performed to assess differences in correlations between exposure to media portrayals of interpersonal sexual objectification and the W-SOMES subscales. Discriminant validity of the W-SOMES subscales was supported in line with Hypothesis 7, such that exposure to media portrayals of interpersonal sexual objectification had a significantly stronger correlation with the W-SOMES Body Evaluation subscale than the Importance of Physical Appearance subscale (= 3.53, <.001), and Sexualised Body Representation subscale (z = 4.35, < .001). 

               However, discriminant validity of the W-SOMES Total as outlined in Hypotheses 8 and 9 was not supported. Specifically, scores on the W-SOMES Total were weakly and positively correlated with anti-social media content exposure, and moderately-to-strongly and positively correlated with interpersonal sexual objectification experience (see Table 4). 

               Differentiation by Known Groups. Independent t-tests were performed to assess differences in the W-SOMES Total and subscale scores as a function of daily media usage. Differentiation by known groups was supported in line with Hypothesis 10, such that heavy media users scored significantly higher on the W-SOMES Total and three subscales than light media users (see Table 5). 

Incremental Validity. Hierarchical multiple regressions were conducted to examine the incremental validity of the W-SOMES Total in predicting body image concerns, with internalisation of sociocultural ideals of appearance, self-objectification, and body surveillance as outcome variables, and sexual objectification experience and exposure measures as predictor variables. Scores on the ISOS were entered in Step 1, followed by exposure to media portrayals of interpersonal sexual objectification in Step 2, and the W-SOMES Total in Step 3. The average variance inflation factor of each predictor in the model were below or approximately 2, confirming that collinearity was not an issue (Bowerman & O’Connell, 1992). 

The ISOS significantly predicted all three body image concerns in Step1 and Step 2 (see Table 6). Adding exposure to media portrayals of interpersonal sexual objectification in Step 2 resulted in a significant Rchange for the internalisation of sociocultural ideals of appearance and self-objectification, not body surveillance. Adding the W-SOMES Total in Step 3 did not result in a significant Rchange for any body image concern. The ISOS was the only significant predictor of three body image concerns in Step 3. Incremental validity of the W-SOMES Total as outlined in Hypotheses 10, 12, and 13 was therefore not supported. 

Exploratory hierarchical multiple regressions were conducted to further explore the incremental validity of each W-SOMES subscale in predicting body image concerns. The ISOS and exposure to media portrayals of interpersonal sexual objectification were entered in Step 1 and Step 2 respectively, followed by the scores on three subscales of the W-SOMES in Step 3. The ISOS and the W-SOMES Body Evaluation subscale significantly predicted men’s body surveillance in Step 3 (see Table 7). However, adding three subscales of the W-SOMES in Step 3 did not result in a significant Rchange in any body image concern. 

Internal Consistency. In line with Hypothesis 14, Cronbach’s α of the W-SOMES Total and subscales indicated good internal consistency, with all Cronbach’s α values exceeding .84 (see Table 8). 

Test-retest Reliability. In line with Hypothesis 15, the ICC of the W-SOMES Total and subscales indicated moderate to good test-retest reliability, with all ICC exceeding .69 (see Table 8). 

M-SOMES.

Unless otherwise stated, the statistical analyses applied to the M-SOMES below were the same as those reported above for the W-SOMES.

Convergent Validity. In line with Hypothesis 6, scores on the M-SOMES were moderately and positively correlated with exposure to media portrayals of interpersonal sexual objectification (see Table 9). However, at odds with Hypotheses 2, 3, 4, and 5, scores on the M-SOMES were weakly correlated with internalisation of sociocultural ideals of appearance, body shame, self-objectification, and body surveillance. 

Discriminant Validity. Discriminant validity of the M-SOMES as outlined in Hypotheses 8 and 9 was not supported. Specifically, scores on the M-SOMES were weakly and positively correlated with anti-social media content exposure, and moderately and positively correlated with interpersonal sexual objectification experience (see Table 9). 

Differentiation by Known Groups. Differentiation by known groups was supported in line with Hypothesis 10, such that heavy media users scored significantly higher on the M-SOMES than light media users (see Table 5). 

Incremental Validity. The average variance inflation factor of each predictor in the model were below or approximately 2, confirming that collinearity was not an issue (Bowerman & O’Connell, 1992). 

The ISOS significantly predicted men’s internalisation of sociocultural ideals of appearance, self-objectification, and body surveillance in Step1 and Step 2 (see Table 10). Adding exposure to media portrayals of interpersonal sexual objectification in Step 2 resulted in a significant Rchange for all body image concerns. Adding the M-SOMES in Step 3 resulted in a significant Rchange for internalisation of sociocultural ideals of appearance and self-objectification, not body surveillance. The ISOS remained a significant predictor of all body image concerns in Step 3, and exposure to media portrayals of interpersonal sexual objectification remained a significant predictor of internalisation of sociocultural ideals of appearance and body surveillance in Step 3. Incremental validity of the M-SOMES was therefore supported in line with Hypotheses 11 and 12, while incremental validity as outlined in Hypothesis 13 was not supported. 

Internal Consistency. In line with Hypothesis 14, Cronbach’s α of the M-SOMES indicated excellent internal consistency (see Table 8). 

Test-retest Reliability. In line with Hypothesis 15, the ICC of the M-SOMES indicated moderate test-retest reliability (see Table 8). 

Discussion 

Study 3 supported a 15-item higher-order structure for the W-SOMES, with three subscales (Importance of Physical Appearance, Sexualised Body Representation and Body Evaluation) loading into a general sexually objectifying media exposure scale; and a 4-item single factor structure (Sexualised Body Representation) for the M-SOMES. Both the W-SOMES Total and the M-SOMES displayed satisfactory internal consistency, test-retest reliability, convergent validity with exposure to media portrayals of interpersonal sexual objectification and differentiation by daily media usage. The M-SOMES also displayed good incremental validity in predicting men’s body image concerns (except body surveillance) above and beyond exposure to media portrayals of interpersonal sexual objectification. However, the W-SOMES and M-SOMES did not display adequate convergent validity with internalisation of sociocultural ideals of appearance, self-objectification, body shame and body surveillance, nor adequate discriminant validity from antisocial media content exposure and interpersonal sexual objectification experience.

Given the strong correlation among the subscales on the W-SOMES (r = .71-.76) and large overlapping variance (50% - 58% shared variance among subscales), the distinctiveness among the three subscales may be challenged. However, the model fit index indicates that one general scale with three specific subscales, is a better fit for the data than one general scale. In addition, it is possible that high correlations among subscales reflect the nature of sexually objectifying media content targeting women, in which different forms of sexual objectification often co-occur. For instance, images that sexually portray women’s bodies are often paired with articles emphasizing the importance of physical appearance (Aubrey & Hahn, 2016). 

While the W-SOMES Total and M-SOMES demonstrated satisfactory convergent validity with exposure to media portrayals of interpersonal sexual objectification, neither scale had adequate convergent validity with internalisation of sociocultural ideals of appearance, body shame, self-objectification, or body surveillance, operationalised as correlations exceeding r ≥ .3 (Cohen, 1992). However, the small positive correlations were consistent with prior research investigating the relationship between sexually objectifying media exposure and body image concerns (Aubrey, 2006b; Aubrey, 2007; Vandenbosch & Eggermont, 2012; Vandenbosch & Eggermont, 2013). As such, it is possible that our criteria for convergent validity were too stringent.

Neither the W-SOMES nor M-SOMES displayed adequate discriminant validity from sexual objectification directly experienced during interpersonal interaction, such that participants with greater interpersonal sexual objectification experience also reported higher exposure to sexually objectifying media. Selective exposure theory (Zillmann & Bryant, 2008) may explain this relationship. Research demonstrates that individuals who experience interpersonal sexual objectification may develop negative attitudes or beliefs about their bodies, which then increase the likelihood of seeking out sexually objectifying media content in order to support these views of the self (Davids et al., 2018; Moradi et al., 2005). However, in contrast, Aubrey (2006a) argues that self-objectification predicts avoidance of exposure to sexually objectifying media. Discriminant validity of the W-SOMES and M-SOMES from exposure to antisocial-media was also poor, such that total scores on the W-SOMES and M-SOMES were positively associated with exposure to antisocial media. In hindsight, items in the antisocial media exposure scale (C-ME2-AS; Den Hamer et al., 2017) also capture evaluation of the body and sexuality and visual representations of sexualised bodies (i.e., “How often do you watch on the Internet/TV/games/mobile phone/DVD…people who openly talk about sex?” and “...people who are having sex?”), and thus the correlation may be explained by conceptual overlap. 

Although the W-SOMES conceptually captures more facets of sexually objectifying media exposure, relative to media portrayals of interpersonal sexual objectification, the W-SOMES did not demonstrate incremental validity in predicting body image concerns in women. It is possible that exposure to media content that sexualises women’s bodies and emphasises the importance of physical appearance does not have a salient effect on women’s body image concerns. For example, women may not perceive the sexualised models and celebrities portrayed in magazines and advertisement as relevant to the self, and therefore may be less likely to perceive their own body negatively as a result (Mussweiler et al., 2004). Alternatively, women may be aware of the commercial purposes of such imagery when it is embedded in commercial content such as magazine advertisements (Huang et al., 2021). 

The M-SOMES displayed satisfactory incremental validity in predicting the internalisation of sociocultural ideals of appearance and self-objectification in men. However, the M-SOMES did not have adequate incremental validity in predicting body surveillance, at odds with Aubrey’s (2006b) finding that exposure to sexually objectifying magazines increased body surveillance for men. The measurement used by Aubrey (2006b) allows for broader capture of sexually objectifying media based on the judges’ knowledge, while the M-SOMES only captures specific forms of sexual objectification (i.e., Sexualised Body Representation). It is possible therefore, body surveillance is differentially associated with the forms of sexual objectification captured by Aubrey (2006b) relative to sexualised body representations captured by the M-SOMES.

General Discussion

Three studies were conducted to develop and evaluate two scales that measure exposure to sexual objectification in media environments for cisgender heterosexual women (W-SOMES) and cisgender heterosexual men (M-SOMES). Study 1 developed item pools for the W-SOMES and the M-SOMES. Study 2 reported an EFA, identifying three conceptually meaningful factors in the W-SOMES: Importance of Physical Appearance, Sexualised Body Representation, and Body Evaluation, and a single factor for the M-SOMES, representing Sexualised Body Representation. Study 3 reported the CFA and supported a 15-item higher-order structure for the W-SOMES, with three first-order specific subscales, one second-order general sexually objectifying media exposure scale, and a 4-item single factor structure for the M-SOMES, capturing Sexualised Body Representation. The W-SOMES Total and the M-SOMES have satisfactory internal consistency, test-retest reliability, and differentiation by known groups, while convergent and discriminant validity are less well supported. Incremental validity was mostly supported for the M-SOMES while not supported for the W-SOMES (see Online Resource 7 for a summary of the psychometric properties of the W-SOMES and M-SOMES: https://osf.io/em9tk?view_only=574f2f730bd9466eb4643ce82363b94a).

In line with the literature on sexualised media (Ward, 2016), the current programme of research indicates that core facets of sexual objectification in the media include content that underscores the importance of physical appearance, visually represents women’s bodies in sexualised ways, and evaluates women’s bodies. Although existing content analysis of sexually objectifying media suggests men’s bodies are also evaluated by others (Gill, 2009), and that the importance of physical appearance is also emphasised for men (Ricciardelli et al., 2010), the current work suggests these two forms of sexual objectification are not as salient in media contexts as content that the portrays men’s bodies in sexualised ways.

Both the W-SOMES and the M-SOMES incorporate exposure to sexualised representations of the body in advertising and magazines, reflecting that one purpose of sexually objectifying media is to sell products (Conley & Ramsey, 2011). However, in contrast to the M-SOMES, the W-SOMES also captures exposure to media content that promotes the importance of physical appearance or depicting women’s bodies being evaluated by others. Sociocultural factors such as traditional gender norms and differential social power between women and men may explain these differences in factor structure in the W-SOMES and M-SOMES. As there is a social expectation for women to be physically attractive (Crawford, 2022), contemporary media may serve to reinforce those cultural norms (Holbrook, 1987), communicating the message that women are valued first and foremost for their bodies and appearance. In contrast, men are less socially pressured to prioritise their physical appearance (Mahalik et al., 2003), and accordingly, the media is less likely to highlight the importance of physical appearance for men. With respect to body evaluation, while men’s bodies may be evaluated by women (Gill, 2009), given men’s higher social power, men are still more likely to engage in body evaluation as perpetrators than experience body evaluation as targets (Aubrey & Frisby, 2011; Lampman et al., 2002). In contrast, women’s bodies are frequently evaluated in the media (Martino et al., 2006; Montemurro, 2003) and more likely than men to be commented on or be looked at by men in sexualised ways (Grauerholz & King, 1997).

Strengths, Limitations And Future Directions

Since the release of the APA Task Force Report on the sexualisation of girls (APA, 2007), there has been a rapid increase in peer-reviewed papers with a focus on sexually objectifying media (Ward, 2016). The development of the W-SOMES and the M-SOMES can support advances in this field, by providing validated scales for measuring cisgender heterosexual women and men’s exposure to sexual objectification in the media. The W-SOMES and M-SOMES items can be also used to increase participants' awareness and ability to identify sexually objectifying media content or help participants resist self-blame when consuming sexually objectifying media (Shepherd, 2018). In addition, these scales could also help identify at-risk individuals who are exposed to more sexual objectification in the media and support relevant interventions to mitigate the impact on body image.

Interestingly, while the initial aim of our research was to capture both experiences of and exposure to sexually objectifying media, neither the W-SOMES nor M-SOMES captured individuals’ experiences of interpersonal sexual objectification in online contexts, instead focusing solely on exposure to sexually objectifying media without an interpersonal component. While we attempted to generate items relating to experiences of sexual objectification online in Studies 1a and b, these items were later removed as they were relatively less frequently experienced by participants. This may be partially explained by our sampling strategy, which sought to recruit an age-representative sample in the UK. This meant that we oversampled Generations X and Y (i.e., participants that were between 26 and 57 years in 2022 when the data was collected), while undersampling Generations Z and Alpha (i.e., participants that were under 26 years in 2022). Given more recent trends towards digital communication displacing face-to-face interactions (Twenge & Martin, 2020), Generations Z and Alpha’s may experience more online interpersonal sexual objectification than the older generations. As such, while the W-SOMES and M-SOMES capture exposure to sexually objectifying media, future research may focus on developing measures of experiences of interpersonal sexual objectification online.

The current research also samples only cisgender heterosexual women and men, and participants predominantly identified as White. While validation of the scales would ideally be carried out in all samples, resources were limited, and thus the focus on cisgender heterosexual samples was a pragmatic choice. To address this, further research should investigate the psychometric properties of the W-SOMES and M-SOMES in samples comprising other gender, ethnic and racial identities.

Interestingly, while initial item pools for the M-SOMES covered a broad range of instances of sexual objectification in all forms of media, after item reduction, the content of the M-SOMES was restricted to measuring exposure to sexualised body representations in magazines and advertisements. On one hand, these items may reflect the nature of men’s exposure to sexually objectifying media, in that magazines and advertisements are popular media that tend to sexually objectify men’s bodies using sexualised images (Hatton & Trautner, 2011; Mager & Helgeson, 2011). Alternatively, the literature reviewed in the current research may not have adequately captured the breadth of sexual objectification experienced by men. Most items were drafted based on literature specifically targeting women’s experiences (e.g., APA, 2007; Ward, 2016), as there is limited literature exploring how men are sexually objectified in different forms of media. It is possible, therefore, that some unique aspects of men’s exposure to or experiences of sexually objectifying media were not adequately captured, thus explaining the small number of items contained in the M-SOMES. While open-ended questions were included in the initial surveys to capture additional examples of sexual objectification, this may not have been adequate. More relevant items may be generated using focus groups, allowing participants to produce richer and more nuanced examples in collaboration than survey methods (Braun & Clarke, 2013).

It is important to note that experiences of sexual objectification may be subjective, depending on both the objective nature of media content and recipients’ subjective attitudes, preferences and sensitivities to the messages (Barak, 2005). The current study used respondent-based evaluation methods in item reduction, and items which were not considered sexually objectifying by participants were removed. This method allows us to understand participants’ subjective perceptions of items and ensures scale items are grounded in the real-life experiences of cisgender heterosexual women and men (Bleakley et al., 2008). However, without training, participants may be less sensitive or fail to identify subtle sexually objectifying media messages. It may be beneficial for future research to incorporate a brief media literacy training session before participants rate the extent of sexually objectifying items and additionally incorporate expert panels to assess each item according to their own knowledge and expertise (DeVellis, 2022).

Finally, we did not evaluate the predictive validity of the W-SOMES and M-SOMES in the current programme of research. Future research should therefore seek to examine the extent to which individuals’ exposure to sexually objectifying media predicts future body image concerns in women and men. Moreover, although the W-SOMES and M-SOMES capture exposure to sexualised body representations, given the highly gendered nature of these representations, such that women are portrayed in a submissive and powerless position (Stankiewicz & Rosselli, 2008) while men are portrayed as confident and muscular (Hatton & Trautner, 2011), it would be particularly interesting to explore whether exposure to sexualised body representations have similar effects on body image across genders.

Conclusion

The W-SOMES and M-SOMES are novel and valid measures for assessing exposure to sexual objectification in contemporary media for cisgender heterosexual women and men. The W-SOMES advances measurement techniques by capturing women’s exposure to different forms of sexual objectification (i.e., Importance of Physical Appearance, Sexualised Body Representation, and Body Evaluation) across contemporary media types, including visual media, text media and digital media. Given the limited factor structure and number of items in the M-SOMES, additional research is warranted to examine content validity for the M-SOMES. Further research is also needed to examine the predictive validity of the W-SOMES and M-SOME.

References

American Psychological Association, Task force on the sexualization of girls. (2007). Report of the APA task force on the sexualization of girls. Washington, DC: American Psychological Association. Retrieved from http://www.apa.org/pi/women/programs/girls/report-full.pdf

Andrew, R., Tiggemann, M., & Clark, L. (2016). Predicting body appreciation in young women: An integrated model of positive body image. Body Image, 18, 34–42. https://doi.org/10.1016/j.bodyim.2016.04.003

Aubrey, J. S. (2006a). Effects of sexually objectifying media on self-objectification and body surveillance in undergraduates: Results of a 2-year panel study. Journal of Communication, 56(2), 366–386. https://doi.org/10.1111/j.1460-2466.2006.00024.x

Aubrey, J. S. (2006b). Exposure to sexually objectifying media and body self-perceptions among college women: An examination of the selective exposure hypothesis and the role of moderating variables. Sex Roles, 55(3-4), 159–172. https://doi.org/10.1007/s11199-006-9070-7

Aubrey, J. S. (2007). The impact of sexually objectifying media exposure on negative body emotions and sexual self-perceptions: Investigating the mediating role of body self-consciousness. Mass Communication & Society, 10(1), 1–23. https://doi.org/10.1080/15205430709337002

Aubrey, J. S. (2010). Looking good versus feeling good: An investigation of media frames of health advice and their effects on women’s body-related self-perceptions. Sex Roles, 63(1-2), 50–63. https://doi.org/10.1007/s11199-010-9768-4

Aubrey, J. S., & Frisby, C. M. (2011). Sexual objectification in music videos: A content analysis comparing gender and genre. Mass Communication & Society, 14(4), 475–501. https://doi.org/10.1080/15205436.2010.513468

Aubrey, J. S., & Hahn, R. (2016). Health versus appearance versus body competence: A content analysis investigating frames of health advice in women's health magazines. Journal of Health Communication, 21(5), 496–503. https://doi.org/10.1080/10810730.2015.1103328

Barak, A. (2005). Sexual harassment on the internet. Social Science Computer Review, 23(1), 77–92. https://doi.org/10.1177/0894439304271540

Barlett, C. P., Vowels, C. L., & Saucier, D. A. (2008). Meta-analyses of the effects of media images on men's body-image concerns. Journal of Social and Clinical Psychology, 27(3), 279–310. https://doi.org/10.1521/jscp.2008.27.3.279

Barr, M., & Copeland-Stewart, A. (2022). Playing Video Games During the COVID-19 Pandemic and Effects on Players’ Well-Being. Games and Culture, 17(1), 122–139. https://doi.org/10.1177/15554120211017036

Barrett, P. (2007). Structural equation modelling: Adjudging model fit. Personality and Individual Differences, 42(5), 815–824. https://doi.org/10.1016/j.paid.2006.09.018

Bartky, S. L. (1990). Femininity and domination: Studies in the phenomenology of oppression. Routledge.

Barzoki, M. H., Mohtasham, L., Shahidi, M., & Tavakol, M. (2016). Self-objectification and self-sexualization behavior within consumer culture. Applied Research in Quality of Life, 12(2), 425–438. https://doi.org/10.1007/s11482-016-9468-5

Bleakley, A., Fishbein, M., Hennessy, M., Jordan, A., Chernin, A., & Stevens, R. (2008). Developing Respondent-Based Multi-Media Measures of Exposure to Sexual Content. Communication Methods and Measures, 2(1-2), 43–64. https://doi.org/10.1080/19312450802063040

Boateng, G. O., Neilands, T. B., Frongillo, E. A., Melgar-Quiñonez, H. R., & Young, S. L. (2018). Best practices for developing and validating scales for health, social, and behavioral research: A primer. Frontiers in Public Health, 6, 149–149. https://doi.org/10.3389/fpubh.2018.00149

Bowerman, B. L., & O'Connell, R. T. (1992). Linear Statistical Models: An Applied Approach [Review of Linear Statistical Models: An Applied Approach]. Biometrics, 48(1), 333–333. Biometric Society. https://doi.org/10.2307/2532769

Braun, V., & Clarke, V. (2013). Successful qualitative research: A practical guide for beginners. Sage Publications.

Brown, J. D. (2009). Choosing the right type of rotation in PCA and EFA. Shiken: JALT Testing & Evaluation SIG Newspaper, 13(3), 20-25. https://hosted.jalt.org/test/PDF/Brown31.pdf

Brown, J. D., L'Engle, K. L., Pardun, C. J., Guo, G., Kenneavy, K., & Jackson, C. (2007). Sexy media matter: Exposure to sexual content in music, movies, television, and magazines predicts black and white adolescents' sexual behavior. Journal of the American Academy of Child and Adolescent Psychiatry, 46(1), 24.

Browne, M. W., & Cudeck, R. (1992). Alternative ways of assessing model fit. Sociological Methods & Research, 21(2), 230–258. https://doi.org/10.1177/0049124192021002005

Buja, A., & Eyuboglu, N. (1992). Remarks on parallel analysis. Multivariate Behavioral Research, 27(4), 509–540. https://doi.org/10.1207/s15327906mbr2704_2

Cellini, N., Canale, N., Mioni, G., & Costa, S. (2020). Changes in sleep pattern, sense of time and digital media use during COVID‐19 lockdown in Italy. Journal of Sleep Research, 29(4), e13074–n/a. https://doi.org/10.1111/jsr.13074

Cicchetti, D. V. (1994). Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychological Assessment, 6(4), 284–290. https://doi.org/10.1037/1040-3590.6.4.284

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155–159. https://doi.org/10.1037/0033-2909.112.1.155

Conley, T. D., & Ramsey, L. R. (2011). Killing us softly? Investigating portrayals of women and men in contemporary magazine advertisements. Psychology of Women Quarterly, 35(3), 469–478. https://doi.org/10.1177/0361684311413383

Crawford, M. (2022). ISE ebook online access for transformations: Women, gender, and psychology. McGraw-Hill.

Crawford, M., Lee, I. C., Portnoy, G., Gurung, A., Khati, D., Jha, P., & Regmi, A. C. (2009). Objectified body consciousness in a developing country: A comparison of mothers and daughters in the US and Nepal. Sex Roles, 60(3-4), 174-185. https://doi.org/10.1007/s11199-008-9521-4 

Cronbach, L.J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika 16(3), 297–334. https://doi.org/10.1007/BF02310555

Dakanalis, A., Di Mattei, V. E., Bagliacca, E. P., Prunas, A., Sarno, L., Riva, G., & Zanetti, M. A. (2012). Disordered eating behaviors among italian men: Objectifying media and sexual orientation differences. Eating Disorders, 20(5), 356–367. https://doi.org/10.1080/10640266.2012.715514

Davids, C. M., Watson, L. B., & Gere, M. P. (2018). Objectification, masculinity, and muscularity: A test of objectification theory with heterosexual men. Sex Roles, 80(7-8), 443–457. https://doi.org/10.1007/s11199-018-0940-6

Den Hamer, A. H., Konijn, E. A., & Bushman, B. J. (2017). Measuring Exposure to Media with Antisocial and Prosocial Content: An extended version of the Content-based Media Exposure Scale (C-ME2). Communication Methods and Measures, 11(4), 289–299. https://doi.org/10.1080/19312458.2017.1375089

Den Hamer, A., Konijn, E. A., & Keijer, M. G. (2014). Cyberbullying behavior and adolescents' use of media with antisocial content: A cyclic process model. Cyberpsychology, Behavior and Social Networking, 17(2), 74–81. https://doi.org/10.1089/cyber.2012.0307

DeVellis, R. F., & Thorpe, C. T. (2022). Scale development: theory and applications (5th edition.). SAGE Publications, Inc.

Eaton, B. C. (1997). Prime-time stereotyping on the new television networks. Journalism & Mass Communication Quarterly, 74(4), 859–872. https://doi.org/10.1177/107769909707400413

Fardouly, J., Diedrichs, P. C., Vartanian, L. R., & Halliwell, E. (2015). The mediating role of appearance comparisons in the relationship between media usage and self-objectification in young women. Psychology of Women Quarterly, 39(4), 447–457. https://doi.org/10.1177/0361684315581841

Flynn, M. A., Craig, C. M., Anderson, C. N., & Holody, K. J. (2016). Objectification in popular music lyrics: An examination of gender and genre differences. Sex Roles, 75(3-4), 164–176. https://doi.org/10.1007/s11199-016-0592-3

Fouts, G., & Vaughan, K. (2002). Television situation comedies: Male weight, negative references, and audience reactions. Sex Roles, 46(11-12), 439–442. https://doi.org/10.1023/A:1020469715532

Fredrickson, B. L., & Roberts, T.A. (1997). Toward understanding women's lived experiences and mental health risks. Psychology of Women Quarterly, 21(2), 173–206. https://doi.org/10.1111/j.1471-6402.1997.tb00108.x

Gestos, M., Smith-Merry, J., & Campbell, A. (2018). Representation of women in video games: A systematic review of literature in consideration of adult female wellbeing. Cyberpsychology, Behavior and Social Networking, 21(9), 535–541. https://doi.org/10.1089/cyber.2017.0376

Gill, R. (2009). Beyond the ‘Sexualization of Culture’ Thesis: An intersectional analysis of ‘sixpacks’, ‘midriffs’ and ‘hot lesbians’ in advertising. Sexualities, 12(2), 137–160. https://doi.org/10.1177/1363460708100916

Gordon-Messer, Deborah, M. P. H, Bauermeister, Jose Arturo, M. P. H., Grodzinski, Alison, M. L. I. S, & Zimmerman, M. (2013). Sexting among young adults. Journal of Adolescent Health, 52(3), 301–306. https://doi.org/10.1016/j.jadohealth.2012.05.013

Grauerholz, E., & King, A. (1997). Prime time sexual harassment. Violence Against Women, 3(2), 129–148. https://doi.org/10.1177/1077801297003002003

Guadagnoli, E., & Velicer, W. F. (1988). Relation of sample size to the stability of component patterns. Psychological Bulletin, 103(2), 265–275. https://doi.org/10.1037/0033-2909.103.2.265

Guizzo, F., Canale, N., & Fasoli, F. (2021). Instagram sexualization: When posts make you feel dissatisfied and wanting to change your body. Body Image, 39, 62–67. https://doi.org/10.1016/j.bodyim.2021.06.005

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2014). Multivariate data analysis (7th edition) Pearson.

Hatton, E., & Trautner, M. N. (2011). Equal opportunity objectification? The sexualization of men and women on the cover of Rolling Stone. Sexuality & Culture, 15(3), 256–278. https://doi.org/10.1007/s12119-011-9093-2

Henry J. Kaiser Family Foundation. (2010, January). Generation M2: Media in the Lives of 8- to 18- years old. https://files.eric.ed.gov/fulltext/ED527859.pdf

Holbrook, M. (1987). Mirror, mirror, on the wall, what’s unfair in the reflections on advertising. Journal of Marketing, 51(3), 95–103. https://doi.org/10.2307/1251650

Holland, E., Koval, P., Stratemeyer, M., Thomson, F., & Haslam, N. (2017). Sexual objectification in women's daily lives: A smartphone ecological momentary assessment study. British Journal of Social Psychology, 56(2), 314-333. https://doi.org/10.1111/bjso.12152 

Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis. Psychometrika 30, 179–185. https://doi.org/10.1007/BF02289447

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

Huang, Q., Peng, W., & Ahn, S. (2021). When media become the mirror: A meta-analysis on media and body image. Media Psychology, 24(4), 437–489. https://doi.org/10.1080/15213269.2020.1737545

Kline, R. B. (2011). Principles and practice of structural equation modelling (3rd ed.). Guilford.

Kozee, H. B., Tylka, T. L., Augustus-Horvath, C. L., & Denchik, A. (2007). Development and psychometric evaluation of the Interpersonal Sexual Objectification Scale. Psychology of Women Quarterly, 31(2), 176–189. https://doi.org/10.1111/j.1471-6402.2007.00351.x

Lampman, C., Rolfe-Maloney, B., David, E. J., Yan, M., McDermott, N., Winters, S., Davis, J., & Lathrop, R. (2002). Messages about sex in the workplace: A content analysis of primetime television. Sexuality & Culture, 6(4), 3–21. https://doi.org/10.1007/BF02719213

Lin, S., Li, L. and Jiang, L. (2022). Online interpersonal sexual objectification experiences and teenage girls’ self-objectification: The role of broad conceptualization of beauty, Behavioral sciences, 12(7), 210. https://doi: 10.3390/bs12070210.

Lindner, D., & Tantleff-Dunn, S. (2017). The development and psychometric evaluation of the Self-Objectification Beliefs and Behaviors Scale. Psychology of Women Quarterly, 41(2), 254-272. https://doi.org/10.1177/0361684317692109 

Lindner, K. (2004). Images of women in general interest and fashion magazine advertisements from 1955 to 2002. Sex Roles, 51(7-8), 409–421. https://doi.org/10.1023/B:SERS.0000049230.86869.4d

Luo, Y., Niu, G., Kong, F., & Chen, H. (2019). Online interpersonal sexual objectification experiences and Chinese adolescent girls' intuitive eating: The role of broad conceptualization of beauty and body appreciation. Eating Behaviors: An International Journal, 33, 55–60. https://doi.org/10.1016/j.eatbeh.2019.03.004

Mager, J., & Helgeson, J. G. (2010). Fifty years of advertising images: Some changing perspectives on role portrayals along with enduring consistencies. Sex Roles, 64(3-4), 238–252. https://doi.org/10.1007/s11199-010-9782-6

Mahalik, J. R., Locke, B. D., Ludlow, L. H., Diemer, M. A., Scott, R. P. J., Gottfried, M., & Freitas, G. (2003). Development of the Conformity to Masculine Norms Inventory. Psychology of Men & Masculinity, 4(1), 3–25. https://doi.org/10.1037/1524-9220.4.1.3

Mardia, K. V. (1970). Measures of multivariate skewness and kurtosis with applications. Biometrika, 57(3), 519–530. https://doi.org/10.1093/biomet/57.3.519

Martino, S. C., Collins, R. L., Elliott, M. N., Strachman, A., Kanouse, D. E., & Berry, S. H. (2006). Exposure to degrading versus nondegrading music lyrics and sexual behavior among youth. Pediatrics (Evanston), 118(2), e430–e441. https://doi.org/10.1542/peds.2006-0131

McKinley, N. M., & Hyde, J. S. (1996). The Objectified Body Consciousness Scale: Development and validation. Psychology of Women Quarterly, 20(2), 181–215. https://doi.org/10.1111/j.1471-6402.1996.tb00467.x

Montemurro, B. (2003). Not a laughing matter: Sexual harassment as "material" on workplace-based situation comedies. Sex Roles, 48(9-10), 433–445. https://doi.org/10.1023/A:1023578528629

Moradi, B., Dirks, D., & Matteson, A. V. (2005). Roles of sexual objectification experiences and internalization of standards of beauty in eating disorder symptomatology. Journal of Counseling Psychology, 52(3), 420–428. https://doi.org/10.1037/0022-0167.52.3.420

Mussweiler, T., Rüter, K., & Epstude, K. (2004). The ups and downs of social comparison. Journal of Personality and Social Psychology, 87(6), 832–844. https://doi.org/10.1037/0022-3514.87.6.832

Muthen, B., & Kaplan, D. (1992). A comparison of some methodologies for the factor analysis of non-normal Likert variables: A note on the size of the model. British Journal of Mathematical and Statistical Psychology, 45(1), 19–30. https://doi.org/10.1111/j.2044-8317.1992.tb00975.x

Pew Research Center. (2015, April 09). Teenager, social media & technology overview 2015. https://www.pewresearch.org/internet/2015/04/09/teens-social-media-technology-2015/

Pew Research Center. (2021, April 07). Social Media Use in 2021. https://www.pewresearch.org/internet/2021/04/07/social-media-use-in-2021/

Ricciardelli, R., Clow, K. A., & White, P. (2010). Investigating hegemonic masculinity: Portrayals of masculinity in men’s lifestyle magazines. Sex Roles, 63(1-2), 64–78. https://doi.org/10.1007/s11199-010-9764-8

Rohlinger, D. A. (2002). Eroticizing men: Cultural influences on advertising and male objectification. Sex Roles, 46(3-4), 61–74. https://doi.org/10.1023/A:1016575909173

Rosen, L. D., Whaling, K., Carrier, L. M., Cheever, N. A., & Rokkum, J. (2013). The Media and Technology Usage and Attitudes Scale: An empirical investigation. Computers in Human Behavior, 29(6), 2501–2511. https://doi.org/10.1016/j.chb.2013.06.006

Shepherd, L. (2018). Responding to sexual objectification: The role of emotions in influencing willingness to undertake different types of action. Sex Roles, 80(1-2), 25–40. https://doi.org/10.1007/s11199-018-0912-x

Stankiewicz, J. M., & Rosselli, F. (2008). Women as sex objects and victims in print advertisements. Sex Roles, 58(7-8), 579–589. https://doi.org/10.1007/s11199-007-9359-1

Stevens, J. P. (2002). Applied multivariate statistics for the social sciences (4th ed.). Lawrence Erbaum Associates

Strahan, E. J., Lafrance, A., Wilson, A. E., Ethier, N., Spencer, S. J., & Zanna, M. P. (2008). Victoria's dirty secret: How sociocultural norms influence adolescent girls and women. Personality & Social Psychology Bulletin, 34(2), 288–301. https://doi.org/10.1177/0146167207310457

Tabachnick, B., G., & Fidell, L.S. (2007). Using multivariate statistics (5th ed.). Pearson Education, Limited.

Tebbe, E. A., Moradi, B., Wilson, E., Bell, H. L., Connelly, K., & Lenzen, A. (2021). Development and psychometric evaluation of the Sexual Minority Women's Sexual Objectification Experiences Scale. Journal of Counseling Psychology, 68(5), 501–514. https://doi.org/10.1037/cou0000547

Terry, G., & Braun, V. (2011). 'I'm committed to her and the family': Positive accounts of vasectomy among New Zealand men. Journal of Reproductive and Infant Psychology, 29(3), 276–291. https://doi.org/10.1080/02646838.2011.592976

Thompson, J. K., & Stice, E. (2001). Thin-ideal internalization: Mounting evidence for a new risk factor for body-image disturbance and eating pathology. Current Directions in Psychological Science: A Journal of the American Psychological Society, 10(5), 181–183. https://doi.org/10.1111/1467-8721.00144

Thompson, J. K., Berg, P. van den, Roehrig, M., Guarda, A. S., & Heinberg, L. J. (2004). Sociocultural Attitudes Towards Appearance Scale-3 (SATAQ-3): Development and validation. The International Journal of Eating Disorders, 35(3), 293–304. https://doi.org/10.1002/eat.10257

Twenge, J. M., & Martin, G. N. (2020). Gender differences in associations between digital media use and psychological well-being: Evidence from three large datasets. Journal of Adolescence (London, England.), 79(1), 91–102. https://doi.org/10.1016/j.adolescence.2019.12.018

Vandenbosch, L., & Eggermont, S. (2012). Understanding sexual objectification: A comprehensive approach toward media exposure and girls' internalization of beauty ideals, self-objectification, and body surveillance. Journal of Communication, 62(5), 869–887. https://doi.org/10.1111/j.1460-2466.2012.01667.x

Vandenbosch, L., & Eggermont, S. (2013). Sexualization of adolescent boys: Media exposure and boys' internalization of appearance ideals, self-objectification, and body surveillance. Men and Masculinities, 16(3), 283–306. https://doi.org/10.1177/1097184X13477866

Vandenbosch, L., Muise, A., Eggermont, S., & Impett, E. A. (2015). Sexualizing reality television: Associations with trait and state self-objectification. Body Image, 13, 62–66. https://doi.org/10.1016/j.bodyim.2015.01.003

Ward, L. M. (2016). Media and sexualization: State of empirical research, 1995-2015. The Journal of Sex Research, 53(4-5), 560–577. https://doi.org/10.1080/00224499.2016.1142496

Williams, B., Onsman, A., & Brown, T. (2010). Exploratory factor analysis: A five-step guide for novices. Australasian Journal of Paramedicine, 8(3), 1–13. https://doi.org/10.33151/ajp.8.3.93

Winship, J. (1987). Handling sex. In Betterton, R. (Ed.), Looking on: Images of femininity in the visual arts and media (pp. 25–41). Pandora.

Wright, P. J., & Tokunaga, R. S. (2015). Activating the centerfold syndrome: Recency of exposure, sexual explicitness, past exposure to objectifying media. Communication Research, 42(6), 864–897. https://doi.org/10.1177/0093650213509668

Zillmann, D., & Bryant, J. (2008) Selective exposure to communication. Routledge.

Tables

Table 1

Study 2: Exploratory Factor Pattern Matrix Loading and Cross Loading for the W-SOMES items

W-SOMES Factors and Items

Factor loading


1

2

3

Factor 1: Importance of Physical Appearance




34. How often have you noticed in Print/online articles and online comments, women are encouraged to attract romantic partners by improving their physical attractiveness?

.79

-.02

.07

32. How often have you noticed in Print/online articles and online comments, people are only interested in women if they are physically attractive?

.77

.01

-.01

29. How often have you noticed in Print/online articles and online comments, women are encouraged to do something to look more physically attractive (e.g., exercising, dieting, or wearing certain clothing)?

.75

.02

-.08

28. How often have you noticed in Print/online articles and online comments, people talk about the importance of physical appearance in women’s desirability?

.75

-.03

.09

33. How often have you noticed in Print/online articles and online comments, people talk about women’s main concern should be their appearance?

.74

.09

-.01

Factor 2: Sexualised Body Representation




22. How often have you seen in advertisements, female models’ bodies or sexuality are highlighted (e.g., body exposure, revealing clothes), while the product itself is less focused on?

-.02

.77

-.01

19. How often have you noticed in magazines and advertisements, female models pose in a sexually suggestive way?

.06

.73

.01

18. How often have you noticed in magazines and advertisements, female models wear revealing clothes, or expose their bodies?

.12

.70

-.06

21. How often have you noticed in advertisements, female models are represented as decorations, and their body or sexuality are used to sell the products?

-.04

.67

.11

20. How often have you noticed in magazines and advertisements, female models are featured with sexually suggestive facial expressions (e.g., flirtatious winking, licking lips)?

-.03

.61

.11

Factor 3: Body Evaluation




8. How often have you noticed in music videos, female models or music artists are looked at in a sexual way by others (e.g., staring, leering, gazing, ogling)?

-.07

.08

.77

4. How often have you noticed in TV programs and movies, female characters’ physical appearance is evaluated by other characters?

.10

-.06

.75

1. How often have you noticed in TV programs and movies, female characters’ bodies are looked at in a sexual way by other characters (e.g., staring, leering, gazing, ogling)?

-.03

.07

.70

12. How often have you heard the music lyrics mentioning the sexualised body or body parts of women?

-.06

.08

.68

27. How often have you seen on social media, people make inappropriate sexual comments about women’s bodies or their sexuality?

.19

-.04

.58

Note. = 340. The extraction method was principal axis factoring with an oblique (Promax) rotation. Factor loadings above .40 are in bold.

 

 

Table 2

Study2: Scale Correlations, Descriptive Statistics, Internal Consistency and Factor Intercorrelations for the W-SOMES              

Variable

1

2

3

4

1. W-SOMES Total





2. W-SOMES Importance of Physical Appearance

.84*


(.59)

(.61)

3. W-SOMES Sexualised Body Representation

.84*

.54*


(.69)

4. W-SOMES Body Evaluation

.86*

.57*

.64**


Cronbach’s α

.88

.85

.85

.92

M

3.71

3.79

3.71

3.63

SD

0.58

0.75

0.62

0.68

Scale range

1-5

1-5

1-5

1-5

Note. N = 340. Correlations in parentheses above the diagonal reflect EFA factor intercorrelations. W-SOMES Total = Women-Sexually Objectifying Media Exposure Total Score; W-SOMES Importance of Physical Appearance = Women-Sexually Objectifying Media Exposure Importance Physical Appearance subscale; W-SOMES Sexualised Body Representation = Women-Sexually Objectifying Media Exposure Sexualised Body Representation subscale; W-SOMES Body Evaluation = Women-Sexually Objectifying Media Exposure Body Evaluation subscale.

* p < .01, ** p < .001. 

 

Table 3

Study 2: Exploratory Factor Pattern Matrix Loading for the M-SOMES items

M-SOMES Items

Factor Loading

6. How often have you noticed in advertisements, male models are represented as decorations, and their body or sexuality are used to sell the products?

.80

7. How often have you noticed in advertisements, male models’ bodies or sexuality are highlighted (e.g., body exposure, revealing clothes), while the product itself is less focused on?

.76

5. How often have you noticed in magazines and advertisements, male models pose in a sexually suggestive way?

.74

4. How often have you noticed in magazines and advertisements, male models wear revealing clothes, or expose their bodies?

.73

2. How often have you noticed in music videos, male models or music artists wear revealing clothing, or expose their bodies?

.72

3. How often have you noticed in music videos, male models are represented as decorations, and their bodies and sexuality are used to attract audiences?

.66

Note. N = 100. The extraction method was principal axis factoring with an oblique (Promax) rotation.

 

Table 4

Study 3: Descriptive Statistics and Bivariate Correlations among Variables in Women

Variable

n

M

SD

1

2

3

4

5

6

7

8

9

10

11

12

1. W-SOMES Total

331

3.78

0.62












 

2. W-SOMES

Importance of Physical Appearance

331

3.80

0.71

.91***











 

3. W-SOMES 

Sexualised Body Representation

331

3.80

0.66

.89***

.71***










 

4. W-SOMES 

Body Evaluation

331

3.75

0.68

.92***

.76***

.74***









 

5. SATAQ3-IG

331

3.10

1.12

.29***

.27***

.22***

.31***








 

6. SOBBS

331

2.82

0.79

.26***

.23***

.20***

.29***

.73***







 

7. OBC-Body Shame

331

4.00

1.38

.22***

.20***

.17**

.22***

.65***

.73***






 

8. OBC-Body Surveillance

331

4.45

1.19

.24***

.20***

.16**

.28***

.70***

.76***

.64***





 

9. ISOS

331

2.18

0.73

.52***

.43***

.44***

.56***

.28***

.30***

.25***

.29***




 

10. Exposure to Media Portrays of 

Interpersonal Sexual Objectification

331

3.05

0.83

.66***

.58***

.55***

.68***

.30***

.28***

.24***

.25***

.56***



 

11. CME-2-AS

331

2.44

0.77

.25***

.18**

.21***

.31***

.14**

.13*

.100

.11*

.23***

.31***


 

12. Daily Media Usage

331

14.09

8.41

.26***

.23***

.19***

.28***

.19**

.24***

.20***

.21***

.27***

.31***

.30***

 

Note. W-SOMES Total = Women-Sexually Objectifying Media Exposure Total Score; W-SOMES Importance of Physical Appearance = Women-Sexually Objectifying Media Exposure Importance of Physical Appearance subscale; W-SOMES Sexualised Body Representation = Women-Sexually Objectifying Media Exposure Sexualised Body Representation subscale; W-SOMES Body Evaluation = Women-Sexually Objectifying Media Exposure Body Evaluation subscale; SATAQ3-IG= Internalisation General subscale of the Sociocultural Attitudes Towards Appearance Scale-3; SOBBS = Self-Objectification Beliefs and Behaviors Scale; OBC-Body Shame= Objectified Body Consciousness Body Shame Scale; OBC-Surv = Objectified Body Consciousness Body Surveillance Scale; ISOS = Interpersonal Sexual Objectification Scale; CME-2-AS= The Antisocial subscale of Content-based Media Exposure Scale 2.

*p< .05, **p < .01, ***p< .001.


Table 5

Study 3: Independent T-tests of differences in scores of the W-SOMES and M-SOMES between heavy media users and light media users

Variable

Heavy media user

Light media user

df

t

Cohen's d


n

M

SD

n

M

SD




W-SOMES Total

70

3.99

0.61

67

3.55

0.67

135

4.05***

.66

W-SOMES Importance of Physical Appearance

70

4.03

0.67

67

3.54

0.79

135

3.93***

.62

W-SOMES Sexualised Body Representation

70

3.93

0.67

67

3.63

0.73

135

2.53*

.41

W-SOMES Body Evaluation

70

4.02

0.68

67

3.47

0.68

135

4.65***

.79

M-SOMES

68

3.15

0.88

69

2.85

0.81

135

2.08*

.37

Note. W-SOMES Total = Women-Sexually Objectifying Media Exposure Total Score; W-SOMES Importance of Physical Appearance = Women-Sexually Objectifying Media Exposure Importance of Physical Appearance subscale; W-SOMES Sexualised Body Representation = Women-Sexually Objectifying Media Exposure Sexualised Body Representation subscale; M-SOMES = Men-Sexually Objectifying Media Exposure Scale. 

*p< .05, **p < .01, ***p< .001.


Table 6

Study 3: Hierarchical Regression Analysis for Incremental Validity of the W-SOMES Total Scores relative to Exposure to Media Portrayals of Interpersonal Sexual Objectification

Variable

∆R2

∆R2(F)

B

SE B

ß

t

Internalisation of sociocultural ideals of appearance







Model 1

.08

28.15***





ISOS



0.43

0.08

0.28

5.31***

Model 2

.03

10.60**





ISOS



0.25

0.10

0.17

2.61*

Exposure to Media Portrayals of

Interpersonal Sexual Objectification



0.28

0.09

0.21

3.26**

Model 3

.01

3.43





ISOS



0.21

0.10

0.14

2.09*

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.18

0.10

0.13

1.81

W-SOMES Total



0.24

0.13

0.13

1.85

Self-objectification







Model 1

.09

31.58***





ISOS



0.32

0.06

0.30

5.62***

Model 2

.02

6.45*





ISOS



0.22

0.07

0.21

3.25**

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.15

0.06

0.16

2.54*

Model 3

.01

1.18





ISOS



0.20

0.07

0.19

2.84**

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.11

0.07

0.11

1.48

W-SOMES Total



0.12

0.09

0.09

1.30

Body surveillance







Model 1

.09

31.1***





ISOS



0.48

0.09

0.29

5.58***

Model 2

.01

3.44





ISOS



0.37

0.10

0.23

3.58***

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.17

0.09

0.12

1.85

Model 3

.00

1.12





ISOS



0.34

0.11

0.21

3.22**

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.11

0.11

0.08

1.02

W-SOMES Total



0.15

0.14

0.08

1.06

Note.  = 331. Degrees of freedom for comparisons: Model 1 (1, 329); Model 2 (2, 328); Model 3 (3, 327). ISOS = Interpersonal Sexual Objectification Scale; W-SOMES Total= Women-Sexually Objectifying Media Exposure Total Score. ∆R= R squared change made by adding new predictors to the model; A significant ∆R2(F) indicates the difference made by adding new predictors to the model is significant; ∆R2(F) = F change of R squared change; = unstandardized regression weight; SE B = standard error of unstandardized regression weight; ß = standardised regression weight.

*< .05, **p < .01, ***p< .001. 


Table 7 

Study 3: Hierarchical Regression Analysis for Incremental Validity of the W-SOMES subscale scores relative to Exposure to Media Portrayals of Interpersonal Sexual Objectification

Variable

∆R2

∆R2(F)

B

SE B

ß

t

Internalisation of sociocultural ideals of appearance







Model 1

.08

28.15***





ISOS



0.43

0.08

0.28

5.31***

Model 2

.03

10.60**





ISOS



0.25

0.10

0.17

2.61*

Exposure to Media Portrayals of

Interpersonal Sexual Objectification



0.28

0.09

0.21

3.26**

Model 3

.01

1.77





ISOS



0.20

0.10

0.13

1.95

Exposure to Media Portrayals of

Interpersonal Sexual Objectification



0.16

0.10

0.12

1.57

W-SOMES 

Importance of Physical Appearance



0.12

0.14

0.08

0.89

W-SOMES 

Sexualised Body Representation



-0.10

0.14

-0.06

-.69

W-SOMES 

Body Evaluation



0.23

0.17

0.14

1.40

Self-objectification







Model 1

.09

31.58***





ISOS



0.32

0.06

0.30

5.62***

Model 2

.02

6.45*





ISOS



0.22

0.07

0.21

3.25**

Exposure to Media Portrayals of

Interpersonal Sexual Objectification



0.15

0.06

0.16

2.54*

Model 3

.01

1.82





ISOS



0.19

0.07

0.18

2.67**

Exposure to Media Portrayals of

Interpersonal Sexual Objectification



0.09

0.07

0.09

1.24

W-SOMES 

Importance of Physical Appearance



0.06

0.10

0.05

0.60

W-SOMES 

Sexualised Body Representation



-0.08

0.10

-0.07

-0.81

W-SOMES 

Body Evaluation



0.16

0.12

0.13

1.33

Body surveillance







Model 1

.09

31.1***





ISOS



0.48

0.09

0.29

5.58***

Model 2

.01

3.44





ISOS



0.37

0.10

0.23

3.58***

Exposure to Media Portrayals of

Interpersonal Sexual Objectification



0.17

0.09

0.12

1.85

Model 3

.02

2.08





ISOS



0.31

0.11

0.19

2.88**

Exposure to Media Portrayals of Interpersonal Sexual Objectification



0.07

0.11

0.05

0.61

W-SOMES Importance of Physical Appearance



0.00

0.14

0.00

0.03

W-SOMES Sexualised Body Representation



-0.21

0.15

-0.12

-1.40

W-SOMES 

Body Evaluation



0.40

0.18

0.23

2.26*

Note.  = 331. Degrees of freedom for comparisons: Model 1 (1, 329); Model 2 (2, 328); Model 3 (5, 325). ISOS = Interpersonal Sexual Objectification Scale; W-SOMES Importance of Physical Appearance = Women-Sexually Objectifying Media Exposure Importance of Physical Appearance subscale; W-SOMES Sexualised Body Representation = Women-Sexually Objectifying Media Exposure Sexualised Body Representation subscale; W-SOMES Body Evaluation = Women-Sexually Objectifying Media Exposure Body Evaluation subscale. ∆R= R squared change made by adding new predictors to the model; A significant ∆R2(F) indicates the difference made by adding new predictors to the model is significant; ∆R2(F) = F change of R squared change; = unstandardized regression weight; SE B = standard error of unstandardized regression weight; ß = standardised regression weight. 

*< .05, **p < .01, ***p< .001.

 

Table 8

Study 3: Internal Consistency and Intraclass Correlation Coefficients of the W-SOMES and the M-SOMES

Variable 

Time 1

Time 2

 

 


N

Cronbach’s α

N

Cronbach’s α

ICC

95% CI

W-SOMES Total

331

.94

288

.94

.78

[.71, .83]

W-SOMES Importance of Physical Appearance

331

.87

288

.86

.70

[.63, .76]

W-SOMES Sexualised Body Representation

331

.86

288

.86

.72

[.65, .78]

W-SOMES Body Evaluation

331

.84

288

.85

.74

[.66, .80]

M-SOMES

328

.87

299

.91

.69

[.62, .74]

Note. ICC estimates and their 95% confidence intervals were based on single measure, absolute agreement, 2-way mixed effects model. W-SOMES Total = Women-Sexually Objectifying Media Exposure Total Score; W-SOMES Importance of Physical Appearance = Women-Sexually Objectifying Media Exposure Importance of Physical Appearance subscale; W-SOMES Sexualised Body Representation = Women-Sexually Objectifying Media Exposure Sexualised Body Representation subscale; W-SOMES Body Evaluation = Women-Sexually Objectifying Media Exposure Body Evaluation subscale; M-SOMES = Men-Sexually Objectifying Media Exposure Scale. 95% CI = 95% confidence interval.


Table 9

Study 3: Descriptive Statistics and Bivariate Correlations among Variables in Men

Variable

n

M

SD

1

2

3

4

5

6

7

8

9

1. M-SOMES

328

2.90

0.80










2. SATAQ3-IG

328

2.58

1.01

.26***









3. SOBBS

328

2.46

0.79

.28***

.61***








4. OBC-Body Shame

328

3.31

1.25

.22***

.52***

.71***







5. OBC-Body Surveillance

328

3.68

1.29

.21***

.57***

.78***

.65***






6. ISOS

328

1.39

0.39

.34***

.30***

.38***

.27***

.26***





7. Exposure to Media Portrayals of

Interpersonal Sexual Objectification

328

2.71

0.80

.35***

.23***

.21***

.22***

.25***

.30***




8. CME-2-AS

328

2.39

0.72

.27***

.30***

.31***

.25***

.26***

.29***

.35***



9. Daily Media Usage

328

13.89

7.67

.18**

.13*

.16**

.19**

.12*

.29***

.26***

.26***


Note. M-SOMES = Men-Sexually Objectifying Media Exposure Scale; SATAQ3-IG = Internalisation General subscale of The Sociocultural Attitudes Towards Appearance Scale-3; SOBBS = Self-Objectification Beliefs and Behaviors Scale; OBC-Body Shame = Objectified Body Consciousness Body Shame Scale; OBC-Surv = Objectified Body Consciousness Body Surveillance Scale; ISOS = Interpersonal Sexual Objectification Scale; CME-2-AS = The Antisocial Subscale of Content-based Media Exposure Scale 2.

*p< .05, **p< .01, ***p< .001.


Table 10

Study 3: Hierarchical Regression Analysis for Incremental Validity of the M-SOMES relative to Exposure to Media Portrayals of Interpersonal Sexual Objectification

Variable

∆R2

∆R2(F)

B

SE B

ß

t

Internalisation of sociocultural ideals of appearance







Model 1

.09

33.04***





ISOS



0.79

0.14

0.30

5.75***

Model 2

.02

8.29**





ISOS



0.66

0.14

0.26

4.67***

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.20

0.07

0.16

2.88**

Model 3

.02

6.61*





ISOS



0.57

0.15

0.22

3.88***

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.15

0.07

0.12

2.09*

M-SOMES 



0.19

0.07

0.15

2.57*

Self-objectification







Model 1

.14

54.79***





ISOS



0.77

0.10

0.38

7.40***

Model 2

.01

3.95*





ISOS



0.71

0.11

0.35

6.49***

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.11

0.05

0.11

1.99*

Model 3

.02

7.06**





ISOS



0.63

0.11

0.31

5.63***

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.07

0.06

0.07

1.20

M-SOMES 



0.15

0.06

0.15

2.66**

Body surveillance







Model 1

.07

23.90***





ISOS



0.86

0.18

0.26

4.89***

Model 2

.03

12.14**





ISOS



0.67

0.18

0.20

3.69***

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.31

0.09

0.19

3.48**

Model 3

.01

2.71





ISOS



0.59

0.19

0.18

3.14**

Exposure to Media Portrayals of 

Interpersonal Sexual Objectification



0.27

0.09

0.17

2.91**

M-SOMES 



0.15

0.09

0.10

1.65

Note.  = 328. Degrees of freedom for comparisons: Model 1 (1, 326); Model 2 (2, 325); Model 3 (3, 324). M-SOMES = Men-Sexually Objectifying Media Exposure Scale. ∆R= R squared change made by adding new predictors to the model; A significant ∆R2(F) indicates the difference made by adding new predictors to the model is significant; ∆R2(F) = F change of R squared change; = unstandardized regression weight; SE B = standard error of unstandardized regression weight; ß = standardised regression weight.

*< .05, **p < .01, ***p< .001

Declarations

Competing interests: The authors declare no competing interests.