Avatar Creation: The Social Construction of “Beauty” in Second Life

Abstract

Rooted in the theory of Social Construction of Reality and informed by media portrayal of female beauty and virtual community research, this study examined how beauty is socially constructed by gatekeepers in Second Life. A content analysis of 360 still images of female avatars was conducted to understand the extent to the beauty types that appear in the virtual world. Findings suggest trendy and sex kitten/sensual beauty types were the most-portrayed beauty types. Most female avatars had the ideal body size and light colored skin. In addition, this study found a significant difference in beauty type among different types of products.

Keywords

theory of social construction of reality Second Life female beauty content analysis

What is beauty? How does one define what is beautiful? It is a term that bears many meanings. For some, beauty is about the feeling that comes from inside-having a pretty heart, mind, and soul. For others, it is about having confidence and letting one’s personality shine. Unfortunately, these conceptualizations of beauty bear no fruit for advertisers. For them, beauty means to look physically attractive as it is used as an indicator to drive sales for beauty-related companies (Xie & Zhang, 2013). Willem, Arauna, Crescenzi, and Tortajada (2010) suggest this may explain why women appear to be sex objects, in advertisements, that simply please the male gaze. As a result, female models appear to look like sex kittens (i.e., big boobs, big butts, slim waist, curves, and wears little clothing; Chan & Cheng, 2012). The everyday consumer consumes these images and is fed the beauty agenda advertisers set. Given the nature of the advertising industry, it is imperative to apprehend how this concept plays a role in socially constructed beauty standards.

While traditional advertisers have set the stage for what is considered “beautiful,” Second Life (SL), a virtual world, is perceived to present different options to its users. Developed in 2003, by Linden Labs, the purpose of SL is to give its users the option to live a SL (Brennen & dela Cerna, 2010; Huang, 2011). As part of their SL, they have the power to create an avatar that looks anyway users’ desire. Although it may seem that SL users hold immense power and can change how beauty is perceived with the click of a button, the reality is that the gatekeepers who create the avatars and the products that enhance avatar appearance are setting the beauty agenda. Bloustien and Wood (2013) contend SL gives its users the ability to re-create the avatar they want, but they must work within the parameters SL creates. This realization calls attention to the importance of understanding the role of the gatekeepers in this virtual world.

Behm-Morawitz and Mastro (2008) suggest that media impact on individuals’ social construction of reality, real-world gender-based attitudes, beliefs, and behaviors, even when people think they have behavioral agency and freedom. However, few researchers have examined the role gatekeepers’ plays in creating and reinforcing beauty portrayal in virtual communities. Moreover, among the emerging virtual communities research, scant attention has been given to the gender portrayal in SL. Thus, this study aims to address the gap by examining how beauty is socially constructed in SL by its gatekeepers. Specifically, this study focuses on how female avatars were featured in the SL Marketplace (SLM).

Theoretical Conceptualization

The Theory of Social Construction of Reality

Social construction of reality (SCR) suggests that meaning is unfixed (Berger & Luckmann, 1966) and that reality is socially constructed during communication (Wu, 2006). Within this theory, there are two fundamental concepts, reality and knowledge (Berger & Luckmann, 1966; Brabant, 2010). Reality refers to “a quality appertaining to a phenomenon that we recognize as having a being independent of our own volition (we cannot wish them away)” (Brabant, 2010, p. 222). Knowledge pertains to “the certainty that phenomena are real and that they possess specific characteristics” (Brabant, 2010, p. 223). When exploring the notions of reality and knowledge, Berger and Luckmann (1966) suggest people will never develop full awareness of these concepts because they are not forced to think about them. People take for granted the things that go on around them and the decisions that they make every day. It is difficult to know what can be justified as reality or knowledge because each person’s perspective is unique.

Berger and Luckmann (1966) posit mass media have the ability to shape a person’s perception of what constitutes knowledge and reality. They argue that the media constructs ideas through re-presentation of information. Through this process, they encourage the meanings of the dominant social groups, while discouraging opposing viewpoints from everyday people. Ibroscheva and Ramaprasad (2007) add to this contention by suggesting media exposure, the media source, and media content all impact perceived stereotypes. Put simply, stereotyping people is not a universal concept, but one that is socially constructed. As a source that is accessible by many, information confirming stereotypes can spread like a wildfire. If enough people are exposed to stereotypical content, then they are more likely to act upon those stereotypes, become solidified, and the meaning behind the stereotypes are reified overtime (Pickering, 1995).

Stereotypes become more prevalent in an online context; media users can now connect with anyone from around the world, which adds to the challenge of managing stereotypes. Ananthaswamy (2007) claims people now have the ability to construct their reality both on and offline. When they participate in virtual worlds, they post information that may or may not represent who they are (Zheng, Young, Wagner, & Brewer, 2009). For example, if online users choose to disclose photographs of themselves, they may be more apt to show pictures that will receive positive appraise from their network. This symbolic act of communication (Ellis, 2010) has the ability to create a cycle that promotes a socially constructed reality. If online users like the posted pictures, then the user may think she or he is physically attractive. In a world where gatekeepers control the standard of beauty, it is important to note that online users use the beauty industry’s standard of beauty to dictate their perception of it, which may impact whether images in a picture are liked.

Media Portrayal of Female Beauty

Conrad, Dixon, and Zhang (2009) discovered the notion of the beauty myth. The beauty myth suggests that there is a stereotypical beauty standard portrayed in the media. This standard is represented by Eurocentric features such as white skin, blond hair, blue eyes, as well as youth, and low body weight. Researchers suggest individuals with lighter colored skin are used more often in the media over those with dark colored skin (Conrad et al., 2009). Mbure (2009) was curious as to whether the stereotype remained within the media portrayal of African American beauty. She conducted a study of female model representation in six African-origin magazines and found that the majority of African American models had light skin and long hair. Even within the non-Caucasian community, there is heavy emphasis placed on the media’s construction of beauty. Supporting the contention that lighter colored skin is valued more highly is Xie and Zhang (2013). They conducted a content analysis on skin advertisements in women’s magazines (Cosmopolitan and Vogue) from China and the United States. Not surprisingly, they found that light colored skin was shown more often in Chinese advertisements, but were surprised to find that U.S. magazines did not show tan or dark models nearly as often as they thought. In fact, the majority of them were coded as having a very fair or fair complexion. Although perceptions of beauty and skin color differ among Western and Eastern countries, what is important to mention is that the gatekeepers from both countries and different races promote the same agenda.

Studies also suggest that the media portrays women as sex objects or “decoration” in the media. Lin and Yeh (2009) support this claim by suggesting the majority of women used in magazine advertisements are at least partially nude. By showing off their bodies, Emons, Wester, and Scheepers (2010) posit that partial nudity can impact whether a person is perceived to be attractive. Even when women are not partially nude, their bodies still remain a focal point. Morris (2006) posits when female models are depicted in professional roles in advertisements, they are not perceived to be as serious as men. Although males were photographed wearing a suit a tie, females were photographed in less professional, tight-fitted clothing (Signorielle & McLeod, 1994). This type of exposure continues to reify the notion that females are not as qualified to work in professional jobs as their male counterparts. In order to be noticed, women must be willing to showcase their bodies promiscuously.

Researchers also suggest that regardless of how a woman wants to be perceived, one of the most powerful weapons any woman can own is beauty (Joseph, 1982; Patzer, 1985). Beauty has a strong, pervasive impact on people’s lives. A beautiful woman can gain power, social influence, and perceived intelligence (Patzer, 1985). Joseph (1982) investigated the role of media and socialization. Through his analysis, he found that attractive females were perceived to be more credible and likeable than their unattractive counterparts. In addition, Benoy (1982) contends attractive females were perceived to have positive attributes and desirable traits such as sexual warmth, sensitivity, strength, and sensitivity. Moreover, when females embody desirable traits, they lead happier and more fulfilling lives than unattractive ones. What this demonstrates is that a woman’s worth is tied to her looks rather than her mind.

Media are instrumental in shaping and framing standards of beauty. They control what people see and how they perceive beauty. Solomon, Ashmore, and Longo (1992) conducted an important study in gendered media scholarship. They discovered that beauty is not seen as a singular concept. Rather, the researchers found that eight different beauty types exist: classic (perfect physical, especially facial features), cute (child-like physical features and/or attire), sex kitten (sexual looks, overt and youthful), sensual (sexual looks), girl next door (natural, unmade-up appearance and simple attire), exotic (non-Causasian), feminine (a soft and/or romantic look), and trendy (an off-beat look, perhaps flawed or asymmetrical, in contrast of a classic beauty type). Now, beauty is no longer a vague thought in the mind of consumers. Consumers can now indicate which beauty type is being portrayed and how they identify with the beauty types.

Englis, Solomon, and Ashmore (1994) examined how Solomon, Ashmore, and Longo’s (1992) beauty types were distributed across different media platforms. They found that beauty was not evenly distributed; it depended on the type of magazine. Through their examination, they found that fashion magazines used the exotic beauty type most often, while youth magazines focused on classic and girl next door beauty types. Goodman, Morris, and Sutherland (2008) conducted a study on women’s emotional responses to different beauty types presented in magazine advertisements. They found that women had the most positive emotional response when they saw a classic beauty type; the most negative emotional response was to sensual/exotic beauty type. Through the researchers’ exploration of the different beauty types, it is apparent that specific beauty types are recognizable. Regardless of beauty type that is portrayed, the media gatekeepers still control the standard of beauty.

Virtual Communities and SL

A virtual community is defined as a community founded intentionally by people who share a set of similar interests, often revolving around noncomputer mediated communication venues (McQuail, 2010). It is a computer generated version of a real-life setting that can be experienced in a variety of formats (e.g., individualized computer video game, 3D game, or as a collaborative learning environment) (O’Brien & Levy, 2008). They can be formed by any number of individuals who have Internet access (McQuail, 2010). Defining characteristics of virtual communities are interactivity, three-dimensionality, and real-time feedback. Users feel a sense of presence when they are in a virtual community. This means that users feel they are inside of the computer world rather than just looking at a video game display (O’Brien & Levy, 2008). People are able to communicate among those who would never be able to communicate in person and it is a revolutionary way to explore worlds that are too far away, too small, too experimental, or too dangerous for ordinary access (Harrison, 2009; O’Brien & Levy, 2008). In other words, real-life communities are often hard to enter; virtual communities are more open and accessible (McQuail, 2010).

A virtual community where users can participate in is SL. Developed in 2003 by Linden Labs (Huang, 2011), the three-dimensional world it became a place where people could live a SL rather than play a game that has predetermined goals. SL’s technology allows users to build anything imaginable, explore various lands and from seaside towns to educational institutions, speak with other users, and purchase products (Atkinson, 2008). As users explore different aspects of SL, they may come into contact with organizations who have integrated their services into this world, a few of which include implementing learning activities for distance education (Hartley, Ludlow, & Duff, 2015), simulating medical activities to show patients what they will experience before undergoing procedures (Best & Butler, 2013; Hudson, Taylor, Kozachik, Shaefer, & Wilson, 2015), reducing speech anxiety through virtual world workshops (Guzel & Aydin, 2014), and Starwood brand hotels building virtual hotels for users to stay in (Jana, 2006; Lynch, 2006). Freedom to explore anything in SL and the possibility of coming into contact with outside organizations suggests that it is place where its users must figure out what to do as they immerse themselves in SL, just like they would in real life.

Regardless who a user comes into contact with or where the contact happens, users represent themselves in SL by creating an avatar (Barker, 2008). The creation of avatars in cyberspace is not incidental; it is an intentional act that is enacted by the SL user (Huang, 2011). Since there are no boundaries or limits, with a few clicks of the mouse, anyone can be who they want to be (Ananthaswamy, 2007; Brookey & Cannon, 2009). Lee and Park (2009) state avatars are highly customizable, as describe below:

Your avatar can look any way you want it to, up to the limitations of your equipment. If you’re ugly, you can make your avatar beautiful. If you’ve just gotten out of bed, your avatar can still be wearing beautiful clothes and professionally applied makeup. (p. 6)

By creating a highly customizable avatar, users can create an avatar that complements the social interactions that take place with other users in the networked setting. Their creations can be based on what they look like in real life or based on unrealistic standards of beauty that are unattainable in real life (i.e., hair, structured face, and a body that looks like a 1950’s Barbie doll). Thus, the appearance of an avatar is a reflection of the person who created it (Bélisle & Bodur, 2010).

Ayiter (2010) found that just as people choose their self-representations in virtual environments, their self-representations shaped their behaviors in turn. These changes happen not over hours or weeks, but within minutes. In the real world, people may be used to the disappointments that come from the inability to achieve a certain look (Rak, 2009). When people become immersed in SL, they quickly realize that a large proportion of the avatars are physically fit, fake looking (heavy makeup and hair extensions), and enhance the avatar (big breasts and big butt) to their satisfaction. Essentially, these users experience what it is like to look a certain way, even if the experience is virtualized (Hall, 2008).

Within this online community, Martey and Consalvo (2011) contend that certain looks may be more desirable than others. In their exploration of avatar appearance with those who had SL accounts, they discovered that while users perceived to have freedom in appearance, users tended to create avatars that complemented the societal norms of those they interacted with on SL. If users wore the right clothes, then they would be perceived as likeable and were more likely to fit in. Interestingly, a female avatar who dressed properly (i.e., adorned in accessories and wore skimpy clothing) and had light skin color was perceived to be likeable by those in their SL community. Their findings revealed that in a space where SL users perceive having the freedom to look any way, traditional media gatekeepers have impacted their social construction of beauty offline and their beauty perception may have been brought onto SL.

As a large online virtual world, there are many opportunities to create and maintain impressions with other users. Previously, there were 1.1 million users. However, in 2015, there were 900,000 active users. Although the number of users has decreased, they generate a profit of US$60 million dollars per year (real-world money; Weinberg, 2015). Although the community of users have been able to sustain this virtual world, SL may have been ahead of its time initially. However, virtual reality and 3D technology is increasing in popularity. Weinberg suggests the increasing popularity in virtual reality is attracting companies like Facebook, HTC Sony (Weinberg, 2015) and Google Glass (Au, 2013) to purchase products that support 3D and virtual reality products. The rise (Weinberg, 2015) and advancement (Au, 2013) of 3D and virtual technology signals a shift in what consumers want to see. Despite coming into the market sooner than it should have, SL has the opportunity to capitalize on this technological transition. Thus, this gives them the leverage they need to bring in more users and create more avatars and products to enhance avatar appearance.

Literature on the Theory of the Social Construction of Reality focuses on how mass media plays a central role in socially constructing meanings of beauty; little attention has been given to the creation of avatars in SL. In addition, few researchers have studied gender portrayal in SL. As a virtual community that creates the perception that SL users have freedom to look any way they want, it is important to understand how the gatekeepers control what SL users have access to for creating their avatars, which can impact how much agency users perceive they have when they create their avatars. Researchers suggest that constant exposure to beauty stereotypes can cause one to become angered, have high levels of anxiety, and possibly depression (Robinson, Callister, Clark, & Phillips, 2009), which can impact how they choose to look and interact with other users in SL. There is clearly a need to examine the issue of socialized beauty construction in SL. Thus, this study aims to examine how beauty is socially constructed in SL by its gatekeepers. This focus serves as a guide for the following research questions:

RQ1: Which beauty type is portrayed most often for creating a female avatar in SL?

RQ2a: Is there a significant difference in beauty type between avatars with different skin color?

RQ2b: Is there a significant difference in beauty type among avatars with different body size?

RQ2c: Is there a significant difference in beauty type among avatars with different amount of coverage?

RQ2d: Is there a significant difference in beauty type among different types of products?

Method

This study conducted a content analysis to examine how beauty is socially constructed in SL by the SL gatekeepers. Wrench, Thomas-Maddox, Richmond, and McCroskey (2008) suggest a content analysis can “summarize visual images, characterizations, and other messages that are sent” (p. 244). Thus, a content analysis of 360 still images of full-bodied, female avatars in SL was conducted in January 2012. The unit of analysis for this study was full-bodied, still images of female avatars. Only female avatars were selected because this study is focused on female beauty portrayal.

Data for this study were collected via the SLM, which is the universal location for all SL residents to purchase products for creating an avatar (Huang, 2011). Three categories were selected for coding: (a) avatar accessories, (b) avatar apparel, and (c) avatar appearance. These categories were selected because are the only ones that allow residents to purchase items for creating a female avatar. The best-selling items were selected for coding because these are the items people purchased most often. Also, a general maturity level was selected for coding (used to determine how promiscuous an avatar looks; others maturity levels in the Marketplace available to select were moderate and adult). According to Linden (2015), a general maturity level is defined as one that does not display products for purchase that display full nudity of an avatar and promote violence. In addition, it was selected because a general maturity level provided the most number of results and the widest range of products that were available to purchase. Overall, the first 120 images of full-bodied, still, female avatars from each of the three categories (accessories, apparel, and appearance) were selected for coding. To achieve a medium effect size (0.3) with an alpha level of .05 and a projected power of .8 with a df of 3, an a priori calculation through g power suggested a sample size of 120 per group.

This study coded the following variables: beauty type, skin color, body size, coverage, and product type. To code beauty type, Solomon, Ashmore, and Longo’s (1992) study on multiple beauty types was utilized (their study included the following beauty types: classic, cute, exotic, girl next door, feminine, sex kitten, sensual, and trendy). From the eight they used, this study used a total of four beauty types, which were classic, cute/girl next door, sex kitten/sensual, and trendy. The feminine beauty type was eliminated because all avatars were posed in a feminine way. The exotic beauty type was excluded because it could have been easy to combine it with any beauty type, which would make intercoder reliability impossible to achieve at an acceptable level. In addition, based on their study, the beauty types cute and girl next door portrayed the same physical features as did sex kitten and sensual beauty types. Therefore, the researcher combined cute/girl next door and sex kitten/sensual beauty types.

For this study, classic beauty was defined as an avatar wearing clothing and accessories that displayed a classic or elegant look and displayed classy, formal attire and accessories. The cute/girl next door beauty type was defined as wearing clothing that looked outdoorsy, athletic, childish (i.e., cartoon pajamas), and/or displayed casual attire (i.e., jeans and tennis shoes together). Sex kitten/sensual beauty type was defined as wearing sexy attire that was revealing or very tight. These avatars could have been dressed in normal clothes, but posed in an unnatural, uncomfortable way. The trendy beauty type was defined as displaying faddish clothes (clothing items that would be out of style after one season) and oversized accessories.

In addition, skin color was coded as light or dark skin. Light skin was defined as porcelain, light ivory, sun beige, warm beige, and nude. Dark skin was defined as light cocoa, caramel, honey beige, cocoa, tawny, tan, espresso, and nut brown (Covergirl, 2017; Maybelline, 2017; L’Oréal Paris, 2017). Body size was categorized as thin, average, and overweight. If an avatar had no curves, the avatar was coded “thin.” If an avatar had excess body fat, stomach rolls, and enlarged thighs, the avatar was coded as “overweight.” If the avatar was not considered thin or overweight, it was coded as “average.” Coverage was coded as covered a lot, covered modestly, and covered very little. A lot of coverage was defined as when an avatar showed none to less than half of the total body skin, looked conservative, and had no inappropriate body parts exposed. Moderate coverage was defined as when an avatar showed about half of the total body skin, wore a bra and pants set at the same time and it did not look like she was naked. Little coverage was coded when an avatar exposed more than half of the total body skin, wore skimpy lingerie or a skimpy outfit, and had private parts fall out of the clothing.

Finally, product type was coded as accessories, clothing, lingerie, and shoes. Clothing was defined as selling shirts, pants, panty hose, dresses, skirts, shorts, or a one-piece outfit. Lingerie was defined as underwear, braziers, swimsuits, sexy nighttime outfits, strap ups, or any clothing item used for mature activities. Shoes were defined as selling gothic boots, sexy boots, high heels, low heels, tennis shoes, sandals, low tops, high tops, platforms, and wedges. Accessories were defined as selling any other item, such as hair, nails, jewelry, a body enhancement, makeup, wigs, and a complete avatar package. The researcher completed all of the coding for this study. All coded images were downloaded to a computer hard drive for the purpose of coding and intercoder reliability test to ensure consistency.

Prior to coding, a pilot study was conducted to test for intercoder reliability. For the pilot study, 20% of the sample (72 still images of female avatars, 24 per category) was coded. Intercoder reliability was conducted by three female coders who had previous exposure to SL. All coders were trained, and the coding book and examples were provided. After the training process was complete, each coder coded a total of 72 female avatars independently (i.e., 20% of the sample). Coders were instructed to code the avatars based on what clothes and accessories they were wearing and not by their facial features. The Cohen’s Kappa results indicated that there was a high level of reliability on the coding instrument and among coders. Intercoder reliability was as follows: body size (1.00), product type (.97), coverage (.93), skin color (.93), and beauty type (.91). Once the pilot study was finished, the same four coders coded 360 images. All categories produced 1.00 intercoder reliability. After this process was finished, a total of 360 new images were used for this study. For these images, the researcher coded all images. In addition, to ensure consistency for the study, each coder from the pilot study coded 90 total images (25% of the total sample for each coder). Each coder and the researcher maintained 100% reliability for the final set of coded images.

Data Screen and Analysis

The researcher used Frequency and Explore in SPSS for data screening. Frequency tests were used to answer RQ1. Chi-square tests were conducted to examine the differences in beauty type to answer RQ2a to RQ2d (i.e., Is there a significant difference in beauty type between avatars with different skin color? with different body size? with different amount of coverage? and among different types of products?).

Results

A total of 360 still images of full-bodied, female avatars in SL were coded to examine how beauty is socially constructed in SL. Overall, this study found that trendy beauty was the most often used beauty type (151 avatars), followed by sex kitten/sensual (136 avatars), cute/girl next door (40 avatars), and classic (33 avatars). In terms of skin color, more than 80% of the avatars coded in the study (293) had light skin, while only 19% (67) of the avatars featured dark skin. In addition, almost all of the female avatars had an average body size-358 out of 360 coded avatars, while two were overweight, and zero were thin. In terms of the body coverage, 139 avatars had little coverage, followed equally by a lot of coverage (111 avatars) and moderate coverage (110 avatars). Furthermore, accessories (238) were the most purchased products to create a female avatar in SL, followed by clothing (95), lingerie (23), and shoes (4).

To answer RQ2a, a chi-square test was conducted to see whether there was a significant difference in beauty type between avatars with different skin color. The results indicate that the majority of the avatars with both light skin and dark skin featured the trendy beauty type, followed by sex kitten/sensual, cute/girl next door, and classic (see Table 1). Specifically, among the avatars with light skin, 41% featured the trendy beauty type, 37% were sex kitten/sensual, 12% were cute/girl next door, while only 10% appeared to be the classic beauty type. Among avatars with dark skin, 48% were the trendy beauty type, 43% appeared to be sex kitten/sensual, 8% were cute/girl next door, and only 1% featured the classic beauty type. The chi-square test suggests that there was no significant difference in beauty type between avatars with different skin color (χ² = 7.60, d = 3, p = .055).

Table 1.

Differences in Beauty Type Between Avatars With Different Skin Color.

Beauty type	Dark skin		Light skin
Beauty type	Frequency	%	Frequency	%
Classic	1	1	32	10
Cute/girl next door	5	8	35	12
Sex kitten/sensual	29	43	107	37
Trendy	32	48	119	41
Total	67	100	293	100

Note. χ² = 7.600, d = 3, p = .055.

RQ2b was concerned with the difference in beauty type among avatars with different body size. The study found that the most often-used beauty type among people with average body size was trendy, followed by sex kitten/sensual, cute/girl next door, and classic (see Table 2). Specifically, 42% of the avatars with average size appeared to be the trendy beauty type, 38% featured sex kitten/sensual, 11% were cute/girl next door, and 9% showed the classic beauty type. Between the two overweight avatars, one showed cute/girl next door, while the other was sex kitten/sensual. The chi-square test result was not significant (χ² = 3.85, d = 3, p = .279), which suggests that there was no significant difference in beauty type among avatars with different body size.

Table 2.

Differences in Beauty Type Among Avatars With Different Body Size.

Beauty type	Average		Overweight
Beauty type	Frequency	%	Frequency	%
Classic	33	9	0	0
Cute/girl next door	39	11	1	50
Sex kitten/sensual	135	38	1	50
Trendy	151	42	0	0
Total	358	100	2	100

Note. χ² = 3.85, d = 3, p = .279.

RQ2c was concerned with the difference in beauty type among avatars with different amount of body coverage. Results indicate that the majority of the avatars with a lot of coverage featured the trendy beauty type (54%), followed by classic (28%), cute/girl next door (15%), and sex kitten/sensual (3%) (see Table 3). The majority of the moderate covered avatars were also the trendy beauty type (61%), followed by sex kitten/sensual (24%), cute/girl next door (14%), and classic (2%). Not surprisingly, the avatars with little amount of body coverage featured the sex kitten/sensual beauty type most frequently (77%), followed by trendy (17%), and cute/girl next door (6%). None of the 139 avatars with little body coverage featured a classic beauty type. The majority of the classic beauty type were covered a lot, the majority of the cute/girl next door beauty type were covered a lot or moderately covered, the sex kitten/sensual beauty type were covered little most of the time, and the trendy beauty type was covered a lot of moderately covered. The chi-square test result was statistically significant (χ² = 199.84, d = 6, p < .0001). There was a significant difference in beauty type among avatars with different amount of body coverage.

Table 3.

Differences in Beauty Type Among Avatars With Different Body Coverage.

Beauty type	A lot of coverage		Little coverage		Moderate coverage
Beauty type	Frequency	%	Frequency	%	Frequency	%
Classic	31	28	0	0%	2	2
Cute/girl next door	17	15	8	6%	15	14
Trendy	60	54	24	17%	67	61
Total	111	100	139	100%	110	100

Note. χ² = 199.84, d = 6, p < .0001.

RQ2d was concerned with the difference in beauty type among different product types. This study found that the majority of the products available for purchase were for the sex kitten/sensual beauty type (105), followed by trendy (86), cute/girl next door (31), and classic (16) (see Table 4). Among accessories, sex kitten/sensual and trendy were the most frequently used beauty types (44% and 36%, respectively), followed by cute/girl next door (13%) and classic (7%). Among all the clothing items coded, the majority of clothing was purchased for the trendy beauty type (65%), followed by classic (18%), sex kitten/sensual (9%), and cute/girl next door (8%).

Table 4.

Differences in Beauty Type Among Different Types of Products.

Beauty type	Accessories		Clothing		Lingerie		Shoes
Beauty type	Frequency	%	Frequency	%	Frequency	%	Frequency	%
Classic	16	7	17	18	0	0	0	0
Cute/girl next door	31	13	7	8	2	9	0	0
Sex kitten/sensual	105	44	9	9	21	91	1	25
Trendy	86	36	62	65	0	0	3	75
Total	238	100	95	100	23	100	4	100

Note. χ² = 79.68, d = 9, p < .0001.

A total of 91% of all the lingerie products available for purchase displayed the sex kitten/sensual beauty type. The only other beauty type shown in the lingerie category was cute/girl next door (9%). In terms of shoes, 75% of shoes showed the trendy beauty type, while 25% showed the sex kitten/sensual beauty type. The chi-square test was statistically significant (χ² = 79.68, d = 9, p < .0001). It suggests that there was a significant difference in beauty type among different types of products.

Overall, in terms of classic beauty, only accessories and clothing products used this beauty type. For cute/girl next door, the majority of the products purchased were accessories and clothing. The sex kitten/sensual beauty type showed accessories most often. In addition, out of the 23 lingerie items coded, 21 showed the sex kitten/sensual beauty type. In terms of trendy beauty, the majority of items available for purchase featuring this beauty type were accessories. Also, 65% of clothing items and 75% of shoes showed a trendy beauty type.

Discussion

This research examined how beauty is socially constructed in SL by its gatekeepers. Overall, findings suggest that stereotypical beauty types are portrayed in the SLM. Trendy and sex kitten/sensual beauty types were the most-portrayed beauty types. Almost all the female avatars had a perfect body size and more than 80% of the avatars coded in this study had light skin. In addition, there were significant differences in beauty type among avatars with different amounts of body coverage among different types of products. However, no significant difference was found in beauty type between avatars with different skin color nor was there a significant difference in beauty type among avatars with different body sizes.

Interestingly, this study found that almost 70% of all products sold that were coded were accessories. Because residents can create any avatar they desire, accessories, compared with clothing, lingerie, or shoes, can provide residents with the “ultimate” look that would be impossible or too costly to achieve in the real life. For example, while a person would have to go through surgery or begin a weight loss program if she or he wanted to look thinner, a resident in SL can simply purchase a different body size (e.g., thin body shape) with the click of a button. Also, in the SLM, the most expensive jewelry item was L$2,499, which is equivalent to a little over US$10.00. Thus, residents can give their avatars the “luxurious” look which many of them desire, but cannot afford outside of the virtual world.

Moreover, this study found no significant difference in beauty types among avatars with different skin color. This could be seen as confirmation that SL users prefer Eurocentric physical features over other types. As suggested by previous researchers (e.g., Conrad et al., 2009; Mbure 2009), lighter skin models were used more often in advertising and were preferred by the general public. This may explain why 81% of the coded avatars had light skin. This, ultimately, shows that the gatekeepers rely on and reinforce the stereotype that having lighter colored skin is what is considered to be beautiful.

In SL and the Marketplace, SL users can create any avatar they desire. However, this study found that the average body shape was still dominant in SL. In fact, all but two avatars coded in this study had an average body shape. This finding supports the notion that a thinner body type is considered beautiful. As users see this body type over time, this beauty ideal becomes “reified” within the mind (Berger & Luckmann, 1966). Thus, when people are in this virtual world, they already have an idea of what is considered to be an attractive body type. The gatekeepers creating products and showcasing these types of models continue to perpetuate body stereotypes in the Marketplace.

Trendy beauty was the most often portrayed beauty type in SL. It was also the dominant type used for “selling” fashionable clothing, oversized accessories, and dramatic hairstyles in the SLM. This suggests virtual communities are using advertising tactics which are applied in real-world applications. By using avatars which are the most popular to “model” or “sell” particular accessories, the SLM is relying on the users to go with what is popular, rather than be independent in their fashion decisions. Essentially, the propensity for users to follow the majority whether it is in gender portrayal or accessory purchasing suggests online personas are more similar to offline personas than originally thought.

When people access the virtual world, they seek to create an avatar that is with a perfect shape, fake-looking, and seek any way to enhance the avatar to their satisfaction. Essentially, people want to experience what it is like to look a certain way, even if the experience is virtualized (Hall, 2008). Cooke (2010) says the ability to change one’s appearance, with massive possibilities, in a short amount of time showcases that “anything is possible” in the virtual world (p. 396). In SL, people can create any avatar they envision or desire, without the costs associated with changing one’s appearance in real life. In a matter of minutes, a SL user can display sex kitten/sensual beauty type characteristics and change to trendy beauty type characteristics in the next minute. Having this freedom allows users to experiment with and explore beauty type options for a fraction of the cost.

In SL, users from all over the world join together to create their own economy; part of the process involves creating an avatar. Given the options presented in the game, users strategically created their avatars based on the Western ideals of beauty. Regardless of which beauty type was selected by users, the stereotypes employed in magazines, movies, and other media platforms were employed in the virtual world. Like other media avenues, women are still depicted as sex objects, noted primarily for their looks. Even though people can choose to look anyway they desire, they still reinforce gender stereotypes. Such stereotypes created a skewed view of what Westerners look like and how they behave. However, as long as new media platforms are created and/or enhanced, there will be possibilities for improvement and change. In the meantime, it gives researchers the opportunity to explore virtual phenomenon that would be impossible to explore otherwise.

This study was not without limitations. First, although this study provides important insights in understanding beauty portrayal in SL, a content analysis cannot provide in-depth understanding of how and why certain beauty types are selected over other ones. It is impossible to explore how SL players use apparel or accessories for their avatars, if and how users choose to adjust their avatar’s image based on their projection of beauty at a given moment, and if users portray different beauty types with multiple SL accounts. Future studies should institute different methods such as interviews, netnography, SL focus groups, or surveys to provide more accurate and in-depth information and explanation in understanding the social construction of beauty in the virtual world.

This study analyzed the SLM. Although it is a methodological step in the right direction, Linden Labs had the control over the standard of beauty. As Biu (2010) suggests, gatekeepers have the power to control the world. People who work for the Marketplace ultimately dictate how beauty is perceived in the Marketplace and in SL. Within the SLM, images were shown in an attempt to get SL users to purchase their products. Coding static images provided a unique perspective into understanding the social construction of beauty in SL, but it cannot explain the real-time decisions SL users make. Future research could code avatars in their native environment. By capturing the avatars in their native environment, it will provide richer sampling and analysis of the data. Moreover, this study coded 360 still images of full-bodied, female avatars in the SLM. Future studies could use a larger sample and focus on word choice that accompanies the product being sold in addition to still images. In addition, the same study could be used on still images of full-bodied, male avatars. Instead of examining different female beauty types, researchers could examine the social construction of male images. Furthermore, price and product reviews may impact the sample of this study. Thus, future research is encouraged to examine the relationship between price and product reviews and people’s choices on beauty type in SL. Moreover, data for this study were collected in 2012. In the case of many research studies, this data would no longer be timely. However, Yan and Bissell (2014) suggest that North American and European beauty standards will continue to dominate how models are portrayed and how they showcase products for many years. After revisiting the SLM in 2017, an informal assessment indicated that starkly similar results would have been generated today. Future researchers should examine the SLM longitudinally to determine the degree to which North American and European beauty standards still dominate the virtual world.

Despite the limitations, this study examined how beauty is socially constructed in SL by its gatekeepers. Ultimately, it suggests that gender stereotypes are reified and perpetuated in the SLM, even though people think they have the freedom to create any avatar. While great strides have been made to create gender equality, gender stereotypes are being reinforced in SL. By examining female avatar construction in the Marketplace, this study yields insights for understanding the social construction of avatar creation and gender norms within the Marketplace and SL. Furthermore, it provided understanding, coherence, and value to communication and gender inquiry, and will lead to advance the Theory of Social Construction of Reality in this dynamic new media environment.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biography

Hailey L. Mills, MA, received her BA in communication from the University of Mount Union and MA from the University of Akron. As a PhD student at Ohio University, she investigates the ways in which individuals manage the messages they put on social media via social media silencing and the impact it has on people.

References

Ananthaswamy

(2007). A life less ordinary offers far more than just escapism. New Scientist, 195(2618), 26-27.

Atkinson

(2008). Inside linden lab. Tech Trends: Linking Research and Practice to Improve Learning, 52(2), 18-21.

W. J.

(2013). Second life turns 10: What it did wrong, and why it may have its own second life. Retrieved from https://gigaom.com/2013/06/23/second-life-turns-10-what-it-did-wrong-and-why-it-will-have-its-own-second-life/

Ayiter

(2010). Embodied in a metaverse: “Anatomia” and “body parts.” Technoetic Arts: A Journal of Speculative Research, 8, 181-188.

Barker

(2008). Presentation of the virtual beyond-self on cyber stage: Real, constructed, staged, and/or masked? Communicatio: South African Journal for Communication Theory and Research, 34, 189-209.

Behm-Morawitz

Mastro

D. E.

(2008). Mean girls? The influence of gender portrayals in teen movies on emerging adults’ gender-based attitudes and beliefs. Journalism & Mass Communication Quarterly, 85, 131-146.

Bélisle

J. F.

Bodur

H. O.

(2010). Avatars as information: Perception of consumers based on their avatars in virtual worlds. Psychology & Marketing, 27, 741-765.

Benoy

J. W.

(1982). The credibility of physical attractive communicators: A review. Journal of Advertising, 11(3), 15-24.

Berger

P. L.

Luckmann

(1966). The social construction of reality. Garden City, NY: Doubleday.

10.

Best

Butler

(2013). Second Life avatars as extensions of social and physical bodies in people with myalgic encephalomyelitis/chronic fatigue syndrome. Continuum: Journal of Media & Cultural Studies, 27, 837-849. doi:10.1080/10304312.2013.794190

11.

Biu

(2010). How online gatekeepers guard our view-News portals’ inclusions and ranking of media and events. Global Media Journal: American Edition, 9(16), 1-41.

12.

Bloustien

G. F.

Wood

(2013). Face, authenticity, transformations, and aesthetics in Second Life. Body & Society, 19(1), 52-81. doi:10.1177/1357034X12462250

13.

Brabant

(2010). Death: The ultimate social construction of reality. Omega: Journal of Death and Dying, 62, 221-242.

14.

Brennen

dela Cerna

(2010). Journalism in second life. Journalism Studies, 11, 546-554.

15.

Brookey

R. A.

Cannon

K. L.

(2009). Sex lives in second life. Critical Studies in Media Communication, 26, 145-164.

16.

Chan

Cheng

(2012). Portrayal of females in magazine advertisements in Hong Kong. Journal of Asian Pacific Communication, 22(1), 78-96. doi:10.1075/japc.22.1.05cha

17.

Conrad

Dixon

Zhang

(2009). Controversial rap themes, gender portrayals and skin tone distortion: A content analysis of rap music videos. Journal of Broadcasting & Electronic Media, 53, 134-156.

18.

Cooke

(2010). De-inter-facement. Convergence: The International Journal of Research into New Media Technologies, 16, 395-410.

19.

Covergirl. (2017). Retrieved from http://www.covergirl.com/en_us/beauty=products/face-Makeup/foundation-makeup/lean-liquid-foundation

20.

Ellis

(2010). Be who you want to be: The philosophy of Facebook and the construction of identity. Screen Education, 58, 36-41.

21.

Emons

Wester

Scheepers

(2010). “He works outside the home; She drinks the coffee and does the dishes.” Gender roles in fiction programs on Dutch television. Journal of Broadcasting & Electronic Media, 54(1), 40-53.

22.

Englis

B. G.

Solomon

M. R.

Ashmore

R. D.

(1994). Beauty before the eyes of the beholder: The cultural encoding of beauty types in magazine advertising and music television. Journal of Advertising, 23(2), 49-64.

23.

Goodman

J. R.

Morris

J. D.

Sutherland

J. C.

(2008). Is beauty a joy forever? Young women’s emotional responses to varying types of beautiful advertising models. Journalism & Mass Communication Quarterly, 85, 147-168.

24.

Guzel

Aydin

(2014). A review of research on the effects of second life on speaking anxiety. Journal of Graduate School of Social Sciences, 18, 233-242.

25.

Hall

G. P.

(2008). Choosing life or second life? Discipline and agency in a mediated culture. International Review of Mission, 97(384/385), 7-20.

26.

Harrison

(2009). A virtual society. Rocky Mountain Communication Review, 6(1), 5-7.

27.

Hartley

M. D.

Ludlow

B. L.

Duff

M. C.

(2015). Second Life®: A 3D virtual immersive environment for teacher preparation courses in a distance education program. Rural Special Education Quarterly, 34(3), 21-25.

28.

Huang

J. S.

(2011). An examination of the business strategies in the second life virtual market. Journal of Media Business Studies, 8(2), 1-17.

29.

Hudson

Taylor

L. A.

Kozachik

S. J.

Shaefer

S. J.

Wilson

M. L.

(2015). Second Life simulation as a strategy to enhance decision-making in diabetes care: A case study. Journal of Clinical Nursing, 24, 797-804. doi:10.1111/jocn.12709

30.

Ibroscheva

Ramaprasad

(2007, June). A social construction model of stereotypes of foreigners. Paper Presented at the National Communication Association, San Francisco, CA.

31.

Jana

(2006, August). Starwood hotels explore second life first. Businessweek Online, 24.

32.

Joseph

W. B.

(1982). The credibility of physically attractive communicators: A review. Journal of Advertising, 11(3), 15-24.

33.

Lee

J. E. R.

Park

S. G.

(2009, June). Identity threat versus safety for racial minorities in virtual worlds: Effects of numerical representation of race as a diversity cue. Paper Presented at the International Communication Association, Chicago, IL.

34.

Lin

C. L.

Yeh

J. T.

(2009). Comparing society’s awareness of women: Media-portrayed idealized images and physical attractiveness. Journal of Business Ethics, 90(1), 61-79.

35.

Linden

(2015). Maturity ratings. Retrieved from https://community.secondlife.com/t5/English-Knowledge-Base/Maturity-ratings/ta-p/700119#Section_.2.2

36.

L’Oréal Paris. (2015). Retrieved from http://www.lorealparisusa.com/products/makeup/face/foundation-makeup/true-match-super-blendable-compact-makeup.aspx?shade=Buff-Beige-N4

37.

Lynch

B. M.

(2006). New Starwood brand lives second life first: Aloft: The virtual version. Meetings & Conventions, 41(12), 19.

38.

Martey

R. M.

Consalvo

(2011). Performing the looking-glass self: Avatar appearance and group identity in second life. Popular Communication, 9, 165-180. doi:10.1080/15405702.2011.583830

39.

Maybelline. (2015). Retrieved from http://www.maybelline.com/products/156/face/foundation/dream-nude-airfoam/classic-ivory/150?shadeId=856

40.

Mbure

(2009, June). Beauty through the eyes of cosmetics: The construction of beautiful faces through cosmetic advertising in black magazines. Paper presented at the International Communication Association, Chicago, IL.

41.

McQuail

(2010). McQuail’s mass communication theory (6th ed.). London, England: SAGE.

42.

Morris

(2006, June). Gender in print advertisements: A snapshot of representations from around the world. Paper presented at the International Communication Association, Dresden, Germany.

43.

O’Brien

M. G.

Levy

R. M.

(2008). Exploration through virtual reality: Encounters with the target culture. Canadian Modern Language Review, 64, 663-691.

44.

Patzer

G. L.

(1985). The physical attractiveness phenomena. New York, NY: Plenum Press.

45.

Pickering

(1995). The politics and psychology of stereotyping. Media, Culture & Society, 17, 691-700.

46.

Rak

(2009). The electric self: Doing virtual research for real in second life. Biography: An Interdisciplinary Quarterly, 32, 148-160.

47.

Robinson

Callister

Clark

Phillips

(2009). Violence, sexuality, and gender stereotyping: A content analysis of official video game websites. Web Journal of Mass Communication Research, 13, 1-17.

48.

Signorielli

McLeod

(1994). Gender stereotypes in MTV commercials: The beat goes on. Journal of Broadcasting & Electronic Media, 38, 91-112.

49.

Solomon

M. R.

Ashmore

Longo

L. C.

(1992). The beauty match-up hypothesis: Congruence between types of beauty and product images in advertising. Journal of Advertising, 21(4), 23-34.

50.

Weinberg

(2015). This company was 13 years early to virtual reality-and it’s getting ready to try again. Retrieved from http://www.businessinsider.com/second-life-is-still-around-and-getting-ready-to-conquer-virtual-reality-2015-3

51.

Willem

Arauna

Crescenzi

Tortajada

(2010). Girls on Fotolog: Reproduction of gender stereotypes or identity play? Interactions: Studies in Communication & Culture, 2, 225-242. doi:10.1386/iscc.2.3.225_1

52.

Wrench

J. S.

Thomas-Maddox

Richmond

V. P.

McCroskey

J. C.

(2008). Quantitative research methods for communication. Oxford, NY: Oxford University Press.

53.

(2006, June). The construction of consumer population: A gender analysis of advertising in the 1920s China. Paper presented at the International Communication Association, Dresden, Germany.

54.

Xie

Q. V.

Zhang

(2013). White or tan? A cross-cultural analysis of skin beauty advertisements between China and the United States. Asian Journal of Communication, 23, 538-554. doi:10.1080/01292986.2012.756046

55.

Yan

Bissell

(2014). The globalization of beauty: How is ideal beauty influenced by globally published fashion and beauty magazines? Journal of Intercultural Communication Research, 43, 194-214. doi:10.1080/17475759.2014.917432

56.

Zheng

Young

M. F.

Wagner

M. M.

Brewer

R. A.

(2009). Negotiation for action: English language in game-based virtual worlds. Modern Language Journal, 93, 489-511.