Is attractiveness an arbitrary social construct?

From "Maxims or Myths of Beauty? A Meta-Analytic and Theoretical Review" Langlois et al 2000:

"Common maxims about beauty suggest that attractiveness is not important in life. In contrast, both fitness-related evolutionary theory and socialization theory suggest that attractiveness influences development and interaction. In 11 meta-analyses, the authors evaluate these contradictory claims, demonstrating that (a) raters agree about who is and is not attractive, both within and across cultures; (b) attractive children and adults are judged more positively than unattractive children and adults, even by those who know them; (c) attractive children and adults are treated more positively than unattractive children and adults, even by those who know them; and (d) attractive children and adults exhibit more positive behaviors and traits than unattractive children and adults. Results are used to evaluate social and fitness-related evolutionary theories and the veracity of maxims about beauty.

..."Beauty is good" was empirically tested in a seminal study by Dion, Berscheid, and Walster (1972), who demonstrated that, even in modem times, human beings attribute positive qualities to attractive people and negative qualities to unattractive people. Since this important study, much research has focused on the effects of attractiveness, especially facial attractiveness, on, the attributions, impressions, and stereotypes of strangers. These studies primarily have investigated attributions made by college students about attractive and unattractive strangers based on a photograph of the face and, sometimes, minimal printed "background information" about the hypothetical individuals. This stranger-attribution literature has been summarized by earlier meta-analyses (see, e.g., Eagly, Ashmore, Makhijanl, & Longo, 1991; Feingold, 1992b), which have confirmed the association between attractiveness and many attributions of positive characteristics. Defining attractiveness theoretically is a topic of great interest and controversy. Until recently, empirical work proceeded without any conceptual or scientific definition of attractiveness: Researchers simply defined people as attractive when raters agreed they were attractive. Although this issue is beyond the scope of the current article, the interested reader is referred to Cunningham (1986); Farkas, Munro, and Kolar (1987); Langlois and Roggman ( 1990); and Langlois, Roggman, and Musselman (1994), among others.

...The stranger-attribution literature and meta-analyses of it have firmly established the existence of the "beauty is good" stereotype (see, e.g., Adams & Crane, 1980; Berscheid & Walster, 1974; Dion, 1973; Downs & Harrison, 1985; Eagly et al., 1991; Feingold, 1992b; Langlois, 1986; Ritter, Casey, & Langlois, 1991; Snyder, Berscheid, & Glick, 1985). At least among strangers, attractiveness clearly elicits differential expectations for the behavior and traits of attractive and unattractive targets.

...different cultures presumably have very dissimilar standards of beauty (see, e.g., Darwin, 1871; Ford & Beach, 1951). Assessing the veracity of this maxim is important because without consensus in judgments of attractiveness, there can be no consistent effect of attractiveness on social judgments, interactions, or behavior. Empirically, if beauty is only in the eye of the beholder, judgments of attractiveness should show little consistency among raters and, thus, low reliability coefficients. Only one meta-analysis we know of has quantitatively evaluated the reliability coefficients of attractiveness judgments. Feingold (1992a) examined panels of raters from within the United States and Canada and found a high level of agreement among adult raters. However, we do not know of any meta-analysis that has examined reliability coefficients of attractiveness judgments made by or about children nor do we know of any meta-analysis investigating
agreement across raters of different ethnicities and cultures.

...several individual studies have indicated that people do interact differently with others based on attractiveness (Dion, 1974; Langlois, Ritter, Casey, & Sawin, 1995; Snyder, Tanke, & Berscheid, 1977; Stewart, 1980; 1984), no meta-analysis is currently available to assess judgment and treatment of individuals following actual interactions or to assess judgment and treatment of individuals people know.

...Two meta-analyses have examined the relation between attractiveness and some behaviors and traits (Feingold, 1992b2; L. A. Jackson, Hunter, & Hodge, 1995). Feingold (1992b) reported significant relations between attractiveness and measures of mental health, social anxiety, popularity, and sexual activity but nonsignificant relations between attractiveness and sociability, internal locus of control, freedom from self-absorption and manipulativeness, and sexual permissiveness in adults. Feingold also found a nonsignificant relation between attractiveness and intelligence (r = .04) for adults, whereas L. A. Jackson et al. found a significant relation for both adults (d = .24 overall, d = .02 once selected studies were removed) and for children (d = .41).

...We began by examining the adult and child attractiveness literature obtained from APA databases (PsycLIT, PsycINFO), Cash's (1981) annotated bibliography, and the ERIC database of published, unpublished, and conference papers using key terms beauty,facial attractiveness, and physical attractiveness. We also examined every relevant article from the reference sections of all retrieved articles. This search resulted in obtaining references from 1932 through June 1999. ...From an initial sample of over 1,800 empirical articles, we obtained a final data set of 919 usable effect sizes with some studies contributing more than one independent sample.

...All reliability coefficients of attractiveness ratings from retrieved studies were analyzed. The vast majority of these studies asked raters to evaluate attractiveness from photographs (generally these raters were not the same participants analyzed for judgment or treatment); the remaining studies asked raters to evaluate attractiveness in situ or from videotapes. In almost all cases, judges used either a Likert-type scale or rank orders to evaluate attractiveness.
We were interested in whether children would evaluate attractiveness similarly to adults, whether raters of different ethni¢ity would evaluate attractiveness similarly when residing in similar cultures with similar exposure to media standards of beauty, and whether raters of different ethnicity within different cultures and presumably different media exposure would evaluate attractiveness similarly. Thus, four analyses were performed to evaluate: (a) Adult within-culture, within-ethnic agreement, including only studies in which adults residing in the same culture were rated by others of the same ethnicity as the targets; (b) child within-culture, within-ethnic agreement, including only studies in which children residing in the same culture were rated by others of the same ethnicity as the targets; (c) adult within-culture, cross-ethnic agreement, in which correlations among raters residing within the same culture but from different ethnic groups were analyzed (e.g., African Americans judging European Americans); and (d) adult cross-cultural, cross-ethnic agreement, in which correlations among raters residing in different countries and from different ethnic groups were analyzed (e.g., Koreans judging African Americans). We located insufficient numbers of studies of children from different ethnic groups from either their own or other cultures, precluding analyzing cross-ethnic reliability for children.

...Using DSTAT (B. T. Johnson, 1989), we first calculated one effect size for the overall category (e.g., treatment), collapsing across all subcategories. Because each sample of participants should contribute only one entry per analysis (Hedges & Olkin, 1985), multiple effect sizes obtained from the same participants were converted to Z scores, averaged, and the average converted back to an r, resulting in only one effect size per sample (Rosenthal, 1995). When this process necessitated averaging _n_s of slightly different sizes, usually due to participant attrition on certain measures but not others (e.g., Leinbach & Fagot, 1991), we used the smallest n to be most conservative. Most often, averaging was performed within individual studies, but on a few occasions, separate studies reported statistics from the same sample (e.g., Jovanovic, Lerner, & Lerner, 1989; Lerner et al., 1990, 1991), and then the _r_s from more than one study were averaged. Second, we calculated separate effect sizes for each subcategory of dependent variables (e.g., social skills). At the subcategory level, multiple measures of the same construct obtained from the same participants were averaged, again resulting in only one effect size per sample in each subcategory.

...The meta-analyses showed that, both within and across cultures, people agreed about who is and is not attractive. Furthermore, attractiveness is an advantage in a variety of important, real-life situations. W e found not a single gender difference and surprisingly few age differences, suggesting that attractiveness is as important for males as for females and for children as for adults. Other moderator variables had little consistent impact on effect sizes, although in some cases there were insufficient data to draw conclusions.
Reliability of Attractiveness Ratings (Within-Culture Agreement): The meta-analysis of effective reliability coefficients revealed that judges showed high and significant levels of agreement when evaluating the attractiveness of others. Overall, for adult raters, r = .90 for ratings of adults and r = .85 for ratings of children, both p < .05 (see Table 3)...These results indicate that beauty is not simply in the eye of the beholder. Rather, raters agreed about the attractiveness of both adults and children. Our findings for reliability of adult raters were consistent with Feingold (1992b), who meta-analyzed reliability coefficients from samples of U.S. and Canadian adults and obtained an average effective reliability of r = .83. More importantly, our cross-cultural and cross-ethnic analyses showed that even diverse groups of raters readily agreed about who is and is not attractive...These analyses seriously question the common assumption that attractiveness ratings are culturally unique and merely represent media-induced standards. These findings are consistent with the fact that even young infants prefer the same faces as adults (Langlois, Ritter, Roggman, & Vaughn, 1991; Langlois et al., 1987; Langlois, Roggman, & Rieser-Danner, 1990).

...The overall d+ (1.11) and the BESD statistic showed that approximately 75% of attractive children, compared with only 25% of unattractive children, were judged to be above the mean for all categories combined. The subcategory analyses showed large effect sizes, especially in the domains of social appeal (d÷ = 1.33; 78% vs. 22%) and academic/developmental competence (d÷ = 1.10; 74% vs. 26%). Attractive children were also judged significantly more positively on the dimensions of adjustment (d÷ = .95; 72% vs. 28%) and interpersonal competence (d+ = .92; 71% vs. 29%). No variable significantly moderated the findings (see Table 6).
Attractive adults were judged more positively than unattractive adults were (overall d+ = .50; 62% vs. 38%), particularly for occupational competence (d÷ = .90; 70% vs. 30%). Attractive adults were also judged as having more social appeal (d÷ = .49; 62% vs. 38%), as more interpersonally competent (d+ = .45; 61% vs. 39%), and as better adjusted (d+ = .25; 56% vs. 44%) than unattractive adults.
The moderator analyses revealed a single significant influence of the coded variables on the effect sizes: Year of publication accounted for 10.7% of the variance. Studies published more recently produced larger effect sizes (see Table 6).
...Overall, these results indicate that despite conventional teachings, people do indeed judge books by their covers even when they have behavioral or other information on which to base their judgments. The differences in the informed judgment about attractive and unattractive children were the strongest effect sizes we obtained and, compared with other effect sizes in the social sciences, were uncommonly large (none being smaller than d÷ = .92). Although the effect sizes for informed opinions about adults were not as large as those for children, they were medium in size and among the larger effect sizes we obtained for adults.

[Uncommonly large indeed. For comparison, see "One Hundred Years of Social Psychology Quantitatively Described" (http://jenni.uchicago.edu/Spencer_Conference/Representative%20Papers/Richard%20et%20al,%202003.pdf), Bond et al 2003.]

...Attractive adults were also treated significantly more favorably than unattractive adults were (overall d+ = .54; 63% vs. 37%). Attractiveness had the largest effect on attention (d+ = 1.09; 74% vs. 26%), followed by reward (d+ = .68; 66% vs. 34%), positive interaction (d÷ = .57; 64% vs. 36%), positive impression management (d÷ = .53; 63% vs. 37%), negative interaction (d÷ = - . 5 4 , 63% vs. 37%), and help-giving/cooperation (d+ = .36; 59% vs. 41%). No moderator variables were significant.'
Surprisingly, in addition to being judged differently as a function of their attractiveness, attractive individuals on average were treated significantly better than unattractive individuals. These findings are powerful evidence that, contrary to popular belief, attractiveness effects extend beyond mere "opinions" of others and permeate actual actions towards others, even though people may not be aware of it...Compared with unattractive adults, attractive adults experienced much more occupational success (d÷ = .76; 68% vs. 32%), were liked more as indicated by the subcategory of popularity (d÷ = .65; 65% vs. 35%), and had more dating experience (d÷ = .55; 63% vs. 37%), more sexual experience (d÷ = .31; 58% vs. 42%), and better physical health (d÷ = .39; 59% vs. 41%). In addition, attractive adults were somewhat more extraverted (d÷ = .26; 56% vs. 44%), had somewhat more traditional attitudes (d÷ = .27; 57% vs. 43%), were somewhat higher in self-confidence/self-esteem (d÷ = .24; 56% vs. 44%), possessed somewhat better social skills (d+ = .20; 55% vs. 45%), had slightly better mental health (d÷ = .16; 54% vs. 46%), and were very slightly more intelligent (d÷ = .07; 52% vs. 48%; see Table 8).

...One moderator accounted for a significant portion of the variance in the 'overall effect size. Type of attractiveness measure accounted for 14.6% of the variance; studies using measures of attractiveness that included the face plus additional cues had higher effect sizes than studies using measures of facial attractiveness only (see Table 6)...When controlling for all variables simultaneously, we found not a single significant gender effect. The overall lack of gender differences in ecologically valid situations suggests that, in most domains, attractiveness is equally important for men and women. These findings are consistent with the lack of gender differences found in other meta-analyses of stranger attribution research (Eagly et al., 1991; Feingold, 1992b).

...The effects of facial attractiveness are robust and pandemic, extending beyond initial impressions of strangers to actual interactions with those whom people know and observe. Contrary to conventional wisdom, there is strong agreement both within and across cultures about who is and who is not attractive. Furthermore, attractiveness is a significant advantage for both children and adults in almost every domain of judgment, treatment, and behavior we examined. The magnitude of attractiveness effects is roughly the same as or larger than that of other important variables in the social sciences (Eagly, 1995). In most cases, the benefits of attractiveness are large enough to be "visible to the naked eye" (Cohen, 1988) and are of considerable practical significance (Lipsey & Wilson, 1993; Rosenthal, 1991, 1995). These meta-analyses starkly illuminate the fundamental contradiction between empirical research and maxims about beauty. On the basis of our results, we conclude that the maxims we examined are myths, not reality: Beauty is more than just in the eye of the beholder; people do judge and treat others with whom they interact based on attractiveness; and, perhaps most surprisingly, beauty is more than just skin-deep...Fitness-related evolutionary theories (good genes, human mate selection, differential parental solicitude) all agree that because morphological characteristics such as facial attractiveness are honest indicators of fitness, health, quality, and reproductive value, attractiveness should be important in human interactions (Barber, 1995; Buss & Schmitt, 1993; Daly & Wilson, 1995; Symons, 1979; Thornhill & Gangestad, 1993). All three models agree that perceivers should consistently detect and recognize attractiveness, both within and across cultures, because humans have evolved universal standards of facial attractiveness based on clues to health and reproductive fitness. Furthermore, as an evolved trait, agreement about attractiveness should be evident within as well as between cultures. As we have shown with reliability, this criterion has been met, and all three fitness-related theories have proven to be more predictively accurate than social theory in this regard."