“Once More: Is Beauty in the Eye of the Beholder? Relative Contributions of Private and Shared Taste to Judgments of Facial Attractiveness”, Johannes Hönekopp2006 ()⁠:

Misconstruing the meaning of Cronbach’s alpha, experts on facial attractiveness have conveyed the impression that facial-attractiveness judgment standards are largely shared. This claim is unsubstantiated, because information necessary for deciding whether judgments of facial attractiveness are more influenced by commonly shared or by privately held evaluation standards is lacking.

3 experiments, using diverse face and rater samples to investigate the relative contributions of private and shared taste to judgments of facial attractiveness, are reported.

These experiments show that for a variety of ancillary conditions, and contrary to the prevalent notion in the literature, private taste is about as powerful as shared taste.

Important implications for scientific research strategy and laypeople’s self-esteem are discussed.

[Keywords: attractiveness, faces, reliability, agreement]

…Whereas the pay of top models bespeaks the importance of shared taste, the recollection of any discussion between yourself and a friend about the attractiveness of passersby probably advocates for the importance of private taste. The aim of this article is to inform about the relative contributions of private and shared taste to judgments of facial attractiveness.

…For the sake of efficient communication, I call each face’s average rating its face score; I call the average of a single judge’s ratings a judge score.

[Examining the range/extremes of differences in judgment:]

Figure 2: The impact of private taste. For each rater, the rank order of faces on the basis of both evaluations of each face was computed. The figures show each face’s most (triangles) and least (squares) favorable evaluations. As an example, take the leftmost triangle and square from Experiment 2: Both pertain to the face with the most favorable face score in Experiment 2. The triangle indicates that there was at least one judge who liked this picture best; the square indicates that this picture obtained the 30<sup>th</sup> rank in the preference order of the judge who liked this face least. As can be seen, judge samples of only moderate size (30, 31, and 50 judges, respectively) already yield extreme evaluation differences for almost all faces.
Figure 2: The impact of private taste. For each rater, the rank order of faces on the basis of both evaluations of each face was computed. The figures show each face’s most (triangles) and least (squares) favorable evaluations. As an example, take the leftmost triangle and square from Experiment 2: Both pertain to the face with the most favorable face score in Experiment 2. The triangle indicates that there was at least one judge who liked this picture best; the square indicates that this picture obtained the 30th rank in the preference order of the judge who liked this face least. As can be seen, judge samples of only moderate size (30, 31, and 50 judges, respectively) already yield extreme evaluation differences for almost all faces.

…What about the prevailing message in the scholarly literature that “standards of beauty are widely shared” (Rhodes et al 200123ya, pg31)? In light of the data presented here, a statement like this is not “wrong”, but it is very likely to bring about a wrong notion about facial attractiveness. It is a bit like telling a little girl that a zoo is a place where many children eat ice cream and have much fun; in saying as much, one says nothing wrong, but the girl will acquire a queer concept of a zoo. After all, it is not less true to say that standards of beauty are widely private. Because both statements are true, seemingly militating phenomena can peacefully coexist: Some people can make a fortune with their good looks because they appeal to a broad public, and friends can endlessly debate about who is attractive and who is not.