It seems widely accepted that, overall, men tend to find women in high heels more attractive. (There will, of course, be many exceptions to this rule.)
"High heels may well be the most potent aphrodisiac ever concocted." -- Rossi 1981
However, the increase in attraction seems to be only indirectly ascribed with wearing high heels. Typically this is described along the lines: high heels cause a change in the posture of the female, which the male finds (sexually) attractive.
Question: How can we exclude the possibility that the high heels themselves are not the main cause of the increase in attractiveness?
That is, how can we exclude the possibility that men, overall, primarily just like the sight of women in high heels, for some reason or another?
If the increase in attractiveness were primarily due to these indirect factors, then we should see a comparable increase in attractiveness e.g. if a woman were merely to stand on a slope. (Although, this would not account for the addition factors, such as gait, and the sound made when walking in high heels.)
There are things that women do, such as fingernail painting, which increases attractiveness, but (arguably) not sexual attractiveness. It seems plausible that the "femininity" associated with wearing high heels (as with fingernail painting) could be a large factor in the increase in attractiveness.
Rossi WA. (1981) High heels: the agony and the extasy. Journal of the American Podiatry Association 71(12), 698-699.