The problem with all this type of research is that they (the authors) do not have any model to account for their findings. I mean, interesting observations, but...nothing more. Not to mention that it does contradict other studies (also within psychology; every day there's a new descriptive study observing something else on two dozens observations and extrapolating to the whole population...). Not to mention that it does not take into account the subjects' own social status and other observables (forget about unobserved heterogeneity...). Not to mention that- and this is linked to the previous critique- this is in no way a 'ceteris paribus' conclusion. Not to mention that the sequential process here (first photos, then info on the persons) could influence the final choice- a lot. Etc. Probably this is why the authors are not (and cannot ever be) economists.
Note1: already from the title, the whole thing is counter-intuitive and I don't think extensive data supports the conclusion asserted here (ie. "women fall for Mr. Average", ceteris paribus, that is controlling for everything else, looks included)
Note2: it could be of course that the BBC distorted the message of the original article, could not find that.