Original title: Over 90% of Research Studies Make Me Want to Scream (P < 3E-12).

Shania Twain is in the news today. No, her new album still isn’t out, but her face is in the spotlight. It turns out someone “applied” the latest “research” to “determine” that she has the perfect face, “scientifically” speaking. The distance between her eyes and mouth are precisely 36% of the length of her face, and her interocular distance is exactly 46% of its width. These proportions, according to an article in press at Vision Research, are universally optimal (among low-resolution, mostly Photoshopped images of a few white women).

Garbage. Poppycock. Nonsense. Balderdash. Crap, crap, crap of a research paper, right from sentence 1: “Humans prefer attractive faces over unattractive ones.”

But you came here for the pictures. First, the 2009 Most Bogus Regression award winner, Figure 2A:


Forget about faces. I’m seeing a butt-ugly relationship between the residuals and the model variable. Not only that, but the choice of a quadratic model has no theoretical basis beyond an expectation of U-shapedness, yet I’m supposed to believe that the vertex of this parabola better represents some ideal ratio than, say, the actual ratio people tended to find most attractive? I don’t think so. The actual preferred ratio looks to be a bit smaller than Shania’s, by the way. The scatterdots aren’t hard to read. Apologies for not having picked a canonical celebrity to call out.

Here’s what I think. The authors, whether intentionally or not, used a sleazy trick to generate model-fitting data. They threw in extreme and unnatural faces on each end to add life to some uncorrelated data having to do with real human face ratios. Let’s put a face on the graph. Heck, let’s put three faces on the graph, including two with more extreme ratios than the authors included. The middle one is from Figure 1 of the paper. The others are modifications of faces from Figure 1.


Given two faces, one human-looking and one Photoshopped to look inhuman, what do you think people answer when asked which is “more attractive”? Whatever it means to say humans “prefer attractive faces,” I’m sure they also prefer human faces, or faces that could by some stretch of the imagination be a living human’s face. Most real human faces fall somewhere between the green lines I drew, where there’s nothing going on, data-model-wise.

To hide the insanity (and here it’s hard to buy a lack of intent), the published paper includes face images with ratios between 0.30 and 0.45 only.

Ladies, you look just fine the way you are.


  1. Pallett, P. M., et al. New ‘‘golden” ratios for facial beauty. Vision Research (2009), doi:10.1016/j.visres.2009/11/003.
  2. Perfect face dimensions measured,” BBC News.