This article at Slate discusses the current fascination certain kind of researchers have with human ovulation. The important part is this:
A lot of this research is pretty much based on speculation, says Harriet Hall, who is an editor at the Web site Science-Based Medicine. She explains that whether their goal is marketing or to better our understanding of the hidden effects of ovulation, all the studies tend to suffer from the same basic flaws. "They are isolated studies that have not been replicated, and the findings could be inaccurate due to chance factors."
Replication matters, a lot, and not only because of chance factors.
Remember those magical hip-to-waist ratio studies certain types of evolutionary psychologists created? The ones which quickly became part of folk mythology at all cocktail parties? How all men all over the world prefer women with a certain hip-to-waist ratio?
I bet you do. But you may not know that replications (by different researchers) failed to support such a simplistic conclusion.