I'm skeptical of this statstical analysis for the following reasons: it appears the authors would have published their findings if the opposite had been found, that is that women waited less time for their coffee.

This requires a two sided analysis with a p-value set at 0.025, rather than 0.05, further, it appears they would have published their findings if some factor other than gender had proven most important. This shotgun approach requires them to divide their p-value by the number of variables considered - the Bonferroni adjustment.

Using a one-tailed p-value at 0.05, one states that a statistically significant event is present because the risk of it happening by chance is less than one in twenty. The authors have considered sex, race, subjective attractiveness, age, drink type, and store workload as independant variables - by chance alone the risk of one of these being "significant" at a p-value of 0.05 is 30%; not accounting for the unconscious bias of the observers in looking for the outcome.

Further, the variable of interest is a poor marker for level of service. The time from ordering a drink to it being served could indicate many things. The authors fail to consider the possibility that a barista spending more time making a drink could be a sign of better service rather than contempt. If the reported extra twenty second discrepency is real - what were baristas doing during that interval, perhaps frothing deflated foam or switching the coffee to a fresh urn?

I agree, this study look like an activity for undergraduates. It's rife with selectional and observational bias.

At a more basic level - the most important variable from a modeling level - the complexity of the drink - was poorly modeled.

"The authors fail to consider the possibility that a barista spending more time making a drink could be a sign of better service rather than contempt. If the reported extra twenty second discrepency is real - what were baristas doing during that interval, perhaps frothing deflated foam or switching the coffee to a fresh urn?"

Excellent point.

Let's see a study about how much pointless studies with questionable findings add to the cost of college tuition?