Back in October I posed a question: Is there any truth in the old wives tale that rubbing your hands on stainless steel gets rid of garlic smells? Various theories as to how steel may achieve this were posited. But I wanted to know if there was a real effect in the first place. Kitchen chemists everywhere helped answer this by taking part in a stinky citizen science challenge. And the results are, well, interesting.
I asked people to conduct a quick experiment whilst prepping dinner. The task was simply to rub the palms of their hands with garlic. Then treat one hand with a wipe from a stainless steel spoon and the other with a wooden spoon. Finally participants asked some other poor soul to take a sniff of their hands and report on whether there was a discernible difference.
Thanks to everyone who took up the garlic challenge (especially the person who did their experimenting whilst cooking Christmas dinner).
And so to the results.
These were collected via surveymonkey, with the question “Which hand smelt more of garlic?” and the answer choices a) The hand rubbed with the wooden spoon, b) The hand rubbed with the stainless steel spoon, c) Couldn’t tell the difference.
44 allium lovers responded. Of those 17 thought the hand treated with the wooden spoon smelt more garlicky, 6 said the stainless steel treated hand was the stinkier. So far, so good. Looks like the stainless steel effect might be real. But here’s the rub, there’s still the other 21 responses, none of whom could tell the difference between the smelly hands.
So we’ve got results that are significantly different from an even distribution between the options (the two-tailed P value equals 0.0163 ,according to a chi squared test) . However, the stainless steel treatment seems to be only about 38% effective, assuming the wooden spoon is a good negative control. But maybe the abrasive, absorbent wooden spoon is also quite good at removing garlic smells? In which case the effectiveness of the stainless steel is an underestimate.
Oh well, sorry people, but it looks like I can’t really offer a definitive answer. In hind sight I think the experimental design could have been better. A before and after spoon treatment sniff test would have been a good idea. And maybe a better negative control was in order.
Looks like another round of experiments could be in order. Or can anyone offer a better way of analysing the data (I suspect sensitivity vs specificity analysis might be more appropriate)?