Please login or create an account to join the discussion.

The inside story of how an Ivy League food scientist turned shoddy data into viral studies

Image: Jorge Franganillo, Drowning by numbers, Flickr, Creative Commons Attribution 2.0 Generic
Image: Jorge Franganillo, Drowning by numbers, Flickr, Creative Commons Attribution 2.0 Generic

This Buzzfeed story follows allegations that a Cornell researcher published studies obtained through the scientifically dubious method of ‘p-hacking’.

'P-hacking' refers to sifting through all combinations of variables in a dataset to find one with a strong correlation (which may happen by chance, given enough data). This goes against the generally accepted scientific method, where researchers should state their hypothesis in advance of carrying out the study rather than retrospectively finding a story in the data. Affected studies include research about eating habits at a buffet, investigations into whether children can be encouraged to eat apples by decorating them with stickers, and whether people eat more when they have been thinking about exercise.

Read the full story here. See also the Foodsource resource What are the influences on our food choices?

Post a new comment »

Login or register to comment with your personal account. Anonymous comments require approval to be visible.
CAPTCHA
Planten