So what we see is, again perfectly reasonable statistical techniques, but not looking at things in the right way.

所以我们这里可以看到，还是很完美很有逻辑的统计学工具，但是他们没有用正确的方式去看待数据。

麻省理工公开课 - 计算机科学及编程导论课程节选

Again, you can't even correct for it because, let's just say, the whole population overestimated calories and things were underestimated by 30% percent, well then you just put in that correction and you know what you got.

你甚至无法纠正这个误差,因为,比方说,整个总体人群都高估了卡路里,或者都低估了30%的卡路里,研究人员纠正误差才能得到正确数据

耶鲁公开课 - 关于食物的心理学、生物学和政治学课程节选