I teach a Stats class Winnie the Pooh would love. Sort of the way the vet used to describe my Wheaten Terrier: when you look in her ear, you see light
( Read more... )
I'm very curious: what result did your students think their blocking would provide, and how did they think their distribution would provide any kind of useful statistics?
The question asked them to design an experiment, then asked if it could be improved by blocking. The expected response was that you should block those who previously had therapy, allowing you to compare the effects of the new drug for people in therapy to the effects of the old drug for people in therapy (and similarly, compare the two drugs for people without therapy.)
Alternately, a case could be made for blocking males and females -- not an uncommon practice -- because drugs will often react differently in the genders.
I think my bunnies actually recognized both those possibilities, but, little overachievers that they are, tried to do both at once. And of course, being bunnies, never stopped to think "oh wait!...."
Comments 2
I'm very curious: what result did your students think their blocking would provide, and how did they think their distribution would provide any kind of useful statistics?
Reply
The question asked them to design an experiment, then asked if it could be improved by blocking. The expected response was that you should block those who previously had therapy, allowing you to compare the effects of the new drug for people in therapy to the effects of the old drug for people in therapy (and similarly, compare the two drugs for people without therapy.)
Alternately, a case could be made for blocking males and females -- not an uncommon practice -- because drugs will often react differently in the genders.
I think my bunnies actually recognized both those possibilities, but, little overachievers that they are, tried to do both at once. And of course, being bunnies, never stopped to think "oh wait!...."
Reply
Leave a comment