Selection Questions

[foodchain]

Looking at selection bias, which if I understand such is basically to mean the impact any particular structured observation has when using statistical data, does this correlate to the concept of selection rules in quantum mechanics or chemistry, or science in general that uses statistics in something?

 

In the wiki article a selection rule is defined for in quantum mechanics as something that yields a nonzero probability. Is this actually structured into the formalism of QM somehow as to related to selection bias? I think the latter was worked on or discovered if you want way after QM had been established. It would seem to be then that if selection bias is true, could this part be confused with measurement problem, such as when you make a measurement collapse could be confused with selection bias?

 

http://en.wikipedia.org/wiki/Selection_rules

 

http://en.wikipedia.org/wiki/Selection_bias

 

http://en.wikipedia.org/wiki/Schrodinger_equation

[JohnB]

Science in general, I would think. It's something to look for when deciding exactly how your selection criteria will work. If a bias is introduced into the initial data, then conclusions will be biased as well.

 

One way bias can be introduced is in a poll. Say you want to know what the average Kalathumpian thinks about a .gov policy. You decide to phone around and poll some people. However, since you work 9-5 you decide to phone during those hours.

 

This will introduce a bias because you are phoning at a time when most people are at work. While there will be some home on holidays, RDOs, or sick, you will be excluding a large group from your original data.

 

Hence your final analysis will not be the opinion of the population, but the opinion of the population that don't work during the day.

 

At least, that's how I understand it. I could be wrong.

 

PS. Why did you post the question in this subforum? I think it's a perfectly valid question.