Using the attached information, I have to discuss a priori (data collectioncategory definition before collection) and post hoc (datacollection category definition after collection) as they relate to analysis of a population sample. Is one method preferable? What are the strengths and weaknesses of each?
I want to first take a step back and look at how an experiment is defined.
Normally, you have a large population of people. You would then randomly selection people from this population and assign them to either the control group or experimental group. Let’s look at an example.
You want to research if having music play while your study will affect grades.
The entire population would be all of the students in a given university. You would then randomly select students in all different classes, topics, levels, backgrounds and randomly assign them to one of two groups = half will listen to music while they study, and half will study in silence. You then have them write their exams, and you look at the results to see if music affects grades.
Here, every single person in the sample will have a chance of getting into either group due to randomization. Therefore each group should be equal in its composition – there should be similar number of people with good grades and poor grades, similar people in year 1 as in year 3… The chance of errors in the results are less since we can assume that both groups are similar in nature, and …
Detailed explanation of problem.