SOCPSY 2K03 Lecture Notes - Lecture 5: Sampling Frame, Psy, Repeated Measures Design
Document Summary
If you have a truly random sample, 95% of the values will fall within the main area. Determine sampling ratio: well ya can"t sample 30,000 students . That would put you at a 10:1 sampling ratio. This tells you how many out of the whole population you"ll get. This means each student has the same chance of being sampled. If you only choose 1st year students you won"t get correct responses. You have to make sure the faculty is well-represented. Age, race, gender, study, blonde, black, sporty, creative. If you sampled your friends they would all have the same interests. Snowball sample - you recommend somebody who recommends somebody etc. You choose from a list of numbers, like 10,20,30. You will get a good random sampling from this. Sometimes we have certain things we want included like a certain # of students from each faculty. You would have to rst separate out all the categories (ex colors)