PSY201H1 Lecture Notes - Lecture 10: Central Limit Theorem, Statistical Parameter, Sampling Distribution
Document Summary
Result: repeat over and over again, shape will approximate population. Pattern that is generalizable: sampling error. Comes from variability, discrepancy, due to chance, not mistake! Samples vary randomly, two samples are rarely identical. But the more we sample, the close the estimate will be to the parameter: sampling distribution of sample means. So far, we have been talking about distribution of scores (aka. Now, let"s focus on distribution of samples. Of means from selscelint all possible samples of a given size n from the given population. Note: it is theoretical because it is impossible to calculate population. We select all possible samples of 5 in the class plot means on a figure . Most of the averages will be close to each other. Closer to the middle = more frequent. Expected difference between any individual sample and population mean: central limit theorem. Previous knowledge can still be applied even if it"s not normal!
