PSYC08H3 Lecture Notes - Lecture 1: Bias Of An Estimator, Sample Size Determination, Null Hypothesis
Document Summary
We ha(cid:448)e a fo(cid:396)(cid:373)ula to (cid:272)o(cid:396)(cid:396)e(cid:272)t fo(cid:396) that (cid:858)(cid:449)(cid:396)o(cid:374)g(cid:374)ess(cid:859) i(cid:374) yba (sample population) The formula is variance: formula variance we do n-1 in the denominator (degrees of freedom); we cannot divide by n because ybar is not mu, ybar. If you put mu into variance formula; use n: variance will not be exact, but unbiased estimate, using n-1 on average will have an unbiased estimator of sigma. There is a population mean; a population is whatever we decide it is. The population is the room; there is a true mean age. Any chances of sample mean being equal to population mean is zero there is only 50% higher and 50% lower at that point. There is no percent for that point; there are no two scores that are exactly the same (haha, like snowflakes) We care because there is variability in everything, we need to understand these differences.
