SOAN 2120 Lecture Notes - Lecture 10: Sampling Distribution, Sample Size Determination, Microsoft Powerpoint
Document Summary
Soan 2120 introductory methods october 3rd, 2017. Just as variables have distributions, sample statistics have distribution as well. If (cid:449)e do(cid:374)"t use (cid:396)a(cid:374)do(cid:373) sa(cid:373)ple, ou(cid:396) (cid:272)o(cid:374)(cid:272)lusio(cid:374)s (cid:272)a(cid:374) (cid:271)e (cid:272)halle(cid:374)ged: there is absolutely no substitute for collecting good data whatsoever. Its standard deviation gets smaller as the size of the sample tends to get larger. It is a distribution of the means from all possible samples. If we repeat a random phenomenon many times, average value will get closer to the parameter. It is because of central limit theorem that we can test our theories - see the rights/wrongs. Sampling error: sampling error - difference between the statistics and the parameter due to random processes. It can be affected by two different things: sampling method random samples have less error than nonprobability ones, sample size sampling error decreases as the sample size goes up (increases) It could be close or it could be far enough away.