PSYC 3031 Lecture Notes - Lecture 6: Standard Deviation, Sampling Distribution, Confidence Interval
10 May 2017
School
Department
Course
Professor
Document Summary
Z = x population / standard deviation. If we assume population is normal, we can determine proportions of the population above/below each of these scores using the normal table. Very important: act of transofmring raw scores to z scores does not change shape of original distribution. Cannot normalize data by transofmring to z scores, shape will be unchanged, will have mean of. Can only use z table to determine probability if your original data was normally distributed before you transformed it to z scores. Then denominator of z formula is standard deviation of mean itself. Heres how: pick a sample size, draw indefinitely many samples of size n with replacement from the same population, calculate mean of each sample, plot means in histogram. Calculate standard deviation of all the means you collected. Formula: sdm = sdz / square root n. With sem we can test whether mean is likely to have come from specified population.
