PSYC 60 Chapter Notes - Chapter 9: Statistical Parameter, Squared Deviations From The Mean, Confidence Interval
Document Summary
9. 1 the t statistic: an alternative to z mean computing z-score test statistic. How much difference is reasonable to expect between sample mean (m) and. Standard error provides measure of how well a sample mean approximates population. Purpose of hypothesis test: hypothesis test is significantly greater than chance. Z-score test requires more: population standard deviation/variance. Sample mean (m) is expected to approximate its population mean () population mean () Compare obtained sample mean (m) with hypothesized population mean ((cid:4667)by or = 2. Sample variance provides accurate and unbiased estimate of population variance unknown; use standard error in denominator t = t = N-1 scores in a sample are independent thus free to vary. Degrees of freedom = df = n - 1. Shape of a t distribution changes with degrees of freedom. Df gets very large t distribution gets closer in shape. Df small t distribution has more variability (flatter and more spread out)