STATS 13 Chapter Notes - Chapter 1.4-1.5: Standard Deviation, Central Limit Theorem, Null Hypothesis
Document Summary
Small p-values and large standardized scores (in absolute value) show strong evidence against the null hypothesis. >0. 5: two-sided test: more extreme must go in both directions b. i. b. ii. b. iii. b. iv. P-value is computed by finding how frequently the observed statistic or more extreme occurred in one tail of the distribution and adding that to the corresponding probability of being at least as extreme in the other direction. P-value is larger than for one-sided test less evidence. Used in science when researcher doesn"t have a prior assumption on the direction of the outcome. Section 1. 5: inference on a single proportion: theory-based approach. Null distributions of sample proportions often exhibit a common and familiar shape. Centered at the null hypothesis value for pi. If the sample size (n) is large enough, the distribution of sample proportions will be bell-shaped, centered at the long run probability ( ), with a standard deviation of.