STATS 13 Study Guide - Midterm Guide: Null Hypothesis, Central Limit Theorem, Simple Random Sample
Document Summary
Standard deviation: average distance between a data value and the mean. Statistic or random variable: numerical description of the sample. Parameter: property of the whole population (does not change between samples) Test statistic (standardized statistic): indicates how many standard deviations the observed statistic is above its hypothesized value under the null hypothesis. Number of standard deviations our sample statistic is above the mean of the null distribution. Standard error: standard deviation of the quantity of interest under the null hypothesis. P-value: the probability, assuming the null hypothesis is true, the test statistic will be at least as extreme as that observed. Difference between observed statistic and null hypothesis parameter. Sample size (larger sample stronger evidence, less variability) Two sided tests increase p-value decrease strength of evidence. Central limit theorem: as n gets large, the sample mean or proportion becomes approximately normally distributed (bell-shaped, centered at the proportion under the null)
