BIO SCI 100 Study Guide - Midterm Guide: Null Hypothesis, Normal Distribution, Chi-Squared Distribution
25 Mar 2017
School
Department
Course
Professor
Document Summary
Hypothesis testing: (know assumption of tests) t-test, chi square, anova-test null hypo, uses: possible diff b/w 2 sample. Don"t follow normal dist:median/range: std dev: variability of indiv data points (around mean of sample of pop), std error(sem): measure of precision for estimated pop mean. Confidence interval(ci): range of value with probability that true parameter is within: 95% ci: estimate that true value lies b/w these intervals. Not 95% probability: assumptions: values=indep, random sampling form pop that has gaussian distribution. More n = better: one sample t-test = diff b/w sample mean/known pop mean. >0. 5=not sig diff: statistical power: power=probability of correctly rejecting false null hypothesis. Miss effect if not enough trials: error occur: type 1 e: false positive(incorrect rejection of true null hypo). Type ii e: false neg(failure to reject false null hypo): student t test: stat significance of diff b/w two sample means. R=welch (2 pop/sample)=default: welch compared to student test does not assume equal variance.
