26134 Lecture Notes - Lecture 11: Statistical Hypothesis Testing, Dependent And Independent Variables, Block Design
Document Summary
Z formula for the difference between 2 sample means (independent, known pop variances): When population variance is not known, we must use: where degree of freedom = n 1 + n 2 - 2. Dependent samples are when observations in one population cannot be made independently of observations in another population, aka matched pair samples. t formula to test the difference between two dependent populations: Independent variable treatment or classification variable; aka factors. o. Three experimental designs that use anova are completely randomised design, randomised block design and factorial experiments. 30: observations are from normally distributed populations, observations represent random samples o observations in the populations have equal variances. In a completely randomised design, subjects are randomly selected and assigned to treatments. If only 2 treatment/classification levels of the independent variable, the design is the same as that used to test the difference between means of two independent populations, using t test.