BIOL 155: when to use statistical tests

goodness of fit

compares observed frequencies to expected frequencies for one categorical variable to test whether observed frequencies differ from expected frequencies

test of independence

tests for a relationship between two or more categorical variables for which frequencies have been tabulated

one sample t test

tests a hypothesis about the mean of a sample (is the sample mean different from the population mean)

paired t test

compares two means from a paired design study to test if they differ

one way anova

compares the means of at least 3 or more independent samples to test if they statistically differ

two way anova

tests for an interaction between two or more categorical IVs and their individual effects

post hoc tests

makes separate comparisons of each pair of levels of the IV (pairwise comparisons)

correlation

tests for association, relationships between two numeric, measured variables, no IV or DV

binomial tests

uses the binomial distribution as the statistical model for testing a hypothesis about a proportion or probability