Statistics Cheatsheet

Appropriate Statistics for Different Scenarios

Correlation
— two continuous variables; the research question is whether the two variables are related

Multiple Regression
–three (or more) continuous variables, one of which is the “dependent,” “criterion,” or “outcome” variable; the others are “predictor” variables. The research question concerns how well the set of predictor variables predict scores on the dependent variable

Between-Groups t-test
–one two-level (dichotomous) categorical variable (e.g., gender) and one continuous variable (e.g., self-esteem scores)

Within-Groups t-test
–two continuous variables (comprising one two-level within-group factor); the research question is whether the means of the two variables are significantly different

One-way ANOVA
–one multi-level categorical variable (e.g., favorite color: red, blue, or orange) and one continuous outcome (dependent) variable (e.g., IQ scores)

Factorial Between-Group ANOVA
–two or more multi-level categorical variables (e.g., favorite color AND gender) and one continuous dependent variable (e.g., SAT scores)

Repeated-Measures ANOVA
–Multiple continuous variables (comprising one within-group factor), all from same participants; research question is whether the means across those variables differ (e.g., heart rates at time 1, 2, and 3)

Mixed ANOVA
–One multi-level categorical variable, one within-group variable, and one continuous dependent variable (e.g., favorite color (red, blue, orange), Time (1, 2, or 3), and SAT score)

Chi Square test for independence
–two or categorical variables; the goal of this analyis is to see if the pattern of frequencies at each level of one variable vary as a function of the levels of the other variable

Comments are closed.