Which test to use?

Question

HI,

I am fairly new to stats and need some help. I can't seem to determine which test to use in a certain question. As soon as I see unknown questions i dont know which test to use. T test, chi square, anova, Correlation or any other.

Answer 1

The type of test to use depends on what kind of analysis you want to perform and what kind of data you have and whether the assumptions for a particular type of test can be met.

Correlational tests look for the association between variables. Pearson correlation is strength of association between two continuous variables. Spearman correlation is strength of association between two ordinal variables. Chi-square is strength of association between two categorical variables.

Comparison of means. Paired t test - differences between two related variables. Independent t-test - differences between two independent variables. ANOVA - differences between group means after variance in outcome is accounted for.

Regression - simple regression - change in predictor variable predicts changed in level of outcome variable. Multiple regression - change in combination of two or more predictor variable predict level of change in outcome variable

Non-parametric - when data does not meet assumptions for above parametric tests. Wilcoxon rank-sum test - difference between two independent variable (takes into account magnitude and direction of difference). Wilcoxon sign-rank test - tests for differences between two related variables - takes into account magnitude and direction of difference. Sign test - tests if two related variables are different - ignores magnitude of change, only takes into account direction.