Biostats Flashcards
One-sample t-test
Parametric test
Compares the mean of the study sample to the population mean (group 1 mean vs. known population mean)
Two-sample t-test
Independent samples
Unpaired test
Parametric test
Compares means of two independent samples (group 1 vs. group 2)
Paired t-test
Related samples test
Parametric test
Compares mean difference of paired or matches samples. (group 1 –> two different measurements)
One-way ANOVA
Independent samples
Parametric test
Compares means of 3 or more groups
Example:
Young (group 1, group 2, group 3)
Two-way ANOVA
Parametric test Compares means of 3 or more groups with a factor Example: Young (group 1, group 2, group 3) Old (group 1, group 2, group 3)
Repeated measures ANOVA
Parametric test
Extension of a paired t-test
Group 1 (3 or more measurements)
Can also have a two-way repeated-measures ANOVA
Wilcoxon Rank Sum
Mann-Whitney U-test
Non-parametric test (ordinal)
Compares two independent samples
Kruskal-Wallis one-way ANOVA by ranks
Non-parametric test (ordinal)
Compares 3 or more independent samples
Wilcoxon Signed Rank Test
Sign Test
Non-parametric test (ordinal)
Compares 2 matched or paired samples
Friedman ANOVA by ranks
Non-parametric test (ordinal)
Compares 3 or more matched/paired groups
Chi-square (X^2) Test
Nominal data, non-parametric
Compares expected and observed proportions between two or more independent groups
- Goodness of fit, test of independence
Fisher Exact Test
Nominal Data, non-parametric
Use Chi-square test for small groups with < 5 observations
McNemar
Nominal Data, non-parametric Paired samples (2 related groups)
Mantel-Haenszel
Nominal Data, non-parametric
Controls for the influence of cofounders
Cox Proportional Hazards Model
Survival Analysis
Allows for calculation of hazard ratio & CI
Compares survival in two or more groups after adjusting for other variables
