This forum is dedicated to understanding and discussing the statistical tests commonly used for analyzing continuous variables. Topics covered include the Student’s t-test, Mann–Whitney U test, and Wilcoxon signed-rank test.
🧪 1. Student’s t-test
Description:
The Student’s t-test is a parametric test used to compare the means between two groups.
Types:
-
Independent t-test: Compares means of two independent groups (e.g., treatment vs. control).
-
Paired t-test: Compares means of the same group at different times (e.g., before and after treatment).
Assumptions:
-
Data is continuous and normally distributed.
-
Homogeneity of variances.
-
Observations are independent.
Common Use Case:
Comparing mean systolic blood pressure between two treatment groups.
Discussion Thread Ideas:
-
How to test for normality before using the t-test?
-
What to do if assumptions are violated?
⚖️ 2. Mann–Whitney U Test (Wilcoxon Rank-Sum Test)
Description:
A non-parametric test used to compare differences between two independent groups when the dependent variable is either ordinal or continuous but not normally distributed.
Assumptions:
-
Independent observations.
-
Ordinal or continuous data.
-
Doesn’t require normal distribution.
When to Use:
-
Data is skewed or has outliers.
-
Sample sizes are small.
Example:
Comparing pain scores (non-normal data) between two different analgesic groups.
Discussion Thread Ideas:
-
Advantages of using Mann–Whitney over t-test.
-
How to interpret U values and p-values?
🔄 3. Wilcoxon Signed-Rank Test
Description:
A non-parametric test used for comparing two related samples or repeated measurements on a single sample to assess whether their population mean ranks differ.
Assumptions:
-
Paired data.
-
Symmetrical distribution of the differences.
-
Data at least ordinal.
When to Use:
-
Paired samples with non-normal distribution.
-
Pre-test and post-test designs.
Example:
Comparing cholesterol levels in patients before and after using a new drug.
Discussion Thread Ideas:
-
When to choose Wilcoxon over paired t-test?
-
Real-world examples using Wilcoxon signed-rank test.
