The Kruskal-Wallis H test is used to compare three or more independent groups by ranking data because it:
Free Applied Statistics in Healthcare practice question on Non-parametric tests and chi-square for BSN and Post-RN nursing students. Choose your answer, then open the answer and the explanation.
Acompares group medians without assuming a normal distribution
Bmeasures the strength of a linear relationship between variables
Ccompares group means while assuming equal population variances
Dtests association between two categorical variables in a table
Show answer and explanation
✓ Correct answerA. compares group medians without assuming a normal distribution
Explanation
The Kruskal-Wallis H test is a non-parametric alternative to one-way ANOVA. It ranks all observations together across the groups and compares the average ranks, so it does not require normally distributed data or equal variances and is used for three or more independent groups.
Why each option is right or wrong
✓A. compares group medians without assuming a normal distribution — Ranks replace raw values, so the test compares group medians and needs no normality assumption.
✗B. measures the strength of a linear relationship between variables — Strength of a linear relationship is measured by correlation, such as Pearson r.
✗C. compares group means while assuming equal population variances — Comparing means with equal variances is the job of one-way ANOVA, a parametric test.
✗D. tests association between two categorical variables in a table — Association between two categorical variables is tested with the chi-square test, not by ranking.
Test
Purpose
Data / assumption
Kruskal-Wallis H
Compare 3+ independent groups
Ordinal or non-normal; no normality needed
One-way ANOVA
Compare 3+ independent group means
Continuous, normal, equal variances
Chi-square
Association of two categorical variables
Frequency counts in categories
Pearson correlation
Strength of linear relationship
Two continuous variables, bivariate normal
EXAM HINT Kruskal-Wallis = non-parametric ANOVA: ranks instead of means, medians instead of averages.
اردو وضاحت
Kruskal-Wallis H test کیا ہے؟
یہ ایک non-parametric test ہے جو تین یا زیادہ آزاد groups کو compare کرنے کے لیے استعمال ہوتا ہے۔ اس میں ڈیٹا کو normality کی شرط کے بغیر ranks میں بدل کر دیکھا جاتا ہے۔
یہ test ranks کیوں استعمال کرتا ہے؟
کیونکہ یہ groups کے درمیان فرق means کی بجائے ranks اور medians کی بنیاد پر دیکھتا ہے، اس لیے normal distribution کی ضرورت نہیں رہتی۔ یہی وجہ ہے کہ skewed یا ordinal ڈیٹا پر بھی یہ درست رہتا ہے۔
Key terms
medians — the middle value of a set of numbers when they are arranged in order (اعداد کو ترتیب سے رکھنے پر درمیان والی value)