Examkin

Question

Difficulty: MediumInferential Statistical Tests

A surgical researcher is evaluating a novel topical biologic dressing against standard petrolatum gauze for the healing of chronic diabetic foot ulcers. Twenty-four patients with comparable baseline ulcer characteristics are randomly assigned in a 1:1 ratio to one of the two treatment protocols. The primary outcome measured is the absolute reduction in ulcer surface area (cm2cm^2) at 8 weeks. Statistical evaluation of the collected data indicates that the surface area reduction values are heavily right-skewed and do not follow a normal distribution. Which of the following statistical tests is most appropriate to determine if there is a statistically significant difference in the primary outcome between the two study groups?

  1. A
    Chi-square test
  2. B
    Independent samples t-test
  3. C
    Kruskal-Wallis test
  4. Mann-Whitney U testAnswer
  5. E
    Wilcoxon signed-rank test

Answer

Mann-Whitney U test
The correct answer is the Mann-Whitney U test. Statistical test selection depends on the data type, the number of groups, and the distribution of the data. In this study, the outcome variable is continuous (reduction in ulcer surface area in cm2cm^2) and is being compared between two independent groups (novel dressing vs. standard gauze). Because the data is explicitly stated to be heavily right-skewed and non-normal, a non-parametric test is required. The Mann-Whitney U test (also known as the Wilcoxon rank-sum test) is the non-parametric equivalent of the independent samples t-test and is the appropriate choice for these parameters.

Step-by-Step Solution

1
Identify the type of dependent variable.
The dependent variable is the absolute reduction in ulcer surface area (cm2cm^2), which is a continuous variable.
Variable type dictates whether to use tests for continuous data (e.g., t-test, ANOVA) or categorical data (e.g., Chi-square).
2
Determine the number and relationship of the study groups.
There are two separate groups of patients (biologic dressing vs. petrolatum gauze), meaning the data involves two independent groups.
This rules out tests for paired data and tests designed for three or more groups.
3
Evaluate the distribution of the data.
The data is heavily right-skewed and does not follow a normal distribution.
A non-normal distribution requires the use of non-parametric statistical tests.
4
Select the appropriate statistical test based on the prior steps.
The Mann-Whitney U test is selected.
It is the standard non-parametric test for comparing a continuous outcome between two independent groups.

Key Concept

Selecting the correct statistical test requires identifying the data type of the dependent variable, the number of independent/dependent groups, and whether the data distribution is normal or skewed.
Estimated Time:1m 0s
Rate this question

Topics

Inferential Statistical Tests

More questions from this topic