Examkin

Question

Difficulty: Very hardInferential Statistical Tests

A public health research team is conducting a cross-sectional study to evaluate patient satisfaction across three distinct models of primary care delivery: federally qualified community health centers, hospital-affiliated outpatient clinics, and private physician practices. A random sample of 150 patients (50 from each care model) completes a standardized evaluation tool. The primary outcome is the patients' overall satisfaction, recorded as a single response on a 5-point Likert scale ranging from 1 ("Completely Dissatisfied") to 5 ("Completely Satisfied"). The investigators aim to determine if there is a statistically significant difference in the distribution of satisfaction scores among the three primary care models. Which of the following is the most appropriate statistical test to conduct this analysis?

  1. Kruskal-Wallis testAnswer
  2. B
    One-way analysis of variance
  3. C
    Chi-square test of independence
  4. D
    Mann-Whitney U test
  5. E
    Friedman test

Answer

Kruskal-Wallis test
The correct answer is the Kruskal-Wallis test. The study involves comparing a single dependent variable across three independent groups. Because the dependent variable (satisfaction) is measured on a 5-point Likert scale, it represents ordinal data. In ordinal data, categories have a meaningful rank order, but the mathematical intervals between them are not necessarily equal, making parametric tests invalid. The Kruskal-Wallis test is the non-parametric equivalent of the one-way ANOVA and is the optimal statistical method to analyze ordinal data across three or more independent groups.

Step-by-Step Solution

1
Identify the type of the dependent (outcome) variable.
The outcome is patient satisfaction measured on a 5-point Likert scale, which is an ordinal variable.
Determining the data type dictates whether parametric or non-parametric statistical testing is required. Ordinal data requires non-parametric methods.
2
Identify the independent (predictor) variable and the number of groups being compared.
The predictor is the primary care model, consisting of three distinct groups.
The number of groups differentiates between tests designed for two groups versus tests designed for three or more groups.
3
Determine if the study groups are independent or dependent (paired/matched).
The groups are independent because they consist of distinct patients (50 unique patients per clinic type).
This establishes whether an independent samples test or a repeated-measures test is indicated.
4
Select the appropriate statistical test based on the identified variables and study design.
The Kruskal-Wallis test is chosen.
It is the designated non-parametric test for comparing an ordinal outcome variable across three or more independent groups.

Key Concept

The Kruskal-Wallis test is the non-parametric equivalent of the one-way ANOVA, used to compare an ordinal dependent variable (or continuous non-normal data) across three or more independent groups.
Rate this question

Topics

Inferential Statistical Tests

More questions from this topic