Trimmed means: comparison of independent samples (Yuen-Welch test)

Command:

Statistics Trimmed means Comparison of independent samples

Description

The Yuen-Welch test is used to compare the trimmed means of two independent samples.

Required input

Select the variables for sample 1 and sample 2. Differences will be calculated as Sample2−Sample1.

Caveat: if the two variables are the same, then the two filters must define distinct groups so that the same case is not included in the two samples.

Options

• % Trimming: select the percentage of observations that will be trimmed away. For example, when you select 20% then the lowest 20% and highest 20% of observations will be dropped for the calculation of the trimmed mean.
• Confidence interval: select the required confidence interval for the difference between the trimmed means. A 95% confidence interval is the usual selection.

Results

The results windows displays the sample size, arithmecic mean and its 95% confidence interval, followed by the trimmed mean, the trimmed mean sample size and the 95% confidence interval of the trimmed mean for the 2 samples.

Yuen-Welch test

The report shows the difference between the trimmed means with its 95% confidence interval.

Next follow the test statistic t, the Degrees of Freedom (DF) and the two-tailed probability P. When the P-value is less than the conventional 0.05, the null hypothesis is rejected and the conclusion is that the two trimmed means do indeed differ significantly.

See Calculation of Trimmed Mean, SE and confidence interval for computational details.

Literature

• Wilcox RR (2022) Introduction to robust estimation and hypothesis testing. 5^th ed. Elsevier Academic Press.

See also

• Comparison of trimmed means of paired samples

Recommended book

Introduction to Robust Estimation and Hypothesis Testing
Rand R. Wilcox

Buy from Amazon

Introduction to Robust Estimating and Hypothesis Testing, Fifth Edition is a useful ‘how-to’ on the application of robust methods utilizing easy-to-use software. This trusted resource provides an overview of modern robust methods, including improved techniques for dealing with outliers, skewed distribution curvature, and heteroscedasticity that can provide substantial gains in power. Coverage includes techniques for comparing groups and measuring effect size, current methods for comparing quantiles, and expanded regression methods for both parametric and nonparametric techniques. The practical importance of these varied methods is illustrated using data from real world studies. Over 1700 R functions are included to support comprehension and practice.