PC program for simultaneously testing equality of means and variances for paired bivariate normal data

Area of Science:

• Statistics
• Biostatistics
• Statistical Software Development

Background:

• Existing statistical methods for independent samples widely cover tests for equality of means and variances.
• For paired bivariate normal data, only the test for mean equality (paired t-test) is commonly implemented.
• Tests for equality of variances in paired data are less recognized and often absent in statistical packages.

Purpose of the Study:

• To describe, illustrate, and provide a menu-driven PC program for simultaneously testing the equality of means and variances.
• To address the underrepresentation of variability testing for paired data in statistical software.
• To offer a practical tool for researchers working with paired bivariate normal data.

Main Methods:

• Development of a menu-driven PC program.
• Implementation of simultaneous tests for mean and variance equality.
• Focus on paired bivariate normal data.

Main Results:

• A functional PC program is now available for simultaneous testing of mean and variance equality in paired data.
• The program facilitates the assessment of both central tendency and dispersion for paired observations.
• Demonstrates the practical application of these statistical tests.

Conclusions:

• The developed program fills a critical gap in statistical software for paired data analysis.
• Researchers can now more readily test for changes in both means and variances in paired bivariate normal data.
• Enhances the capability for comprehensive statistical inference in paired study designs.