Establishing normative data for multi-trial memory tests: the multivariate regression-based approach.

Area of Science:

• Psychometrics
• Neuropsychology
• Statistical modeling

Background:

• Multi-trial memory tests are crucial for assessing memory and learning comprehensively.
• Current normative data analysis for these tests often uses univariate methods, which fail to account for correlated trial scores.
• This suboptimal approach leads to issues with statistical power, multiple testing, and parsimony.

Purpose of the Study:

• To propose and exemplify a multivariate regression-based approach for establishing normative data for multi-trial memory tests.
• To address the limitations of univariate analyses in handling correlated memory test scores.
• To provide a more statistically rigorous and parsimonious method for normative data development.

Main Methods:

• A multivariate regression-based approach was developed to analyze correlated data from multi-trial memory tests.
• The methodology was demonstrated using Rey's Auditory Verbal Learning Test data from 221 Dutch-speaking children (ages 5.82–15.49 years).
• An online appendix provides practical guidance for conducting analyses using R software.

Main Results:

• The proposed multivariate approach effectively accounts for the correlated nature of multi-trial memory test scores.
• The method avoids the pitfalls of multiple testing inherent in univariate analyses.
• The study successfully applied the multivariate method to a sample of children, generating normative data.

Conclusions:

• The multivariate regression-based approach offers significant methodological advantages over univariate methods for normative data.
• This advanced method allows for the testing of complex hypotheses, such as trial by covariate interactions, which are not feasible with univariate approaches.
• The approach provides a more robust and comprehensive framework for analyzing multi-trial memory test data.