Ordinal Pattern Analysis: A Tutorial on Assessing the Fit of Hypotheses to Individual Repeated Measures Data.

Area of Science:

• Statistics
• Quantitative Psychology
• Behavioral Science

Background:

• Parametric statistical methods like ANOVA and mixed-effects models are common.
• Analyzing individual characteristics and relative change requires specialized approaches.
• Existing methods may not fully capture individual-level patterns in relative change.

Purpose of the Study:

• Introduce ordinal pattern analysis as a novel statistical method.
• Demonstrate its utility for quantifying relative change at the individual level.
• Provide a tutorial for applying this method alongside familiar parametric techniques.

Main Methods:

• Tutorial introduction to ordinal pattern analysis.
• Demonstration with three analyses of increasing complexity.
• Use of provided code and data for reproducibility.

Main Results:

• Ordinal pattern analysis can be performed manually for small datasets.
• The method handles concepts like competing hypotheses, missing data, and group comparisons.
• Results are comparable to linear mixed-effects models, highlighting similarities and differences.

Conclusions:

• Ordinal pattern analysis provides interpretable results on hypothesis consistency.
• It quantifies hypothesis performance at individual and group levels.
• This method is a valuable addition for analyzing individual characteristics in relative terms.