Statistical vector field analysis applied to mixed cross-sectional and longitudinal data

Area of Science:

• Psychology
• Biostatistics
• Gerontology

Background:

• Analyzing combined cross-sectional and longitudinal data reveals complex developmental patterns.
• Growth functions are influenced by initial measurement values and chronological age.
• Individual differences in developmental versus chronological age complicate single growth curve fitting.

Purpose of the Study:

• To present a novel visualization method for combined cross-sectional and longitudinal data.
• To address the distortions caused by individual differences in developmental age.
• To illustrate the method with examples of intellectual abilities and aging.

Main Methods:

• Introduction of the statistical vector field (svf) plot.
• Application of svf plots to a dataset of intellectual abilities and aging.
• Development of C source code for svf software.

Main Results:

• The svf plot effectively visualizes complex patterns of change in growth data.
• Simultaneous visualization of sampling densities (cross-sectional) and individual change trajectories (longitudinal).
• Demonstrated utility in analyzing intellectual abilities across the aging spectrum.

Conclusions:

• The statistical vector field plot offers a powerful tool for understanding developmental trajectories.
• This method enhances the analysis of combined cross-sectional and longitudinal datasets.
• The svf approach provides a clearer representation of individual differences in aging and cognitive changes.