Differentiating categorical and dimensional data with taxometric analysis: are two variables better than none?

Area of Science:

• Psychology
• Statistics
• Quantitative Psychology

Background:

• Taxometric analyses are vital for differentiating categorical and dimensional data structures.
• Traditional taxometric methods often necessitate a minimum of three variables for analysis.
• The current study addresses the challenge of performing taxometric analyses with limited data, specifically when only two variables are available.

Purpose of the Study:

• To investigate the efficacy of taxometric analyses using only two variables.
• To determine if informative results can be obtained from two-variable taxometric analyses.
• To assess the utility of two-variable taxometric analysis in psychological research.

Main Methods:

• Extensive simulations were conducted to evaluate taxometric procedures with two variables.
• The mean above minus below a cut (MAMBAC) and maximum slope (MAXSLOPE) procedures were employed.
• Parallel analyses of comparison data were used to assess structural differentiation.

Main Results:

• Both MAMBAC and MAXSLOPE procedures successfully differentiated categorical and dimensional structures with only two variables.
• The study highlights the importance of indicator variability and large sample sizes for two-variable analyses.
• Application to childhood aggression data (Study 2) supported dimensional structure, illustrating practical utility.

Conclusions:

• Taxometric analyses can yield reliable results even with only two conceptually or empirically nonredundant variables.
• High-quality data and adequate sample sizes are essential for successful two-variable taxometric analyses.
• The findings support the confidence in using two-variable taxometric analyses when appropriate conditions are met.