The Orthotran Solution

Area of Science:

• Psychometrics
• Computational Statistics
• Data Analysis

Background:

• Factor analysis is a statistical method used to describe variability among observed, correlated variables in terms of a potentially lower number of unobserved variables called factors.
• Determining oblique factor analytic solutions often involves subjective interpretation, leading to potential inconsistencies.
• Orthogonal transformation matrices are key in simplifying factor structures.

Purpose of the Study:

• To develop a computational algorithm, the orthotran solution, for objectively determining oblique factor analytic solutions.
• To assess the congruence and generalizability of the orthotran solution compared to subjective methods.

Main Methods:

• Development of the orthotran solution algorithm utilizing orthogonal transformation matrices.
• Application of a simple structure criterion within the orthotran solution.
• Determination of solutions for 17 diverse data sets.
• Comparison of orthotran-derived solutions with subjectively determined solutions.

Main Results:

• The orthotran solution successfully determined oblique factor analytic solutions for all 17 data sets.
• Orthotran-derived transformation solutions demonstrated significant congruence with subjectively obtained solutions.
• The algorithm's performance suggests robust data generalizability.

Conclusions:

• The orthotran solution provides an objective and reliable method for oblique factor analysis.
• The algorithm's congruence with subjective solutions validates its utility and generalizability in data analysis.
• This computational approach enhances the objectivity and consistency of factor analytic findings.