Downloads

Return to Main

Technical Reports

• Jones, J. A. & Waller, N. G. (2013). The Normal-Theory and Asymptotic Distribution-Free (ADF) Covariance Matrix of Standardized Regression Coefficients: Theoretical Extensions and Finite Sample Behavior. Technical Report 052513. University of Minnesota, Twin Cities.

Handouts

• Handouts for Steve Boker's OpenMx Workshop

Computer Programs written by Niels Waller

• A Generalized Method for Computing Hierarchical Component Structures by Goldberg's Bass Ackwards Method:

  • Companion document.

• VERIPATH

  • VERIPATH: as described in Meehl and Waller, 2001. VERIPATH is a Mathematica program for testing the verisimilitude of path analysis models
  • The VERIPATH manual (pdf)
  • View the VERIPATH source code in pdf format

• DIFFERENTIAL ITEM FUNCTIONING

  • Programs for Differential Item Functioning (LINKDIF and EZDIF as described in Applied Psychological Measurement)

• FACTOR ANALYSIS: WinMFACT 2.0 Student Edition

  • WINCDStudent.zip
  This is a fully functional version of WinMFACT 2.0, the Windows version of MicroFACT 2.0. The student edition runs all example command files. It is restricted to data sets with fewer than 55 variables. The commercial version of WinMFACT 2.0 --- which handles an unlimited number of variables --- can be purchased through Assessment Systems Corporation.

  MFACT2 Full version

  (Standalone FORTRAN Executable -- no Windows GUI) now available for free (Please note that I am unable to answer software related questions) MFACT2

  MFACT2 Manual

  MFACT2 Manual

  MicroFACT is an easy-to-use factor analysis program that can tackle mainframe size problems on your PC quickly!

  MicroFACT provides a methodologically defensible and computationally efficient method for factor analysis of both dichotomous and polytomous data sets. Using virtual memory, it can handle analyses of data sets of any size. For dichotomous (i.e., test) data MicroFACT can assist you in verifying the dimensionality of your test (which is one of the most basic assumptions of the application of the item response theory model). Assessing the dimensionality of a test can be accomplished using a factor analytic approach if appropriate computations are employed. The linear factor analysis model as typically implemented in most statistical packages, however, often fails to use the appropriate measures of correlation for such a factor analysis (tetrachoric correlations). MicroFACT makes it easy to analyze the interrelationships among your test items to determine the number of underlying dimensions. MicroFACT is simple to use and provides detailed (but understandable) results. For the factor analysis of other types of data, such as polytomous data (i.e., ordered categorical data such as that found in surveys in the social sciences, political science, education, and behavioral sciences, and clinical measures used in medical, psychological and personality research). New Features in version 2

  • MicroFACT can be called from another program for Monte Carlo simulations.
  • Data may be written to external files to make simulations easier to conduct.
  • New rotation options: weighted varimax, Harris-Kaiser ortho-oblique, and Oblimin.
  • The ability to rotate from random orientations to avoid local minimum problems.
  • Can import factor pattern, structure, and factor intercorrelation matrices from previous analyses or published studies.
  • New model evaluation features include residual plots, goodness-of-fit indices, and hyperplane counts for rotated factors -- structures and patterns.
  • New factor loading plots.
  • Greater user control of output format.
  • New factor score plots.
  • Improved error handling routines.

R code for: Enhancing Power and Controlling Family-Wise Error in Electrocortical Studies; Yoder, Blackford, Waller, and Kim (2003).