A Comparison of Alternatives to Conducting Monte Carlo Analyses for Determining Parallel Analysis Criteria.

The parallel analysis method for determining the number of components to retain in a principal components analysis has received a recent resurgence of support and interest. However, researchers and practitioners desiring to use this criterion have been hampered by the required Monte Carlo analyses needed to develop the criteria. Two recent attempts at presenting regression estimation methods to determine eigenvalues were found to be deficient in several respects, and less accurate in general, than a simple linear interpolation of tabled random data eigenvalues generated through Monte Carlo simulation. Other methods for determining the parallel analysis criteria are discussed.

[1]  W. Velicer Determining the number of components from the matrix of partial correlations , 1976 .

[2]  W. Velicer,et al.  Comparison of five rules for determining the number of components to retain. , 1986 .

[3]  D. Lawley TESTS OF SIGNIFICANCE FOR THE LATENT ROOTS OF COVARIANCE AND CORRELATION MATRICES , 1956 .

[4]  Norman Cliff,et al.  The eigenvalues-greater-than-one rule and the reliability of components. , 1988 .

[5]  S. N. Roy On a Heuristic Method of Test Construction and its use in Multivariate Analysis , 1953 .

[6]  M. Bartlett THE EFFECT OF STANDARDIZATION ON A χ2 APPROXIMATION IN FACTOR ANALYSIS , 1951 .

[7]  Richard G. Montanelli,et al.  An Investigation of the Parallel Analysis Criterion for Determining the Number of Common Factors , 1975 .

[8]  D. S. Clemm,et al.  Upper Percentage Points of the Individual Roots of the Wishart Matrix , 1972 .

[9]  H. Kaiser The Application of Electronic Computers to Factor Analysis , 1960 .

[10]  P. Hsu ON THE DISTRIBUTION OF ROOTS OF CERTAIN DETERMINANTAL EQUATIONS , 1939 .

[11]  P. Krishnaiah,et al.  Tables for the extreme roots of the wishart matrix , 1973 .

[12]  Lloyd G. Humphreys,et al.  Latent roots of random data correlation matrices with squared multiple correlations on the diagonal: A monte carlo study , 1976 .

[13]  Charles E. Lance,et al.  Parallel Analysis Criteria: Revised Equations for Estimating the Latent Roots of Random Data Correlation Matrices , 1989 .

[14]  P. Krishnaiah,et al.  On the exact distributions of the extreme roots of the Wishart and MANOVA matrices , 1971 .

[15]  S J Allen,et al.  Notes and Commentary: Regression Equations for the Latent Roots of Random Data Correlation Matrices with Unities on the Diagonal. , 1986, Multivariate behavioral research.

[16]  J. Horn A rationale and test for the number of factors in factor analysis , 1965, Psychometrika.