Abstract
Applications of structural equation modeling (SEM) may encounter issues like inadmissible parameter estimates, nonconvergence, or unsatisfactory model fit. We propose a new factor rotation method that reparameterizes the factor correlation matrix in exploratory factor analysis (EFA) such that factors can be either exogenous or endogenous. The proposed method is an oblique rotation method for EFA, but it allows directional structural paths among factors. We thus referred it to as FSP (factor structural paths) rotation. In particular, we can use FSP rotation to “translate” an SEM model to incorporate theoretical expectations on both factor loadings and structural parameters. We illustrate FSP rotation with an empirical example and explore its statistical properties with simulated data. The results include that (1) EFA with FSP rotation tends to fit data better and encounters fewer Heywood cases than SEM does when there are cross-loadings and many small nonzero loadings, (2) FSP rotated parameter estimates are satisfactory for small models, and (3) FSP rotated parameter estimates are more satisfactory for large models when the structural parameter matrices are sparse.
Similar content being viewed by others
Notes
We count diagonal elements of a correlation matrix as unique elements because they are standardized variances. Rotational indeterminacy requires \(m^2\) constraints in EFA. In orthogonal rotation, there are \(m(m+1)/2\) constraints in the factor correlation matrix and \(m(m-1)/2\) in the factor loading matrix (Archer & Jennrich 1973, Eq. (17)). The \(m(m+1)/2\) constraints in the factor correlation matrix are m diagonal elements of ones and \(m(m-1)/2\) above-diagonal (or below-diagonal) elements of zeros. In oblique rotation, there are m constraints in the factor correlation matrix and \(m(m-1)\) constraints imposed on the factor loading matrix and the factor correlation matrix jointly (Jennrich, 1973, Eq. (28)). The m constraints in the factor correlation matrix are m diagonal elements of ones. Therefore, we need to set the factor variances to be one for both simple interpretation and model identification.
The two discrepancy functions are \(f_{OLS}(\varvec{r}, \varvec{\theta }) = \text {trace} [\varvec{R}- \varvec{P}(\varvec{\theta })]^2\) and \( f_{ML}(\varvec{r}, \varvec{\theta }) = \log |\varvec{P}(\varvec{\theta })|- \log |\varvec{R}| + \text {trace}[\varvec{R}\varvec{P}(\varvec{\theta })^{-1}] - p\), respectively. The matrix function “trace” sums together the diagonal elements of a square matrix.
The derivation presented in Appendix A of Zhang et al. (2019) is also applicable to FSP rotation, and their Equation (20) directly implies the current constraint functions.
The target values for factor loadings and structural parameters are presented in an online support file (https://drive.google.com/file/d/1u-XHXlFHyjDumxJC8UN-JNzjAISuC9t4/view?usp=sharing,Table B1). The file also includes the R code for the illustration.
We also computed mean RMSEAs with all samples. Including or excluding samples with Heywood cases produced essentially the same results.
The FSP.target rotated population values and the FSP.geomin rotated population values of Model II are reported as supplementary materials (Table B4).
The FSP.geomin rotated population values and the FSP.target rotated population values of Model III are reported as supplementary materials (Table B5).
Tables B6 and B7 of Supplemental materials report such an example.
References
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411–423. https://doi.org/10.1037/0033-2909.103.3.411.
Archer, C. O., & Jennrich, R. I. (1973). Standard errors for orthogonally rotated factor loadings. Psychometrika, 38, 581–592. https://doi.org/10.1007/bf02291496
Asparouhov, T., & Muthén, B. (2009). Exploratory structural equation modeling. Structural Equation Modeling, 16, 397–438. https://doi.org/10.1080/10705510903008204
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley. https://doi.org/10.1002/9781118619179.
Browne, M. W. (1972). Oblique rotation to a partially specified target. British Journal of Mathematical and Statistical Psychology, 25, 207–212. https://doi.org/10.1111/j.2044-8317.1972.tb00492.x
Browne, M. W. (2001). An overview of analytic rotation in exploratory factor analysis. Multivariate Behavioral Research, 36, 111–150. https://doi.org/10.1207/s15327906mbr3601_05
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–162). Newbury Park, CA: Sage.
Browne, M. W., & Shapiro, A. (1986). The asymptotic covariance matrix of sample correlation coefficients under general conditions. Linear Algebra and Its Applications, 82, 169–176. https://doi.org/10.1016/0024-3795(86)90150-3.
Clarkson, D. B. (1979). Estimating the standard errors of rotated factor loadings by jackknifing. Psychometrika, 44, 297–314. https://doi.org/10.1007/bf02294695
Crawford, C. B., & Ferguson, G. A. (1970). A general rotation criterion and its use in orthogonal rotation. Psychometrika, 35, 321–332. https://doi.org/10.1007/bf02310792
Cudeck, R., & O’Dell, L. L. (1994). Applications of standard error estimates in unrestricted factor analysis: Significance tests for factor loadings and correlations. Psychological Bulletin, 115, 475–487. https://doi.org/10.1037//0033-2909.115.3.475
Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Mahwah, NJ: Lawrence Erlbaum Associates. https://doi.org/10.4324/9780203781098.
Hayashi, K., & Yung, Y.-F. (1999). Standard errors for the class of orthomax-rotated factor loadings: some matrix results. Psychometrika, 64, 451–460. https://doi.org/10.1007/bf02294566
Ichikawa, M., & Konishi, S. (1995). Application of the bootstrap methods in factor analysis. Psychometrika, 60, 77–93. https://doi.org/10.1007/bf02294430
Jennrich, R. I. (1973). Standard errors for obliquely rotated factor loadings. Psychometrika, 38, 593–604. https://doi.org/10.1007/bf02291497
Jennrich, R. I. (1974). Simplified formulae for standard errors in maximum-likelihood factor analysis. British Journal of Mathematical and Statistical Psychology, 27, 122–131. https://doi.org/10.1111/j.2044-8317.1974.tb00533.x
Jennrich, R. I. (2002). A simple general method for oblique rotation. Psychometrika, 67, 7–19. https://doi.org/10.1007/bf02294706
Jennrich, R. I. (2004). Derivative free gradient projection algorithms for rotation. Psychometrika, 69, 475–480. https://doi.org/10.1007/BF02295647
Jennrich, R. I. (2007). Rotation methods, algorithms, and standard errors. In R. Cudeck & R. C. MacCallum (Eds.), Factor analysis at 100: Historical developments and future directions (pp. 315–335). Mahwah, NJ: Lawrence Erlbaum Associates.
Johnson, R. A., & Wichern, D. W. (2007). Applied multivariate statistical analysis (6th ed.). Upper Saddle River, NJ: Pearson Prentic Hall.
Jöreskog, K. G. (1994). On the estimation of polychoric correlations and their asymptotic covariance matrix. Psychometrika, 59, 381–389. https://doi.org/10.1007/bf02296131
Kaiser, H. F. (1958). The varimax criterion for analytic rotation in factor analysis. Psychometrika, 23, 187–200. https://doi.org/10.1007/bf02289233
Lee, C.-T., Zhang, G., & Edwards, M. C. (2012). Ordinary least squares estimation of parameters in exploratory factor analysis with ordinal data. Multivariate Behavioral Research, 47, 314–339. https://doi.org/10.1080/00273171.2012.658340.
Lee, S. Y., Xong, X.-Y., & Tang, N.-S. (2007). Bayesian methods for analyzing structural equation models with covariates, interaction, and quadratic latent variables. Strucutral Equation Modeling: A Multidisciplinary Journal, 14, 404–434. https://doi.org/10.1080/10705510701301511
MacCallum, R. C. (1986). Specification searches in covariance structure modeling. Psychological Bulletin, 100, 107–120. https://doi.org/10.1037/0033-2909.100.1.107
MacCallum, R. C. (2003). Working with imperfect models. Multivariate Behavioral Research, 38, 113–139. https://doi.org/10.1207/s15327906mbr3801_5
MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis. Psychological Methods, 4, 84–99. https://doi.org/10.1037//1082-989x.4.1.84.
Merkle, E. C., Fitzsimmons, E., Uanhoro, J., & Goodrich, B. (2021). Efficient Bayesian structural equation modeling in Stan. Journal of Statistical Software, 100(6), 1–22. https://doi.org/10.18637/jss.v100.i06.
Merkle, E. C., & Rosseel, Y. (2018). blavaan: Bayesian structural equation models via parameter expansion. Journal of Statistical Software, 85, 1–30. https://doi.org/10.18637/jss.v085.i04.
Mulaik, S. A. (2010). Foundations of factor analysis (2nd ed.). Boca Raton, FL: Chapman & Hall/CRC.
Muthén, B., & Asparouhov, T. (2012). Bayesian structural equation modeling: A more flexible representation of substantive theory. Psychological Methods, 17, 313–335. https://doi.org/10.1037/a0026802.
Ogasawara, H. (2007). Asymptotic expansion of the distributions of the estimators in factor analysis under non-normality. British Journal of Mathematical and Statistical Psychology, 60, 395–420. https://doi.org/10.1348/000711006x110904
Pennell, R. (1972). Routinely computable confidence intervals for factor loadings using the “jackknife”. British Journal of Mathematical and Statistical Psychology, 25, 107–114. https://doi.org/10.1111/j.2044-8317.1972.tb00481.x
Reisenzein, R. (1986). A structural equation analysis of Weiner’s attribution-affect model of helping behavior. Journal of Personality and Social Psychology, 50, 1123–1133. https://doi.org/10.1037/0022-3514.50.6.1123
Steiger, J. H., & Hakstian, A. R. (1982). The asymptotic distribution of elements of a correlation matrix: theory and application. British Journal of Mathematical and Statistical Psychology, 35, 208–215. https://doi.org/10.1111/j.2044-8317.1982.tb00653.x
Swain, A. J. (1975). A class of factor analysis estimation procedures with common asymptotic sampling properties. Psychometrika, 40, 315–335. https://doi.org/10.1007/BF02291761
Thurstone, L. L. (1947). Multiple factor analysis. Chicago: University of Chicago Press.
Yates, A. (1987). Multivariate exploratory data analysis: A perspective on exploratory factor analysis. Albany, NY: State University of New York Press.
Yuan, K.-H., Marshall, L. L., & Bentler, P. M. (2002). A unified approach to exploratory factor analysis with missing data, nonnormal data, and in the presence of outliers. Psychometrika, 67, 95–121. https://doi.org/10.1007/bf02294711
Zhang, G., Hattori, M., Trichtinger, L., & Wang, X. (2019). Target rotation with both factor loadings and factor correlations. Psychological Methods, 24, 390–420. https://doi.org/10.1037/met0000198
Zhang, G., Jiang, G., Hattori, M., & Trichtinger, L. (2020). Utility functions for exploratory factor analysis [Computer software manual]. Retrieved from https://cran.r-project.org/web/packages/EFAutilities/EFAutilities.pdf (version 2.1.1)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix
Derivatives of the FSP.target Rotation Criterion
The derivatives of the FSP.target rotation criterion Q with regard to the rotation matrix \(\varvec{T}\) involve four terms. The first term involves factor loadings, and it was described in Jennrich (2002). The fourth term involves correlations among exogenous factors, and it was described in Zhang et al. (2019). We now derive the second term tr\(\left( \frac{\mathrm{d}Q}{\mathrm{d} \varvec{B}'} \frac{\mathrm{d} \varvec{B}}{\mathrm{d} t_{kl}}\right) \) and the third term tr\(\left( \frac{\mathrm{d}Q}{\mathrm{d} \varvec{\Gamma }'} \frac{\mathrm{d} \varvec{\Gamma }}{\mathrm{d} t_{kl}} \right) \).
We compute both terms using the chain rule, which involves the product of the derivatives of the semtarget rotation criterion function Q with regard to \(\varvec{B}\) or \(\varvec{\Gamma }\) and the derivatives of \(\varvec{B}\) or \(\varvec{\Gamma }\) with regard to \(\varvec{T}\). The derivatives of the semtarget rotation criterion function Q with regard to \(\varvec{B}\) and \(\varvec{\Gamma }\) are
We use the chain rule again to derive the derivatives of \(\varvec{B}\) and \(\varvec{\Gamma }\) with regard to \(t_{kl}\),
Let \(\varvec{w}_i\) contains the all regression weights of predicting \(\eta _i\) from all its predictors. These regression weights correspond to the elements at the ith row of \(\varvec{B}\) and \(\varvec{\Gamma }\). The derivatives \(\frac{\mathrm{d} w_{ij}}{\mathrm{d} \phi _{kl}}\) are
Here, \(w_{ik}\) is the kth element of the vector \(\varvec{w}_i\), and \(\varvec{\Phi }_{i+, i+}\) is defined in Eq. (4).
The partial derivatives of EFA factor correlations \(\varvec{\Phi }_z\) with regard to \(\varvec{T}\) are
Here, \(\phi _{ij}\) is a typical element of \(\varvec{\Phi }_z\). If \(i=l\), the value \(\delta _ {il}\) is 1, and it is zero otherwise.
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhang, G., Hattori, M. & Trichtinger, L.A. Rotating Factors to Simplify Their Structural Paths. Psychometrika 88, 865–887 (2023). https://doi.org/10.1007/s11336-022-09877-3
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11336-022-09877-3