Abstract
In covariance structure analysis statistics like the modification index and the “t-values” can be used for modifying a model. A model with an acceptable fit might be simplified, whereas a rejected model might be expanded or otherwise altered. This paper describes the results of Monte Carlo research in which the usefulness of a number of statistics is examined when a wrongly specified factor analysis model is modified in one parameter. For the samples where the correct model was obtained after a first modification step, the empirical distribution of the parameter estimates is examined. The conclusion is that low power and sampling variability in the decision criteria used for model modification, can be a serious threat to the reliability and validity of the decision which model to retain.
This research was supported by the Foundation of Social-Cultural sciences which is subsidized by the Netherlands Organization for the Advancement of Pure Research (Z. W. O.) under project number 500-278-003. The authors wish to thank the editor for his helpful comments during the research and on earlier versions of this paper.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Anderson, T. W., (1962). The choice of the degree of a polynomial regression as a multiple decision problem. Annals of Mathematical Statistics, 33, 255–265.
Bentler, P., & Bonett, D. (1980). Significance test and goodness of fit in the analysis of covariance structure. Psychological Bulletin, 88, 588–606.
Boomsma, A., (1983). On the robustness of LISREL (maximum likelihood estimation) against small sample size and nonnormality. Amsterdam: Sociometric Research Foundation.
Byron, R., (1972). Testing for misspecification in econometric systems using full information. International Econometric Review, 13, 745–756.
Costner, H., & Schoenberg, R. (1973). Diagnosing indicator ills in multiple indicator models. In A.S. Goldberger & O. Duncan (Eds.). Structural equation models in the social sciences. New York: Seminar Press, 168–200.
Cudeck, R., & Browne, M. (1983). Cross validation of covariance structures. Multivariate Behavioral Research, 18, 147–167.
James, L., Mulaik, A. & Brett, J. (1982). Causal analysis. Beverly Hills: Sage.
Joreskog, K.G., & Sörbom, D., (1984). Lisrel VI: Analysis of linear structural relationships by the method of maximum likelihood: User’s guide. Mooresville, Indiana: Scientific Software.
Kennedy, W. J. & Bancroft, T. A., (1971). Modelbuilding for prediction in regression based upon repeated significance tests. Annals of Mathematical Statistics, 42, 1273–1284.
Lawley, D.N., & Maxwell, A.E., (1971). Factor analysis as a statistical method. London: Butterworth.
Luijben, T.C., (1986a). Modification of models. Heymans Bulletin, HB-86-805-EX Rijksuniversiteit Groningen, Vakgroep Statistiek & Meettheorie.
Luijben, T.C., (1986b). Modification of factor analytical models in covariance structure analysis. Heymans Bulletin, HB-86-813-EX. Rijksuniversiteit Groningen, Vakgroep Statistiek & Meettheorie.
MacCallum, R., (1986). Specification searches in covariance structure modeling. Psychological Bulletin, 100, 107–120.
Malinvaud, E., (1980). Statistical methods of Econometrics, North-Holland: Amsterdam.
Saris, W., Zegwaart, P., & De Pijper, W., (1979). Detection of specification errors. In D.R. Heise (Ed.). Sociological methodology 1979. San Fransisco: Jossey-Bass, 151–171.
Saris, W., & Stronkhorst, H., (1984). Causal modelling in nonexperimental research. Amsterdam, The Netherlands: Sociometric research foundation.
Saris, W., Den Ronden, J., & Satorra, A., (1987). Testing structural equation models. In P. Cuttance & J. Ecob (Eds.). Structural modelling. Cambridge: Cambridge University Press, (in press)
Saris, W.E., Satorra, A., & Sörbom, D., (1987). The detection and correction of specification errors in structural models, (submitted for publication)
Satorra, A., & Saris, W.E., (1985). The power of the likelihood ratio test in covariance structure analysis. Psychometrika, 50, 83–90.
Sörbom, D., (1987). Model modification, (submitted for publication)
Steiger, J. H., Shapiro, A. & Browne, M., (1985). On the multivariate asymptotic distribution of sequential chi-square statistics. Psychometrika, 50, 253–264.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1988 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Luijben, T., Boomsma, A., Molenaar, I.W. (1988). Modification of Factor Analysis Models in Covariance Structure Analysis a Monte Carlo Study. In: Dijkstra, T.K. (eds) On Model Uncertainty and its Statistical Implications. Lecture Notes in Economics and Mathematical Systems, vol 307. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-61564-1_5
Download citation
DOI: https://doi.org/10.1007/978-3-642-61564-1_5
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-19367-8
Online ISBN: 978-3-642-61564-1
eBook Packages: Springer Book Archive