Significance Tests and Goodness of Fit in the Analysis of Covariance Structures

Factor analysis, path analysis, structural equation modeling, and related multivariate statistical methods are based on maximum likelihood or generalized least squares estimation developed for covariance structure models. Large-sample theory provides a chi-square goodness-of-fit test for comparing a model against a general alternative model based on correlated variables. This model comparison is insufficient for model evaluation: In large samples virtually any model tends to be rejected as inadequate, and in small samples various competing models, if evaluated, might be equally acceptable. A general null model based on modified independence among variables is proposed to provide an additional reference point for the statistical and scientific evaluation of covariance structure models. Use of the null model in the context of a procedure that sequentially evaluates the statistical necessity of various sets of parameters places statistical methods in covariance structure analysis into a more complete framework. The concepts of ideal models and pseudo chi-square tests are introduced, and their roles in hypothesis testing are developed. The importance of supplementing statistical evaluation with incremental fit indices associated with the comparison of hierarchical models is also emphasized. Normed and nonnormed fit indices are developed and illustrated.

[1]  E. S. Pearson,et al.  ON THE USE AND INTERPRETATION OF CERTAIN TEST CRITERIA FOR PURPOSES OF STATISTICAL INFERENCE PART I , 1928 .

[2]  S. S. Wilks.  The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses , 1938 .

[3]  D. Lawley.  VI.—The Estimation of Factor Loadings by the Method of Maximum Likelihood , 1940 .

[4]  Herman Rubin,et al.  Statistical Inference in Factor Analysis , 1956 .

[5]  W. W. Rozeboom.  The fallacy of the null-hypothesis significance test. , 1960, Psychological bulletin.

[6]  A. Basilevsky,et al.  Factor Analysis as a Statistical Method. , 1964 .

[7]  W. Hays.  Statistics for psychologists , 1963 .

[8]  D. Bakan,et al.  The test of significance in psychological research. , 1966, Psychological bulletin.

[9]  K. Jöreskog.  Some contributions to maximum likelihood factor analysis , 1967 .

[10]  D. Lykken.  Statistical significance in psychological research. , 1968, Psychological bulletin.

[11]  K. Jöreskog.  A general approach to confirmatory maximum likelihood factor analysis , 1969 .

[12]  H. O. Lancaster.  The chi-squared distribution , 1971 .

[13]  Karl G. Jöreskog,et al.  FACTORIAL INVARIANCE OF ABILITY MEASURES IN GROUPS DIFFERING IN INTELLIGENCE AND SOCIOECONOMIC STATUS , 1971 .

[14]  Denton E. Morrison,et al.  The Significance Test Controversy , 1972 .

[15]  A. Goldberger,et al.  Factor analysis by generalized least squares , 1972 .

[16]  M. Browne.  Generalized Least Squares Estimators in the Analysis of Covariance Structures. , 1973 .

[17]  L. Tucker,et al.  A reliability coefficient for maximum likelihood factor analysis , 1973 .

[18]  A. E. Maxwell,et al.  Factor Analysis as a Statistical Method. , 1964 .

[19]  T. W. Anderson.  Asymptotically Efficient Estimation of Covariance Matrices with Linear Structure , 1973 .

[20]  A. Goldberger,et al.  Structural Equation Models in the Social Sciences. , 1974 .

[21]  L. Guttman.  What is Not What in Statistics , 1977 .

[22]  S. Mulaik,et al.  Single-sample tests for many correlations. , 1977 .

[23]  B. Muthén,et al.  Assessing Reliability and Stability in Panel Models , 1977 .

[24]  William T. Bielby,et al.  Structural Equation Models , 1977 .

[25]  L R Bergman,et al.  A Longitudinal Factor Model For Studying Change In Ability Structure. , 1977, Multivariate behavioral research.

[26]  D. Kandel.  Longitudinal research on drug use: Empirical findings and methodological issues , 1978 .

[27]  K. Jöreskog.  Structural analysis of covariance and correlation matrices , 1978 .

[28]  P. Whittle,et al.  Latent Variables in Socio‐Economic Models , 1978 .

[29]  R. P. Carver.  The Case Against Statistical Significance Testing , 1978 .

[30]  Dennis J. Aigner,et al.  Latent variables in socio-economic models , 1978 .

[31]  D. A. Kenny,et al.  Correlation and causality , 1979 .

[32]  Jay Magidson,et al.  Advances in factor analysis and structural equation models , 1979 .

[33]  Peter M. Bentler,et al.  A statistical development of three-mode factor analysis , 1979 .

[34]  D. G. Weeks,et al.  Interrelations Among Models For The Analysis Of Moment Structures. , 1979, Multivariate behavioral research.

[35]  U. Olsson.  On The Robustness Of Factor Analysis Against Crude Classification Of The Observations. , 1979, Multivariate behavioral research.

[36]  P. Bentler.  MULTIVARIATE ANALYSIS WITH LATENT VARIABLES: CAUSAL MODELING , 1980 .

[37]  D. G. Weeks,et al.  Linear structural equations with latent variables , 1980 .

[38]  J. Geweke,et al.  Interpreting the Likelihood Ratio Statistic in Factor Models When Sample Size is Small , 1980 .

[39]  Bentler,et al.  Some asymptotic properties of constrained generalized least squares estimation in coy ariance structure models , 1980 .