Evaluating and Modifying Covariance Structure Models: A Review and Recommendation

摘要: The purpose of this article is to present a strategy for the evaluation and modification covariance structure models. approach makes use recent developments in estimation under non-standard conditions unified asymptotic theory related hypothesis testing. Factors affecting these models are reviewed terms nonnormality, missing data, specification error, sensitivity large sample size. Alternative model error search strategies also reviewed. modeling advocated utilizes LISREL index assessing statistical power, expected parameter change statistic guiding searches. It argued that common utilizing alternative fit indices does not allow investigator rule out plausible explanations misfit. allows one determine extent size effects by relying on existing underlying