For many models, however, the standard null model is an improper comparison model. The standard null model yields unconstrained estimates of the variance (and mean, if included) of each manifest variable. In structural equation modeling, incremental fit indices are based on the comparison of the fit of a substantive model to that of a null model. Missingness is usu-ally a nuisance, not the main focus of inquiry, but Why do missing data create such difficulty in sci-entific research? Because most data analysis proce-dures were not designed for them. Although not yet in the main-stream, these procedures may eventually extend the ML and MI methods that currently represent the state of the art. Newer developments are discussed, including some for dealing with missing data that are not MAR. They present, in both technical and practical language, 2 general approaches that come highly recommended: maximum likelihood (ML) and Bayes-ian multiple imputation (MI). They summarize the evidence against older procedures and, with few exceptions, dis-courage their use. They clear up common misunderstandings regarding the missing at random (MAR) concept. The authors frame the missing-data problem, review methods, offer advice, and raise issues that remain unresolved. Statistical procedures for missing data have vastly improved, yet misconception and unsound practice still abound.