Full information maximum likelihood is almost universallyabbreviated FIML, and it is often pronounced like "fimmle" if"fimmle" was an English word. FIML is often the ideal too

…l to usewhen your data contains missing values because FIML uses the rawdata as input and hence can use all the available information inthe data. This is opposed to other methods which use the observedcovariance matrix which necessarily contains less information thanthe raw data. An observed covariance matrix contains lessinformation than the raw data because one data set will alwaysproduce the same observed covariance matrix, but one covariancematrix could be generated by many different raw data sets.Mathematically, the mapping from a data set to a covariance matrixis not one-to-one (i.e. the function is non-injective), but rathermany-to-one. Although there is a loss of information between a raw data set andan observed covariance matrix, in structural equation modeling weare often only modeling the observed covariance matrix and theobserved means. We want to adjust the model parameters to make theobserved covariance and means matrices as close as possible to themodel-implied covariance and means matrices. Therefore, we areusually not concerned with the loss of information from raw data toobserved covariance matrix. However, when some raw data is missing,the standard maximum likelihood method for determining how closethe observed covariance and means matrices are to themodel-expected covariance and means matrices fails to use all ofthe information available in the raw data. This failure of maximumlikelihood (ML) estimation, as opposed to FIML, is due to MLexploiting for the sake of computational efficiency somemathematical properties of matrices that do not hold true in thepresence of missing data. The ML estimates are not wrong per se andwill converge to the FIML estimates, rather the ML estimates do notuse all the information available in the raw data to fit themodel. The intelligent handling of missing data is a primary reason to useFIML over other estimation techniques. The method by which FIMLhandles missing data involves filtering out missing values whenthey are present, and using only the data that are not missing in agiven row. ( Full Answer )