This chapter deals with the potential that structural equation modeling (SEM) offers to behavioral and social researchers involved in scale construction and development, an activity in which they frequently engage in their quest for high-quality measuring instruments. The plan of the chapter is as follows. It begins with a discussion of a modeling approach that permits point and interval estimation of important indices characterizing an original set of items considered for inclusion in a scale of interest, attending also to issues pertaining to the study of their latent structure. Revision of resulting tentative instrument versions is discussed next, including reliability and criterion validity evaluation, as well as discussion of their change due to removal or addition of scale components. The discussion is subsequently extended to the concepts of maximal reliability and validity. Applications of the general modeling approach with missing data are finally attended to, as is the two-level setup where studied subjects are nested within higher-order units.(PsycINFO Database Record (c) 2019 APA, all rights reserved)