Using data from 135 sixth-, seventh-, and eighth-graders between 11 and 15 years old attending a middle school in a suburban Southwest Ohio school district, two hypothesized models of the factor structures for the Coopersmith Self-Esteem Inventory were tested. One model represents the original Coopersmith factor structure, and the other model is derived from an exploratory factor analysis of the data. Both models were tested using the EQS confirmatory factor analysis algorithm. Neither model defined an acceptable fit to the data. The EQS algorithm was then modified to iterate to a best fit model, non-significant chi square, through the systematic elimination of bad fit variables, statements, in a hypothesized model. The iterations resulted in a modification of both original hypothesized models, with the end result being that both modified models represented acceptable fits to the data. Both confirmed models are discussed in terms of their psychometric properties and in terms of their fit to the theory of self-concept. It is concluded that the confirmed exploratory model is superior to the confirmed Coopersmith model theoretically and psychometrically. Six tables and one graph present study data. A 20-item list of references is included. (Author/SLD)