Using raw-to-scaled-score conversions derived from test-score equating to link item-parameter estimates from the one-parameter (Rasch) and three-parameter logistic models, this study evaluated an indirect method for converting item response theory estimates to a common scale. Data were taken from Petersen's Scholastic Aptitude Test (SAT) scale stability study, which involved the chain equating of a test to itself through five intermediary forms. Sample sizes ranged from 2,527 to 2,879. Six SAT-verbal and six SAT-mathematical test forms were used. To estimate item parameters from individual data, the BICAL computer program was used for the one-parameter logistic model and the LOGIST program for the three-parameter model. The Quantile program estimated item parameters from grouped data for the three-parameter model. IRT true score equating provided a criterion to assess the practical consequences of using the three item calibration methods in conjunction with the score-conversion method of item-parameter transformation. Within-model and between-model comparisons showed close agreement with initial methods. While results were promising, further study of item parameter transformations based on score conversion information is needed before operational application in testing programs. (BS)