An evaluation of several methods of predicting full scale IQ from the ITPA

The purpose of the study was to investigate the usefulness of the ITPA as a predictor of IQ. Three methods of estimating from the ITPA IQ were considered. One of these, the estimate of S-B IQ, is provided by the manual but lacked cross-validation. The other two estimates involve simpler methods which might be used more easily by an ITPA examiner. The ITPA and the Wechsler Intelligence Scale for Children were administered to 138 children. Two subgroups were also considered, the normative subgroup (IQ=85-115) and the low IQ subgroup (IQ=50-84). For each child, three estimates of IQ were obtained from the ITPA, the estimate of S-B IQ, the SS-derived IQ, and the PLQ. These three estimates were compared to the Wechsler FSIQ, Only the estimate of S-B IQ was not significantly different from FSIQ regardless of IQ range. Correlations between the FSIQ and each of the IQ estimates were substantial and comparable to the reliability reported for the ITPA (.87). After discussing the difference in light of statistical significance, the empirical significance of the results was considered. It was concluded that for practical purposes any one of the three IQ estimates will serve equally well.