The reliabilities of 6 item bias indices and correlations among them were investigated for each of the 11 tests of the Iowa Tests of Basic Skills (ITBS). The difficulty and delta indices detected group differences in relative item difficulty. Biserial and point biserial indices detected group differences in item discrimination. The Scheuneman and three-parameter indices detected differences in relative item difficulty by score level and latent ability level, respectively. The sample group consisted of 800 fifth grade students evenly divided between black and white, male and female students, thus examining racial and sexual bias. The results indicated the item bias indices investigated were unreliable when based on sample sizes of 200 minority and 200 majority examinees. The instability of the indices may have resulted from the fact that very few, if any, biased items are included in the ITBS. The study suggested that the use of item bias indices to screen achievement test items could not be expected to lead to consistent decisions about which items are biased with small sample sizes. (DWH)