A Monte Carlo study was conducted to estimate the small sample standard errors and statistical bias of psychometric statistics commonly used in the analysis of achievement tests. The statistics examined in this research were: (1) the index of item difficulty; (2) the index of item discrimination; (3) the corrected item-total point-biserial correlation coefficient; and (4) coefficient alpha. Sample sizes of 5, 10, 20, 40, 80, and 160 were evaluated. One thousand samples of each size were drawn with replacement from each of 10 archival data files from teacher subject area tests. These files represent pseudo-populations whose parameters are directly calculable and from which the sampling bias and errors of statistics are empirically estimable. The behavior of each statistic was evaluated by computing the standard error of the statistic for each sample size and each pseudo-population, and by computing the statistical bias of the statistic for each sample size and each pseudo-population. Results are interpreted in terms of their applications to test development. Nine tables present study data, and nine figures illustrate the discussion. There is a 13-item list of references. (Author/SLD)