OBJECTIVE The distribution of sensitivity across the visual field, as determined by automated threshold static perimetry, can be summarized in terms of visual field indices. Such indices can be weighted for the variation in threshold at each eccentricity. The aims of this study were to determine the influence of the weighting factor, the relationship between the unweighted indices derived from Programs 30-2 and 24-2, and the relationship between the number of double determinations of threshold and the magnitude of the short-term fluctuation. METHODS One visual field derived by Program 30-2 of the Humphrey Field Analyzer was selected from each of 60 consecutive patients with primary open-angle glaucoma. The first two fields from each individual patient were avoided. Unweighted visual field indices were calculated and compared with the Program 30-2 weighted indices using an assessment of agreement evaluated with respect to the 95% confidence limits of the population. RESULTS The weighting function had no influence on the mean deviation, but it caused a slight reduction in the short-term fluctuation and an elevation in the pattern and corrected pattern standard deviations. There was little difference between the indices generated by Programs 30-2 and 24-2. The short-term fluctuation increased with an increase in the number of double determinations of threshold. CONCLUSIONS The weighting function had little clinical influence on the visual field indices. The indices derived from Programs 30-2 and 24-2 were similar, and the short-term fluctuation would better reflect the intratest variability if all available double determinations of threshold were used to calculate the index.