This study compared different strategies for computing sociometric status to determine the extent to which they agree. It is argued that, if the different strategies lead to significant differences in whether a child is classified as popular or unpopular, then the technique for computing social acceptance would be an important issue in interpreting studies of social skills and in clinical assessment. Only strategies using nomination or rating scale approaches were included because these have been most widely used with young children. Using photographs of their classmates, children ages 4 to 6 completed both a nomination and a rating sociometric task. Several strategies were then employed to compute sociometric rankings. In addition, teachers were asked to predict how children would be ranked by their peers. Children were ranked into either three categories (popular, unpopular, or middle) or four categories (popular, neglected, rejected, or middle) sociometric groups depending on the strategy employed. Kappa coefficients and percentages of agreement were then computed between the groupings yielded by the different strategies. While most of the kappa coefficients indicated agreement greater than chance, most of the actual percentages of agreement were below 50. This indicated that, at least for young children, which sociometric strategy is used may make a difference in how popular a child is judged to be. The report suggests that these findings imply that comparisons between studies using different strategies for computing peer acceptance must proceed with caution. (Author/DST)