To test the hypothesis that self-esteem correlates significantly with both achievement and I.Q. in the same direction, that is, either positively or negatively, 318 eighth grade and 478 fifth grade students were randomly selected. Stanford Achievement Tests and the Otis-Lennon Mental Abilities Test were administered to obtain the respective achievement and I.Q. scores. The Coopersmith Self-Esteem Inventory was used to measure the self-esteem of the fifth graders, while a comparable test was developed and validated by the author for the eighth graders. Factor analysis was used to determine whether a significant positive relationship existed between the three variables. I.Q correlated significantly with all the achievement subtests for the fifth graders, as expected, but not with self-esteem. Because self-esteem correlated positively with reading achievement and not I.Q., the hypothesis was rejected for this group. It was again rejected for the eighth graders--I.Q. correlated with only two of the math subtests and self-esteem correlated with none. However, a comparison between the correlation and the rotated factor matrices for eighth graders showed that the original problem and hypothesis were not as simple to resolve as it might appear; although I.Q. did not correlate with self-esteem, it had significant loadings on both factors designated as "ability" and "self-esteem." Math computation had a higher positive loading on the self-esteem factor than the ability factor. With this one exception, results indicated that I.Q. and self-esteem are not related generally. (CP)