The purpose of the study was twofold: first, to determine the reliability of a paired-associate (PA) task when used as a test of learning proficiency; and second, to assess the relationship between performance on the PA task and on IQ tests as a function of grade level (kindergarten, first, third) and socioeconomic status (high, low). The reliability of the PA task was acceptably high and the magnitude of the relationship between learning proficiency and intelligence varied with social class membership. Differences related to social class were also detected in rates of development of the two kinds of abilities assessed. (Author)