TO DETERMINE THE RELATIONSHIPS between measures of oral language production and measures of listening skill, 2 groups, each with 48 black pupils, were randomly selected from the third grades in 2 schools. One was a lower class (Title I) school and the second a low middle class (Non-Title I) school. The subjects in each school were randomly assigned to 3 subgroups of 16 each. A black and a white adult male educator and several black peers administered the measures. For the combined sample, all correlations were significant at the .05 level. A Neuman-Keuls test applied post hoc to the means for the 3 testers revealed that the black adult tester elicited significantly more divergent English than either the white adult tester (q = 2.98, p < .05) or the black peer testers (q = 3.88, p<.05). The results of the 6 Bonferroni tests on total word production indicated that the testers were homogeneous within the Non-Title I school, but that the black adult and black peer testers elicited significantly more words (p < .01) within the Title I school than did the white adult tester.