A pilot study was conducted to examine selected noncognitive attributes to determine (1) if there were significant differences between norm groups for the attribute measures and the sample population, (2) if there were changes in these attributes as a result of participation in basic education instruction, (3) the association between the noncognitive attributes and independent variables, and (4) whether pretest and posttest scores on the noncognitive attribute measures could classify subjects as being participants and nonparticipants in the selected adult basic education programs. In 10 selected communities in North Carolina, a total of 486 disadvantaged adults were pretested and 381 were posttested. The data from 343 (279 participants and 64 nonparticipants) were used in this study. The test battery consisted of Rotter's I-E Scale, Haller's Work Beliefs Checklist, Rundquist and Sletto Minnesota Survey of Opinions (three scales), Moon and McCann's Scale and McClosky and Schaar's Scale, and Fitts' Tennessee Self-Concept Scale. The data were treated by one-way analysis of variance, least squares regression technique, and multivariate discriminate analysis. Results show: (1) no significant differences were detected between participants and nonparticipants in adult basic education; (2) independent variables associated with dependent variables were participation, sex, reported monthly income, and social participation scores; (3) the noncognitive measures were not effective in distinguishing participants. (Author/DB)