The study investigated three models--the traditional classroom reading instructional model, a one to one tutorial model, and a reading resource center model--for providing reading instruction to 110 disadvantaged inner city children in junior high school educable mentally retarded (EMR) classes. The effectiveness of programmed instructional materials was also examined. The sample was divided into four experimental groups (EMRs assigned to individualized tutoring, EMRs assigned to the reading resource center, transition Ss assigned to individualized tutoring, and transition Ss assigned to the reading resource center) and two control groups (EMRs receiving no additional help and transition students receiving no additional help). Findings revealed that both experimental groups assigned to the resource class performed significantly better than did the control group; and because the posttest mean for the experimental tutoring groups was higher than the mean score of the experimental resource groups it can be inferred that the tutorial groups also performed better than the control groups. (SB)