An evaluation of the Pittsburgh Public Schools preprimary program for 2,000 disadvantaged children concluded that the program contributed to the socioemotional maturation of the children involved but did not affect their reading readiness or first grade reading test scores. This finding may be explained in part by a lack of specific academic achievement objectives in the program design. The evaluation processes also showed a need for a more detailed definition of desired teacher behavior. Each teacher was observed for 1 hour on two separate occasions in accordance with an observation schedule, to find out the number and kind of adult-child interactions in her classroom. Each child was rated by his teacher on the Children's Rating Scale developed in the Pittsburgh project. Analysis of the data led to the recommendation that more specific guidelines be established for classroom personnel concerning their duties, ways of reinforcing learning behavior, use of individualized instruction, and academic skill development. Over half of this report is made up of appendixes, which include a detailed description of the primary program and facsimiles of the rating scale and observation schedule. (MS)