Data gathered under various statewide testing programs in South Carolina were used to assess the effects of PET (data base) training for teachers on student achievement. Reading and mathematics achievement data for students in grades 1 through 4 in South Carolina were compared for teachers who had received PET training and those who did not. A total of 4,289 teachers' records were analyzed. The amount of training received by a typical PET trainee was small. Most of the PET teachers were in their first or second post-training year. There was too much residual variation in classroom achievement, as operationalized in the study, for the effects of PET training to be determined. Any differences existing at present are not educationally meaningful. It must be concluded that the Hunter Model has not been adequately evaluated. Three tables and one flowchart summarize the findings. (SLD)