Matching training tasks to abilities of people with mental retardation: a learning test versus experienced staff.

Three sets of predictions were compared concerning the ability of 18 persons with severe or moderate mental retardation to learn 12 training tasks. Predictions were made by (a) experienced teachers who had worked with the clients for a minimum of 8 months and (b) experienced teachers who did not know the clients personally but who interacted with them for 30 minutes and by (c) consideration of client performance on the Assessment of Basic Learning Abilities test. To teach each task to a pass or fail criterion, we applied standard training procedures. The Assessment of Basic Learning Abilities test was significantly more accurate for predicting client performance than was either group of experienced staff. Implications for decreasing the high cost of unproductive training time were discussed.