Four variations of programmed filmstrips were shown to deaf children to improve comprehension of the passive voice. The design included two training strategies that treated the passive voice either as a unitary structure or as a contrast to the active voice. Within strategies two orders of training were compared for the various types of passive sentences: (nonreversible, reversible, agent-deleted) and (agent-deleted, nonreversible, reversible). Six filmstrips were constructed for the two strategies and three types of passive. Evaluation tests were administered before, immediately after, and three months following training. Forty deaf children were selected from two age ranges, 9 to 12 and 13 to 16 years. Within age groups 10 subjects served under each of the four strategy-order conditions. Evaluation tests consisted of a multiple choice picture-sentence matching test and a performance test in which subjects acted out six passive sentences using toys.
Performance test scores increased significantly after training and surpassed pre-training after three months. Highest scores were achieved on nonreversible passives followed by reversible and agent-deleted passives. Reversal of agent and object was the most frequent error. Strategy and order differences were not significant.