Studies on persuasive communication which employed elementary school children as receivers or senders were reviewed.
First, studies where children were only 'receivers' of communication were discussed. The following attributes were found to be problems of interest in previous studies; 1) topic, 2) control group method, 3) message, 4) receiver's age, 5) gender and individual differences, 6) communicator, 7) media, 8) indicator of the persuasion-effect, and 9) persuasion and modeling effect. Most notably, on the factor of 3) message, the tendency that persuasive appeals which cause reactance have small persuasive effect, and that appeals which arouse strong fear have large effect, was found commonly throughout the studies. Other factors did not show any remarkable common tendencies.
Secondly, studies where children were only 'senders' of communication were discussed. Most studies in this field aimed to clarify children's development in producing persuasive messages. Therefore, sender's 1) age, 2) communication-skill, and 3) other factors were discussed in relation to the produced message. As a conclusion, a major dependent variable, variability of message production, was related to age, and this relation was often discussed in relation to the role taking ability of the sender. One limitation of the use of hypothetical method, was also pointed out.
Finally, as for study of receiver, while persuasion itself can be said to be an effective method for children's attitudinal or behavioral change, more systematic research is required. As for study on sender, it was suggested that variability of message production, and the process of its selection and use should be investigated in future studies.