Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




A considerable body of research has been developed in the area of children's problems and their classification. Nevertheless, none of the systems which have been studied have gained widespread acceptance. One of the biggest problems with previous research in this area has been the lack of data tying problem dimensions to age or developmental level. The purpose of the present research was the examination of year by year differences in 25 dimensions of children's behavior. Subjects consisted of 281 school children between the ages of 5 and 14. The Children's Behavioral Classification Project Instrument was administered to the parent or parent-surrogate of each child. Factor scores were calculated for each child on each of the 25 CBCP factors. Regression analysis was used to test the third-order model for each factor. The relationship between age and factor score was obtained for the total sample as well as for male and female samples separately. Significant results of the third-order polynomial regression were obtained on the total sample for only one factor.. Three factors were significant for males, and none was significant for females. Caution was used in interpreting the results on these factors as the percentages of variance accounted for by the third-order model were rather small. The results appear to indicate that the incidence of a variety of childhood problems and positive behaviors as unrelated to age. Although previous research has linked similar problems to age, those studies usually utilized single problems rather than constellations of behaviors. Within constellations of behavior, such as are found on the CBCP, there may be items which are age related as well as items which are independent of age. An instrument utilizing such heterogeneous factors may have better predictive validity insofar as it assesses both types of behaviors.