The purpose of this study was to verify factors which are related to the provision of special education services in Virginia school divisions and to compare obtained results with similar research completed in the era of permissive legislation by Chalfant in 1965 in Illinois. A diagnostic technique was developed to indicate the proportions of children identified as handicapped expected to be found in a school division. Indicator variables selected for inclusion in the study were average daily membership, population per square mile, median school years completed, percentage of total civilian labor force unemployed, true value of property, percent black, median household income, percentage of families in urban residence, and the Composite Index. Criteria measures included the proportion of students identified as mentally retarded, learning disabled, speech impaired, emotionally disturbed, and total proportion of all handicapped students. Indicator variables found significant to the proportion of learning disabled students were median school years completed, percent of the total civilian labor force unemployed, percent black, percent urban, population per square mile, and median household income. Median school years completed, median household income, percent black, and population per square mile were found to be significant in predicting to the proportion of students identified as mentally retarded. Indicator variables related to the proportion of students identified as emotionally disturbed included average daily memh·.~rship, percent urban, median income, and percent of total civilian labor force unemployed. A low correlation was found between proportion of speech impaired students and percentage of families in urban residence, true value of property, and average daily membership. The total proportion of students identified as handicapped was best predicted by the percent of total civilian labor force ·unemployed, percent urban and population per square mile. A special education Expectancy Index was developed to provide a comparative measure for each school division on each criteria measure. The atypical school divisions were diagnosed as having overidentified or underidentified proportions of each criteria measure. The study confirmed the methodology utilized by Chalfant and yielded an operational model for predicting certain special education services. / Ed. D.