ESTIMATING PARENTS' CONTRIBUTION TO COLLEGE COSTS: THE ACCURACY OF THREE MEASURES OF SUCCEEDING YEAR FAMILY NET INCOME

Abstract

ABSTRACTThe accuracy of each of three measures of future income was investigated for a sample of families who filed Parents' Confidential Statements with the College Scholarship Service in two consecutive years. The three income measures investigated were: (1) parents' reports of current (1968) income, (2) parents' estimates of 1969 income, and (3) predictions of future year (1969) income made by application of least squares regression parameters to parents' estimate of income. Two types of accuracy were considered. The first was the accuracy of prediction assessed by the correlation between the predictors (current or estimated income) and the criterion (actual 1969 income). This analysis indicated that the best single predictor of actual income was estimated income, with a validity coefficient of .86, and was followed closely by current income, with a validity coefficient of .83. The accuracy of each income measure was also investigated by assessing the median discrepancies between parents' contribution associated with each income measure and the contribution associated with actual 1969 income. Current and estimated income result with about equal frequency in substantial (>$100) undercalculations of parents' contribution; but use of current rather than estimated income would result in substantial (>$100) overcalculations for approximately 31,000 additional CSS families. In addition, needs analysis procedures that rely on either current or estimated income place low income families at a disadvantage by overstating their future year income. This study suggests least squares regression as one possible approach to remove this inequity.