Abstract
Several econometric issues arise in using nonprofit data. After controlling for unduly influential observations and heteroskedasticity, regression analysis performed on National Center for Charitable Statistics “digitized data” from 2001 to 2003 found mixed evidence of economically significant associations between donations and other revenue streams; regressions without these controls support different conclusions. Testing does indicate that government support sends greater quality signals than program support or investment income.
Get full access to this article
View all access options for this article.
