Audits and logistic regression, deciding what really matters in service processes. A case study of a government funding agency for research grants

Governmental agencies, businesses and nongovernmental organizations experience repetitive and out-of-date processes causing resource waste. Banks, credit card companies and tax authorities must evaluate whether creditors have fulfilled their obligations, and funding agencies should verify the use of funds by grant recipients. Here logistic regression estimates the probability of conformity of grants to the financial obligations of the researcher analyzing the correlation between characteristics of the grant and the grant's final status as approved or not. The logistic equation uncovers relevant characteristics through an analysis of false positives. A ROC curve is constructed which reveals an optimal cutoff for nonconformity.