The consequences of different strategies for measuring tax evasion behavior

Abstract

A critical review of existing measures of tax evasion suggests the need for innovative measurement strategies. Toward this end, a new method for documenting cases of individual tax evasion without compromising confidentiality is described. Results of an investigation in The Netherlands indicated negligible correspondence between scores on this measure and self-reports of tax evasion behavior. This outcome is particularly disturbing since the self-reports were obtained under conditions that should have maximized their veridicality. Additional analyses indicated that different sets of attitudes, norms and personality variables are correlated with admitted tax evasion and with patterns of actual behavior. The implications for future research are discussed.