The IRS 990 Return is becoming an increasingly prominent source of financial data underlying descriptions of the nonprofit sector and studies of nonprofit organizations. However, questions about the quality of the data continue to be of concern. This study of 350 nonprofit organizations investigates the adequacy, reliability, and appropriate interpretation of IRS 990 Return data through comparisons of selected entries with corresponding measures from each organization's audited financial statements. Both quantitative and qualitative methods are used to examine and explain the consistency between the two data sources. The study concludes that the IRS 990 Return can be considered an adequate and reliable source of financial information for many types of investigations, but preparers and users of the data need a clearer understanding of its purposes to enable appropriate interpretations. The study of nonprofit organizations has historically been a daunting task. The nonprofit sector is noted for its tremendous diversity in size, purpose, and formality of organizations. The research challenge has been compounded by a lack of institutionalized mechanisms for sorting nonprofits into homogeneous Note: We wish to thank the Aspen Institute's Nonprofit Sector Research Fund for support of this study, Joe Galaskiewicz for his skepticism about financial measures in earlier research efforts, Curt Doetkott for statistical expertise, Mike Palmer for data management expertise, Mariya Chuykova for meticulous attention to detail throughout all phases of data collection and analysis, and five anonymous reviewers whose suggestions contributed substantially to this effort. Preliminary results were presented at the 1996 annual Association for Research on Nonprofit Organizations and Voluntary Action (ARNOVA) conference; an earlier version of this article was presented at the 1997 Academy of Management national meeting.

This exploratory study examines the reliability and validity of Internal Revenue Service (IRS) 990 data, an increasingly prevalent source of financial measures in the research of nonprofit organizations today. Correlations between selected financial measures taken from the IRS returns of 270 nonprofits and the identical measures gathered from the same organizations via mail questionnaire are examined, followed by examination of financial statements and interviews with participants in the accounting process from a subsample of the organizations. High correlations show the IRS 990 Returns to be a generally reliable source of financial data, although this varies by entry, organizational size, and NTEE category. The validity of this data, however, is questionable. The growing prevalence of expense shifting combined with a lack of rules, guidelines, and monitoring of internal expense allocations allows for increasing distortions in the financial picture portrayed by the IRS 990 Return.

No abstract