Common Source Bias, Key Informants, and Survey-Administrative Linked Data for Nonprofit Management Research. Issue 1 (2nd January 2020)

• Record Type: Journal Article
• Authors: Kim, Mirae; Daniel, Jamie Levine
• Is Part Of: Public performance & management review. Volume 43:Issue 1(2020)
• Page Start: 232
• Page End: 256
• Publication Date: 2020-01-02
• Subjects: administrative records -- common source bias -- key informants -- nonprofits -- IRS 990 -- survey; Civil service -- Labor productivity -- Periodicals; Civil service -- Labor productivity -- United States -- Periodicals; Government productivity -- Periodicals; Government productivity -- United States -- Periodicals; Public administration -- Periodicals; Public administration -- United States -- Periodicals; Civil service -- Labor productivity; Government productivity; Public administration; United States; Periodicals; 351.05
• DOI: 10.1080/15309576.2019.1657915
• Languages: English
• ISSNs: 1530-9576

Abstract

Abstract: In this study, we examine the increasingly popular strategy of supplementing survey data with administrative/archival data as a strategy to overcome common source bias. As the strategy does not present a cure-all solution, we argue it is a questionable recommendation and provide advice on how to approach this issue with due consideration. Focusing on the nonprofit sector, we compare the data drawn from a survey of financial estimates completed by nonprofit leaders and the IRS 990 forms-based financial data of corresponding nonprofit organizations. After presenting the evidence of some differences between the two data sources, we identify a set of key informant characteristics that are correlated with the level of differences. We echo George and Pandey (2018)'s claim that the criticisms toward common source bias have been somewhat exaggerated and urge public and nonprofit management researchers to be aware of other pitfalls they might face when trying to address common source bias.