Measuring Corruption: State of the Art, Challenges, and Advancements

The study of corruption has benefitted significantly from the expansion of survey and experimental research and the availability of national and cross-national datasets, assessing perceptions, attitudes and experiences of corruption. Traditionally, researchers have used perception-based indexes to measure corruption, such as the Corruption Perceptions Index or the Corruption Control indicator of the World Governance Indicators (WGI) (Kaufmann, Kraay, & Mastruzzi, 2009). These indexes aggregate citizens, public servants, business people, and experts’ opinions about the prevalence of corruption in a given organisation, sector or society and have served as proxies to its objective measurement (Gingerich, 2013). Objective indicators, such as official crime statistics, pose various comparability challenges and have often been discarded as a reliable measurement of corruption, leading to academic divestment in the subfield (Cazzola, 1988; De Sousa, 2002; June, Chowdhury, Heller, & Werve, 2008; Piquero & Albanese, 2011) However, perception-based measures raise a number of challenges and criticisms on substantive and methodological grounds (Andersson and Heywood, 2009; Razafindrakoto and Roubaud, 2010; Thomas, 2010; Thompson and Shah, 2005), for often overlooking conceptual gradients of the phenomenon (Gerring and Thacker, 2004; Hellman et al., 2003; Karklins, 2005; Schleiter and Voznaya, 2014) and cultural differences stemming from cynicism, social injustice, economic inequality, social trust, government acceptance, and media reporting (Treisman, 2007). Current discussions in this regard have been held simultaneously (but not always jointly) in academia, specialized agencies, and (non-)governmental organizations, suggesting the need for the interplay of different disciplinary and methodological approaches. There is still much to be done when it comes to build more adjusted measures of corruption and, especially, to determine what dimensions of the phenomena are captured by different indicators (subjective and objective) and the extent to which they travel across contexts for comparative purposes. Besides discussing what has been implemented with greater or lesser success, this workshop also aims at addressing: (a) conceptual constraints/opportunities, since the issues of how to ask, code, and even define corruption over time and across groups and cultures remain underexplored; (b) data collection constraints (opportunities), since comparable crime-related data on corruption is still a challenge to cross-national research but other government data is being produced and release dfor public use, offering new possibilities at developing objective corruption proxies (Fazekas et al., 2016; Fazekas and Kocsis, 2020; Lima and Delen, 2020); (c) design constraints, given that “corruption” is a floating signifier with conceptual nuances not easily captured by single measures and a topic subject to “social desirability” bias; (c) scope constraints, as measurements have often been developed with distinct policy, societal, and/or academic objectives in mind and addressed to specific audiences; (d) and ethical constraints, raises by applied research on criminal behaviours. The workshop will serve to debate such challenges and possible solutions, through a combination of theoretical, methodological, and/or empirical contributions that may help shaping strategies to improve the depth and quality of studies on (anti-)corruption and integrity. This workshop aims to bring together researchers dealing with the challenges of uncovering and measuring corruption using both perception-based and hard data.