Skip to main content

Advertisement

Log in

Measuring Corruption: A Critical Analysis of the Existing Datasets and Their Suitability for Diachronic Transnational Research

  • Original Research
  • Published:
Social Indicators Research Aims and scope Submit manuscript

Abstract

Any researcher on corruption has faced at some point the dataset dilemma. How can one assess the incidence of a phenomenon on corruption levels if we cannot determine how much corruption is there in the first place? The problem compounds when the research has a transnational or comparative element. How can one assess how different corruption levels are in different jurisdictions if we cannot be sure if the measurements are comparable? It becomes critical when the research has a diachronic component, and tries to incorporate changes over time, as the stability and consistency of datasets become essential. This article reviews the literature on the topic from the last fifteen years and evaluates all the main options available today for researchers and policy designers in terms of validity and reliability, explaining first the particularities of these two concepts in the context of measurements of the prevalence of corruption. It pays particular attention to the limitations of the different datasets and determines the validity and reliability of the oft-used Corruption Perceptions Index post-2012 (CPI) and the Control of Corruption (CoC) indicator for the whole data series. This conclusion partially vindicates those researchers and policy designers who have used these datasets in the past, especially when compared to all the other options, while firmly warning users about the kind of conclusions that can be extracted from them.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

  1. Namely the International Country Risk Guide, created in 1980 and still widely used (PRS Group 2018), discussed in Sect. 3.2.3.

  2. To cite two exceptions which systematically identify the problem see: Lancaster and Montinola (1997; 2001) and Hawken and Munck (2009b; 2011).

  3. He was talking about life.

  4. The actual name of the project was Anticorruption policies revisited Global Trends and European Responses to the Challenge of Corruption.

  5. To their credit, in order to address this, they produced another two indicators. One mapping the corruption in Europe and the quality of government, aptly named “European Quality of Government Index” (Charron, Dijkstra, and Lapuente 2014). The other, analysed below in 3.1.3, the Index of Public Integrity (Mungiu-Pippidi and Dadašov 2016). As we will see, even if we accept that these indices are methodologically sounder—a big “if”—their application is limited.

  6. This is a good example of the “science about measuring corruption” presented above, and one that does not help us much, as it requires a proof that cannot be easily obtained. The reason for using a proxy is precisely that the subjacent variable cannot be directly measured. Hence, beyond testing the correlation between surveys of perceptions and surveys of experiences of corruption, this approach will not take us very far.

  7. This term is commonly used in statistics to signify characteristics that can be placed in well-defined groups or categories that do not depend on order. Other statisticians call it “qualitative”.

  8. Their terminology to encompass practices “where processes are non-routine, fuzzy, innovative and conflictual. Means–ends relations, outputs and outcomes are hard to classify, and insights that are brought forward are resented as interests are opposed” (Noordegraaf and Abma 2003, 867). Traffic congestions and unemployment could also be considered non-canonical.

  9. The production phase in the terminology of Merry, Davis, and Kingsbury (2015, p. 10).

  10. The “conceptualisation phase” in the terminology of Merry, Davis, and Kingsbury (2015, 10).

  11. See for example Bardham’s recent discussion of the “Paradox of Singapore”: “Singapore comes out as one of the cleanest. But in The Economist magazine ranking of 22 countries of the world in terms of billionaire wealth from crony rent-thick sectors, Singapore is no. 4 in the crony-capitalism index, only after Russia, Malaysia and the Philippines, and worse than even Ukraine or Mexico” (Bardhan 2018, pp. 115, 126).

  12. For example, over a 5 or 10 year period, taking into account sliding averages of yearly indicators.

  13. Generally speaking the first generation englobes indicators based on experts’ perception and the second indicators based on surveys. Other attempts would be part of a “third generation” (the categories can be traced back to the book of Graycar and Smith 2011 on global research and practice on corruption; but they are still used by leading authors, see for example Mungiu-Pippidi 2015b, 27; 2016, 366). However, as explained below, some indicators based on structural elements predate certain surveys. Methodology overhauls of expert-based indicators are fairly recent in time and posterior to global surveys and structural indicators.

  14. Data mining in corruption has the potential to offer a completely different approach to the issue. Some work has already been done in recent years (Fitzpatrick 2014) although the first studies are already more than ten years old (Huysmans, Baesens, and Vanthienen 2008).

  15. See for example: “According to Transparency International's latest survey on global corruption, an impressive one in four people paid a bribe in to politicians or political appointees in the past year” in the Forbes magazine article “Transparency International Spells It Out: Politicians Are The Most Corrupt” (Rapoza 2013).

  16. See for example the use of its data in the “Informe del Grupo Asesor de Expertos en anticorrupción, transparencia e integridad para América Latina y el Caribe” commissioned by the Inter-American Development Bank (Engel and et al. 2018, p. 4).

  17. Although it is possible to do the same in real time (see for example the review of the literature of real-time data in the fiscal policy context in Cimadomo 2016).

  18. Already in 1976 a study at MIT on foreign investment was considering elements such as “business environment” in its analysis ran with SPSS software (Kobrin 1976).

  19. A report written for TI on the issue of curbing corruption in public procurement openly states that “few government activities create greater temptations or offer more opportunities for corruption” (Susanne and Sherman 2014, 3).

  20. A 2008 study commissioned by the Office of Development Studies United Nations Development Programme listed 178 composite indices, including in that list all regional or global indices elaborated by organizations and academics, “based on several indicators or sub-indices. These indicators and sub-indices are aggregated following some methodology to give an overall score for the country. The country scores are used to either create a ranking to show progress (or setbacks) or to simply present the data—without necessarily ranking the countries” (Bandura 2008, 6).

  21. Though, no discussion is offered about how changes over time in the indicators measured by the components are reflected or reflect—as the causality could be in reverse—actual changes in the prevalence of corruption in a country over time.

  22. He continued: “The international shaming that ensued, encouraged a race to the top, i.e. to lower levels of corruption. The race was on the international stage for some (e.g. a senior advisor to South Korea’s prime minister sharing his stated goal for Korea to be among the top-15 countries within 5 years); frequently the race was regional (e.g. between Hong Kong and Singapore; between Kenya and Uganda; Hungary and the Czech Republic, etc.); and for some selected countries at the bottom of the league table (e.g. Bangladesh, Nigeria and Paraguay) it has spurred a determination to shed the label of being ‘one of the world’s most corrupt countries’ (Galtung 2006, 101).

  23. The press release of the 2009 edition quoted the Permanent Secretary for Governance & Ethics of the Office of the President of Kenya saying: “Challenging the naysayers, the Worldwide Governance Indicators show that governance and corruption can be robustly measured and the lessons drawn can in fact be put to subsequent use by reformist governments, the development community, civil society and the media” (World Bank 2009, 2).

  24. “In the form of excessive patronage, nepotism, job reservations, ‘favor-for-favors’, secret party funding, and suspiciously close ties between politics and business” (PRS Group 2018, 5).

  25. In very simple terms, the experiment proved that the average of the error of each guess was higher that the error of the average of the guesses, proving that in large estimates collective wisdom is better that the aggregation of individual ones.

References

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to José-Miguel Bello y Villarino.

Ethics declarations

Conflict of interest

The author declares that he has no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The views presented in this article are exclusively those of the author and should not be in any case construed to represent the position of any of the organisations in which he has served.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bello y Villarino, JM. Measuring Corruption: A Critical Analysis of the Existing Datasets and Their Suitability for Diachronic Transnational Research. Soc Indic Res 157, 709–747 (2021). https://doi.org/10.1007/s11205-021-02657-z

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11205-021-02657-z

Keywords

Navigation