Norms of corruption among future civil servants

How to cite

Stegmann, Andreas; Marx, Ben; Bazzi, Sam; Shaukat, Mahvish; Hilmy, Masyhur. 2024. "Norms of corruption among future civil servants." AEA RCT Registry. AEARCTR-0013239. https://doi.org/10.1257/rct.13239-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Andreas Stegmann* , Ben Marx , Sam Bazzi , Mahvish Shaukat , Masyhur Hilmy

* Primary investigator

Registry ID
AEARCTR-0013239
DOI
https://doi.org/10.1257/rct.13239-1.1
First registered
Last updated
Status
In development
Trial dates
2024-03-28 – 2025-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/13239

Abstract

The literature has generally documented that perceptions (i.e. beliefs about others) are more slow-moving than individual attitudes, often giving rise to misperceived social norms, i.e. discrepancies between individuals' attitudes and the attitudes they believe are held by others. We study this phenomenon in the context of beliefs about attitudes towards corruption and whistle-blowing among future civil servants in Indonesia. We will conduct a survey experiment in a population of future civil servants to learn about the role of misperceived norms of corruption in shaping the decision to report on corrupt behavior, and more generally, to engage in anti-corruption activities.

Intervention

Information treatment about norms of corruption held by alumni of a civil service academy

Experimental design

We conduct an information provision experiment with students enrolled at a civil service college. After eliciting students' private and second-order beliefs regarding the extent to which engaging in corrupt behavior is perceived to be shameful/embarrassing (memalukan in Bahasa), students are randomly allocated to one of two conditions: 1) no information (control group); 2) information about the true beliefs held in a group of alumni and active civil servants with respect to the extent to which engaging in corrupt behavior is perceived to be either shameful. We cross-randomize whether students receive information related to the relevant whistleblower protection legislation for civil servants. Finally, we further cross-randomize whether we introduce a garbling methodology when eliciting students' observations of corrupt behavior in the past.

Design details

Randomization method
Randomisation done in office by a computer
Randomization unit
We will randomize treatment status at the individual level, stratifying treatment by the faculty in which students are enrolled and by students' region of origin.
Clusters
Randomization is at the individual level.
Observations
We will invite the entire cohort of final-year students of the civil service college to participate in an online survey during which we will administer the above survey experiment. The cohort consists of approximately 1100 students. We expect a 75 percent response rate for a total number of observations of (approximately) 825.
Arms
We will randomly assign one-half of our sample to each of the main treatment conditions described above. We thus expect approximately 410 students per experimental condition.
Primary outcomes
Indicators for hypothetical whistleblowing (in response to a short vignette) and reporting of corruption observed during the past 3 years/students' studies. Intentions to complete an e-learning course designed by the national anti-corruption agency. Indicators for registration to and completion of the e-learning course designed by the national anti-corruption agency. We will also construct an index of whistleblowing behavior.

Cited by

No citation data available for this registration yet.

Keywords