« Back to Results

Research Ethics in Development Economics

Paper Session

Friday, Jan. 7, 2022 3:45 PM - 5:45 PM (EST)

Hosted By: International Network for Economic Method
  • Chair: Seye Abimbola, University of Sydney

A Call for Structured Ethics Appendices in Social Science Papers

Edward Asiedu
,
University of Ghana
Dean Karlan
,
Northwestern University
Monica Lambon-Quayefio
,
University of Ghana
Christopher Udry
,
Northwestern University

Abstract

Ethics in social science experimentation and data collection are often discussed but rarely articulated in writing as part of research outputs. Although papers typically reference human subjects research approvals from relevant institutional review boards, most recognize that such boards do not carry out comprehensive ethical assessments. We propose a structured ethics appendix to provide details on the following: policy equipoise, role of the researcher, potential harms to participants and nonparticipants, conflicts of interest, intellectual freedom, feedback to participants, and foreseeable misuse of research results. We discuss each of these, and some of the norms and challenging situations of each. We believe that discussing such issues explicitly in appendices of papers, even if briefly, will serve two purposes: more complete communication of ethics can improve discussions of papers and can clarify and improve the norms themselves.

How Informed Is Consent? A Field Experiment

Alexandra Avdeenko
,
Center for Evaluation and Development
Matthias Stelter
,
Center for Evaluation and Development

Abstract

With the claim to improve the lives of the poor and vulnerable, economists increasingly apply field experiments to advance economic research and inform policy. We investigate informational constraints to providing consent to participate in research. We show that survey participants in rural Pakistan are insufficiently informed about important aspects of their consent. Only about every fifth person understands the purpose of the data collection. We experimentally test an interactive, audio-visually supported approach to obtain consent within a survey data collection with 3,964 participants. We find that the alternative presentation improves the understanding of the voluntary nature of participation. Beyond ethical aspects, the study addresses methodological implications. We investigate changes in response behavior with potential implications for the external validity of survey based research and thereupon based policies.

Ethics in Development Research: ‘Doing No Harm’ When Conducting Research in the Global South

Ana Garcia-Hernandez
,
Rosario University
Lennart Kaplan
,
University of Gottingen
Jana Kuhnt
,
German Development Institute
Janina Steinert
,
Technical University of Munich

Abstract

In various disciplines such as Economics, Political Science, Public Health, and Sociology, primary data collection in the Global South has expanded rapidly in the past few years. However, while the academic community has acknowledged the ethical challenges of running experiments in the field, the discussion has only recently broadened to considering the safety, wellbeing, and working conditions of research staff. Notably, given ambitious project timelines and budget constraints, rapid data collection and research deliverables may sometimes be prioritized over working conditions and general staff wellbeing. Furthermore, the dynamics that arise from local and international team compositions are of utmost importance in acknowledging contributions, cultural sensitivity, and dissemination of results. Thus, addressing these multifaceted ethical challenges has become more critical than ever and requires clear-cut evidence to guide global action and collaboration. Based on qualitative interviews with research staff at various hierarchical levels, we identify several dimensions of ethical challenges, where emotional wellbeing, security threats, a lack of context sensitivity and unsteady contracts seem pervasive. In order to quantify their prevalence and to identify risk factors of experiencing different ethical challenges, we conduct a global online survey. Results aim to raise awareness of existing challenges and lay the groundwork for conducting more ethical research.

No (Value-)Free Lunch: How Moral Values Matter for Methodology

Donal Khosrowi
,
Leibniz University Hannover

Abstract

In recent decades, prevalent methodology in development, policy evaluation, and microeconometrics more generally has put increasing emphasis on using RFEs/RCTs and sophisticated quasi-experimental strategies. In this paper, I highlight two important ways in which this methodological development raises concerns about the role of moral and political values in economic methodology.
First, methods such as RFEs/RCTs are preferred because they best promote various methodological values, including the unbiasedness and precision of average effect estimates. On the same grounds, the credibility of ancillary methods for elucidating the distribution of effects, such as subgroup analyses, is often discounted. However, average effect estimates alone are not informative for decision-makers that pursue distributive goals, e.g. deploying interventions that prioritize the worst-off in a population. This creates tensions between methodological values, such as unbiasedness and precision, and the moral and political values that decision-makers are able to pursue on grounds of treatment-effect evidence. In the most extreme case,
methodological tenets can bias what kinds of interventions are implemented, e.g. those that are effective on average regardless of their distributive effects.
A second concern arises in the context of familiar tensions between internal and external validity: methods that do well on the former often do poorly on the latter. When methods that promise putatively higher internal validity are prioritized, one implicitly accepts higher risks when making decisions about implementing interventions in novel environments, where they might interact with local features in unforeseen ways. Once again, the choice of some methods over others is not value-neutral, but involves implicit tradeoffs among competing values, including moral and political ones.
Against the background of these concerns, I argue that methodological debates about the relative merits of different methods must consider not only familiar methodological virtues, but also how different methods enable the pursuit of various moral and political ends.

Public Policy Experiments Amid Scarcity: When Is Randomization Permissible?

Douglas MacKay
,
University of North Carolina-Chapel Hill
Emma Cohn
,
University of North Carolina-Chapel Hill

Abstract

Social scientists are increasingly considering the ethics of using randomized controlled trials (RCTs) to evaluate public policy interventions. Random assignment in particular raises questions of fairness for by randomizing participants to either the intervention or control arm, people are subject to different policies and so, often, to different types and levels of benefits. Random assignment is widely understood to be permissible when there is genuine uncertainty regarding which trial arm is superior. However, some argue that random assignment can also be permissible when an intervention is reasonably expected to be superior to other trial arms. Provided that the superior intervention is scarce, there remains some uncertainty regarding the effects of this intervention, and random assignment is a fair way to allocate access to it, proponents argue, randomization is permissible. For example, on this argument, it is permissible to randomly assign people to a generous guaranteed income intervention if the intervention is too expensive to roll-out to all eligible people, there is uncertainty regarding the health and employment effects of the intervention, and a lottery is a fair way to allocate access to it. In this paper, we investigate this line of argument, exploring the ethical literature on fair allocation and the use of lotteries. Our aim is to formulate a set of guidelines government agencies and researchers may use to determine if a lottery is a fair way to allocate a given benefit, or whether the benefit should be allocated according to an alternative principle – e.g. priority to the worse off. In cases where a lottery is not a fair way to allocate access to a superior intervention, we explore whether it is possible to modify a study’s design in an epistemically sound way to ensure fairness, for example, through the use of weighted lotteries or “stepped-wedge” designs.

Discussant(s)
Heather Lanthorn
,
IDinsight
Priya Mukherjee
,
University of Wisconsin-Madison
Malvina Ongaro
,
University of Eastern Piedmont
Marcos Picchio
,
University of Wisconsin-Madison
Janina Steinert
,
Technical University of Munich
JEL Classifications
  • O1 - Economic Development
  • B4 - Economic Methodology