Rayzberg, Margarita S. 2019. “Fairness in the Field: The Ethics of Resource Allocation in Randomized Controlled Field Experiments.” Science, Technology, & Human Values 44 (3): 371–98.





Creative Commons Licence




Contributed date

July 19, 2019 - 6:54pm

Critical Commentary

AO: This Science, Technology, & Human Values article by Margarita Rayzberg (2019) mulls over the line between practice and research in economic and social research and demonstrates that methods like randomized controlled trials are deeply entangled, supporting previous STS work showing that the distinction between practice and research is historically, culturally, institutionally and practically contingent.


"Many in the international development community have embraced the randomized controlled field experiment, akin to a biomedical clinical trial for social interventions, as the new “gold evidential standard” in program impact evaluation. In response, critics have called upon the method’s advocates to consider the moral dimensions of randomization, leading to a debate about the method’s ethics. My research intervenes in this debate by empirically investigating how researchers manage the perception of ran- domization in the field. Without the possibility of a placebo, researchers rhetorically and materially frame the experiment differently for the control and treatment groups. Three technologies allow for this differential framing: geographic separation, temporal delay, and public randomization cere- monies. Geographic separation is a “technology of opacity” designed to obscure unequal resource distribution by disentangling the intervention and research components of the experiment for the control group. The latter two are technologies of transparency designed to expose the element of randomization but downplay conditions that may affect participant buy in. All three technologies work to preclude collective definitions of fair resource allocation, yet they are not fully successful in preventing modes of confrontation and resistance that lie outside of the experiment’s framing."