Rule bending on the frontlines of public service delivery

How to cite

Hansen, Paw. 2020. "Rule bending on the frontlines of public service delivery." AEA RCT Registry. AEARCTR-0005835. https://doi.org/10.1257/rct.5835-1.1

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

Paw Hansen*

* Primary investigator

Registry ID
AEARCTR-0005835
DOI
https://doi.org/10.1257/rct.5835-1.1
First registered
Last updated
Status
Completed
Trial dates
2020-06-01 – 2020-06-30
Countries
Denmark (Capital Region)
Registry record
http://www.socialscienceregistry.org/trials/5835

Abstract

What role does front-line workers' motivation play in known heuristics affecting their decision making with clients? Will different types of motivation, for example, reduce or increase biases and discrimination?

Previous research suggests that when front-line workers face deservingness relevant cues in interaction with clients a psychological `deservingness heuristic' is triggered causing front-line workers to think about their decisions in terms of who deserves help - disregarding recognized legal or political standards.

The study examines this claim by drawing on data from a conjoint experiment embedded in a large-scale survey among Danish social workers.

Intervention

The intervention consists of manipulation of information cues regarding client characteristics in an online survey among Danish social workers.

Experimental design

The study involves an individually randomized trial - building on survey data. The experiment is carried out among a nonprobability
sample of Danish social workers.

All survey respondents are first presented with an introductory text identical across experimental conditions. Then, respondents are presented with descriptions of three fictive clients, one by one, and asked if they would sanction in each of the given cases. The descriptions of clients is randomized using a conjoint approach drawing randomly from a pool of information cues (client characteristics).

Design details

Randomization method
Randomization is carried out by simple randomization by computer.
Randomization unit
Individual survey respondent.
Clusters
300 individual respondents.
Observations
300 individual respondents each making decisions about 3 fictive clients = 900 observations in total.
Arms
Approx. 450 individual respondents in each of the two treatment arms.
Primary outcomes
Social workers willingness to sanction clients on a 10-point scale.

Cited by

No citation data available for this registration yet.

Keywords