The formats of epidemiological risk information disclosure and citizens’ voluntary self-reporting for close contact: report accuracy, time consumption, and privacy concerns

How to cite

Liu, Ning; Wu, Qiying. 2022. "The formats of epidemiological risk information disclosure and citizens’ voluntary self-reporting for close contact: report accuracy, time consumption, and privacy concerns." AEA RCT Registry. AEARCTR-0010036. https://doi.org/10.1257/rct.10036-2.0

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

Ning Liu* , Qiying Wu

* Primary investigator

Registry ID
AEARCTR-0010036
DOI
https://doi.org/10.1257/rct.10036-2.0
First registered
Last updated
Status
Completed
Trial dates
2022-08-08 – 2022-09-20
Countries
China (Beijing)
JEL codes
C93, D83, D91
Registry record
http://www.socialscienceregistry.org/trials/10036

Abstract

Timely contact tracing is important to prevent the occurrence and control the spread of an infectious disease. To complement the active case investigation and contact tracing, one often adopted practice is to publish lists of times and locations of risk and ask citizens to engage in self-reporting in case they find that they have visited the listed places in the corresponding time windows. This study tries to provide the first experimental study on how different formats to display the times and locations of risk affect the accuracy of self-reporting. Our task for the subjects is to read a list of times and locations of risk within a three-day window and compare the list with their own route of visits to decide whether to report being a close-contact or not. We construct the hypothetical lists and routes of visits of the subjects using published lists in the past, varying the length of the list, the routes of visits, and whether there is an overlap between the two for across tasks. The baseline treatment of display format is to sort places and times of risk by individuals with positive nucleic acid test. We compare subjects’ self-report accuracy with two alternative formats: sorting the information by locations, and by time, eliminating link between individuals and the places of risk. We hypothesize that the accuracy in the two alternative formats, in which information is organized in an easier way for search, is higher than the baseline treatment. We will also measure individuals’ privacy concerns regarding the list of risk information and expect the alternative formats to lead to lower concerns of privacy. Another dependent variable to consider is the time spent for the task.

Intervention

We have three between-subject treatment groups, receiving lists of times and locations of risk in three formats, where the information is clustered 1) by the individuals who have positive nucleic acid test results and travelled to these places, 2) by the dates of the time of risk, or 3) by the locations of risk.

Experimental design

Our task for the subjects is to read a list of times and locations of risk within a three-day window and compare the list with their own route of visits to decide whether to report being a close-contact or not. We construct the hypothetical lists and routes of visits of the subjects using published lists in the past, varying the length of the list, the routes of visits, and whether there is an overlap between the two for across tasks. The baseline treatment of display format is to sort places and times of risk by individuals with positive nucleic acid test. We compare subjects’ self-report accuracy with two alternative formats: sorting the information by locations, and by time, eliminating link between individuals and the places of risk.

Design details

Randomization method
Randomization will be done through dingwei.netease survey platform inbuilt randomization feature.
Randomization unit
Individual
Clusters
N/A
Observations
1050 individuals, each answering 6 questions, in which the subjects will compare the given (hypothetical) three-day route track with a list of times and locations of epidemiological risk.
Arms
about 350 individuals in each treatment
Primary outcomes
Accuracy of the self-report, response time, and perceived privacy concerns.
Secondary outcomes
Opinion regarding the information provider

Materials

Cited by

No citation data available for this registration yet.

Keywords