Knowledge versus Skills: Evidence from a Field Study on Health Information in Sierra Leone

How to cite

Horn, Samantha; Karing, Anne; Glennerster, Rachel; Conteh, Fatu. 2024. "Knowledge versus Skills: Evidence from a Field Study on Health Information in Sierra Leone." AEA RCT Registry. AEARCTR-0013488. https://doi.org/10.1257/rct.13488-1.1

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

Samantha Horn* , Anne Karing , Rachel Glennerster , Fatu Conteh

* Primary investigator

Registry ID
AEARCTR-0013488
DOI
https://doi.org/10.1257/rct.13488-1.1
First registered
Last updated
Status
Completed
Trial dates
2024-05-22 – 2025-10-31
Countries
Sierra Leone (Western Area Urban)
Registry record
http://www.socialscienceregistry.org/trials/13488

Abstract

Evaluating the truthfulness of new information is important for knowledge building and often influences the decision to share information. This study compares two approaches to improving judgments of whether information is reliable. The first, grounded in existing literature, teaches individuals to identify specific signals that indicate truthfulness. The second method builds domain-specific knowledge—in this case, about vaccines—to help individuals assess information based on its content. To compare these two approaches, we randomize social media users in Freetown, Sierra Leone, to receive a short interactive training program focusing on i) hand-washing (a placebo health topic for (Control), ii) how to spot typical features of misinformation (Misinformation Skills}), or iii) how vaccines work (Vaccine Knowledge). We examine outcomes immediately after training to assess treatment effects on information discernment, sharing, and vaccine knowledge. Following the training session, we monitor on-platform (WhatsApp) behaviors for one month and evaluate the persistence of immediate treatment effects, as well as treatment effects on vaccine uptake and healthcare utilization one month post-training.

Intervention

Participants will be randomized at the individual level to receive one of three training interventions: a Control training on hand-washing and sanitation, a Misinformation Skills training that teaches how to identify typical features of misinformation, or a Vaccine Knowledge training that explains how vaccines work to build immunity. Each training lasts approximately 30 minutes and consists of a short video and a reinforcement section between two video viewings to clarify the concepts presented.

Experimental design

Please see the accompanying pre-analysis plan.

Design details

Randomization method
Participants will be randomized within SurveyCTO.
Randomization unit
Participants will be randomized at the individual level.
Clusters
1800 individuals
Observations
1800 individuals
Arms
Equal but not blocked randomization across arms so ~600 individuals per arm (Control, Misinformation Skills, and Vaccine Knowledge).
Primary outcomes
Primary outcomes are information discernment for constructed and real headlines, sharing discernment, and learning from headlines (all measured at midline immediately after the training intervention), and on-platform (WhatsApp) behaviors. These outcomes are described in more detail in the pre analysis plan.
Secondary outcomes
Secondary outcomes are information discernment for constructed and real headlines (measured at endline), vaccine beliefs (measured at both midline and endline), healthcare utilization, and Covid-19 vaccine uptake.

Cited by

No citation data available for this registration yet.

Keywords