Memory and Probability: The Formation of Beliefs About Livestock Loss from Droughts among East African Pastoralists

How to cite

Morsink, Karlijn; Phan, Vinh. 2025. "Memory and Probability: The Formation of Beliefs About Livestock Loss from Droughts among East African Pastoralists." AEA RCT Registry. AEARCTR-0016202. https://doi.org/10.1257/rct.16202-1.0

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

Karlijn Morsink* , Vinh Phan

* Primary investigator

Registry ID
AEARCTR-0016202
DOI
https://doi.org/10.1257/rct.16202-1.0
First registered
Last updated
Status
On going
Trial dates
2023-12-01 – 2027-03-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/16202

Abstract

Pastoralists in sub-Saharan Africa are frequently exposed to droughts that pose substantial risk to their primary livelihood: livestock. The way they form beliefs about this risk is critical for their choice of risk-mitigation strategies, and deviations from rational belief updating have been observed in this context. This study investigates how East African pastoralists form beliefs about the likelihood of cattle loss due to droughts, focusing on the role of memory and experience in shaping beliefs about risk. Using a sample of 4,689 households across 391 communities in Ethiopia and Kenya, we explore how past experiences, memory, and similarity of past events influence belief formation. We employ both observational data and a randomized survey experiment to examine the impact of actual and primed experiences with varying degrees of similarity to drought-induced cattle loss.

Experimental design

We implement a survey experiment among 4,689 respondents across 391 communities in Ethiopia and Kenya to study whether recalled experiences influence pastoralists’ beliefs. Within each community, respondents are randomly assigned to a control group or one of the three treatment groups.
In the control group, respondents are not asked to recall any past memory and are simply asked about their beliefs regarding the likelihood of cattle loss due to droughts. In each recall treatment group, we first ask the respondents if they have ever experienced a pre-specified event in the past. If the respondent reports to not have lived the event, they receive no further recall questions. If the respondent reports to have lived the event, we further ask three open-ended questions to help direct respondents’ memories to this past event. In treatment group one, respondents are primed to recall the last time they lost cattle due to droughts. In treatment group two, we prime respondents to recall their experience with losing cattle due to raids. In the last treatment group, we prime respondents to recall their experience with losing goats due to droughts.
For all three recall treatments, we also ask respondents to describe anything else that comes to their mind when they think about the recalled experience. Finally, respondents are asked to give a rating on a scale of 0 to 10, where 0 indicates that they do not remember the event vividly at all, and 10 indicates that they remember the event very vividly.
Later in the survey, we present respondents with a series of positive and negative life events, including the priming events in treatment group two and three. We ask whether the respondents have experienced each of the event and to rate, on a scale of 0 to 10, how similar they think these events are to the anticipated event of losing cattle due to droughts. We do not elicit similarity scores for the priming event of treat group one, i.e., losing cattle due to droughts, since this event is exactly the anticipated event.

Design details

Randomization method
Randomization by STATA
Randomization unit
Individual
Clusters
4,689
Observations
4,689
Arms
1171 in control, 1174 in recall treatment 1, 1173 in recall treatment 2, 1175 in recall treatment 3
Primary outcomes
Beliefs about the likelihood of cattle loss due to droughts

Cited by

No citation data available for this registration yet.

Keywords