The Future in Mind: Aspirations and Forward-Looking Behaviour in the Short and Long Run in Rural Ethiopia
How to cite
Orkin, Kate; Taffesse, Alemayehu; Bernard, Tanguy; Dercon, Stefan.
2017. "The Future in Mind: Aspirations and Forward-Looking Behaviour in the Short and Long Run in Rural Ethiopia." AEA RCT Registry. AEARCTR-0001483.
https://doi.org/10.1257/rct.1483-2.1
Download:
BibTeX
·
RIS— for reference managers without a browser
translator. Zotero and most
others pick up this page's citation metadata directly.
Kate Orkin*
, Alemayehu Taffesse
, Tanguy Bernard
, Stefan Dercon
Poor people often do not make investments, even when returns are high. One possible explanation is that they have low aspirations and form ideas of their future opportunities which ignore some options for investment. We outcome our plan to test this hypothesis through a field experiment to identify the effects of a documentary screening intervention.
This study is a two-level cluster-randomized controlled trial. We randomly selected villages in a remote district in rural Ethiopia. We assign them to pure control or to receive one of two village-level treatments. Then within villages, we selected a random sample of 18 households and assigned them to three groups. A treatment group were randomly invited to watch documentaries. A placebo group in each village watched an Ethiopian entertainment programme and a control group were simply surveyed.
In addition, the number of people treated was varied by village to assess the importance of peer effects in formation of aspirations. We conduct follow-ups in 64 villages after 6 months and in all 84 villages after five years.
In this pre-analysis plan, we describe analysis on data collected via face-to-face household surveys in 2010 (baseline), immediately after screenings in 2010 (midline 1), after six months in 2011 (midline 2), and after five years in 2015 (endline) and through surveys with the kebele (village) leader at baseline and endline. We also use administrative data at village level collected before our baseline.
Intervention
Interventions are detailed in a working paper on the midline results: https://www.csae.ox.ac.uk/workingpapers/pdfs/csae-wps-2014-16.pdf The documentaries, with English subtitles, and one of four placebo segments are at https:// www.youtube.com/channel/UCqfoNjCzt8YPjTRWQaMQfAg.
Experimental design
Design is detailed in a working paper on the midline results: https://www.csae.ox.ac.uk/workingpapers/pdfs/csae-wps-2014-16.pdf
Design details
Randomization method
Village-level randomization done in office by a computer. Within-village randomisation done by public lottery within the village.
Randomization unit
The first level of randomisation was at village level. 20 of the 84 villages were assigned to be pure control villages. The remaining randomly chosen 64 villages were grouped into 16 screening sites with four neighbouring villages in each site. In each screening site of four villages, we randomly assigned two villages to be "intense treatment" villages and the other two to be "intense placebo" villages.
The second level of randomisation was at household level. In all villages, the enumerators compiled a list of all households. In the pure control villages, they randomly selected approximately 15 households to be surveyed. In the "intense treatment" and "intense placebo" villages, they randomly selected 18 households and randomly assigned 6 households to each of the treatment, placebo and within-village spillover groups. Randomisation was at household level, so household heads and their spouses had the same treatment status.
In addition, in the “intense treatment" villages, we randomly selected 18 additional households (roughly 36 individuals) to be invited to the documentary but did not survey these
individuals. In the “intense placebo" villages, we similarly randomly selected 18 additional
households to be invited to the placebo session.
Clusters
84 villages
Observations
Baseline: 2,115 individuals. Endline: dependent on attrition, but potentially up to 2,746 individuals.
Arms
20 of the 84 villages were assigned to be pure control villages. In the pure control villages, we randomly selected approximately 15 households to be surveyed.
The remaining randomly chosen 64 villages were grouped into 16 screening sites with four neighbouring villages in each site. In each screening site of four villages, we randomly assigned two villages to be "intense treatment" villages and the other two to be "intense placebo" villages. In the "intense treatment" and "intense placebo" villages, we randomly selected 18 households and randomly assigned 6 households to each of the treatment, placebo and within-village control groups. In addition, in the "intense treatment" villages, we randomly selected 18 additional households (roughly 36 individuals, or 1,152 individuals in total) to be invited to the documentary but did not survey these individuals. In the "intense placebo" villages, we similarly randomly selected 18 additional (non-surveyed) households to be invited to the placebo session.
Primary outcomes
See attached pre-analysis plan
Papers from this trial
Bernard, T., S. Dercon, K. Orkin, and A. S. Taffesse. 2014. “The Future in Mind: Aspirations and Forward-Looking Behaviour in Rural Ethiopia.” Bureau for Research and Economic Analysis of Development Working Paper 429: 1–42.
http://ibread.org/bread/working/429
Cited by
No citation data available for this registration yet.