Social Influence: The Power of Monetary and Symbolic Social Rank
How to cite
Martinangeli, Andrea; Cicognani, Simona; Benistant, Julien; Sutan, Angela; Meiske, Biljana. 2026. "Social Influence: The Power of Monetary and Symbolic Social Rank." AEA RCT Registry. AEARCTR-0018934. https://doi.org/10.1257/rct.18934-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
We study whether exogenously assigned rank makes advice more influential in an individual decision setting. People rely on rank cues when deciding whose advice to follow. Participants choose what share of an endowment to invest in a lottery that either triples the investment or reduces it to zero with equal probability. Before deciding, they receive unsolicited advice from a prior participant recommending to invest a proportion of the endowment. In the monetary treatment, the advisor is described as having randomly received either a low or high additional budget; in the star treatment, the advisor is randomly given one or two stars. In both cases, rank is explicitly random and therefore not informative about ability. Depending on the advisors' monetary or symbolic rank condition, the decision makers know that they are also randomly allocated an endowment or stars from the same set out of which the advisor's rank is selected. They are however not told which endowment size/number of stars they are allocated. This strategy allows the decision maker to perceive themselves as part of the same hierarchy as the advisors (the hierarchy matters), but avoids envy and rank-homophily. We measure the absolute distance between the participant’s choice and the advice received. Advice is restricted to 30% or 70% to reduce irrelevant variation and keep the content comparable across conditions. We hypothesize that advice from higher-ranked advisors will be followed more closely in both treatments, though possibly more strongly under monetary rank. Additional measures include cognitive reflection, social dominance orientation, recall, and perceptions of the advice. The main analysis estimates pooled and treatment-specific models, with a double-bounded Tobit as the primary specification and OLS and probit as robustness checks.
Intervention
We investigate whether individuals are influenced to different extents by advice from high vs low ranked advisors in their lottery investment decisions.
Experimental design
Participants are randomly assigned to one of two rank environments and then make a single incentivized investment decision under risk. Before choosing, each participant receives advice from a prior participant about how much of an endowment to invest in a lottery. The
experimental variation is the advisor’s randomly assigned rank: in one environment, rank is conveyed by a low or high monetary endowment; in the other, by one or two stars. Advice content is restricted to a small set of pre-selected values and is orthogonal to advisor rank. The main outcome is how closely participants’ choices align with the advice they receive.
Design details
Cited by
No citation data available for this registration yet.