Biased Beliefs About Random Samples: Evidence from Two Integrated Experiments

Daniel Benjamin, Don A. Moore () and Matthew Rabin ()
Additional contact information
Don A. Moore: University of California at Berkeley
Matthew Rabin: Harvard University

No GRU_2018_014, GRU Working Paper Series from City University of Hong Kong, Department of Economics and Finance, Global Research Unit

Abstract: We report two incentivized experiments on four errors in reasoning about random samples: the Law of Small Numbers, Non-Belief in the Law of Large Numbers, exact representativeness, and “bin effects.” We control for a variety of confounds that constrain prior work, test predictions of existing models, and assess the magnitudes of the biases. By asking each participant many different questions about the same data, we disentangle the biases from possible rational alternative interpretations. We find that no coherent model could jointly rationalize people’s beliefs about random sequences with their beliefs about distributions of outcomes.

Keywords: Law of Small Numbers; Gambler’s Fallacy; Non-Belief in the Law of Large Numbers; Big Data; Support Theory (search for similar items in EconPapers)
JEL-codes: B49 (search for similar items in EconPapers)
Pages: 65 pages
Date: 2018-06-22
New Economics Papers: this item is included in nep-exp
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.cb.cityu.edu.hk/ef/doc/GRU/WPS/GRU%232018-014%20Benjamin.pdf (application/pdf)

Related works:
Working Paper: Biased Beliefs About Random Samples: Evidence from Two Integrated Experiments (2017)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:cth:wpaper:gru_2018_014

Access Statistics for this paper

More papers in GRU Working Paper Series from City University of Hong Kong, Department of Economics and Finance, Global Research Unit Contact information at EDIRC.
Bibliographic data for series maintained by GRU ( this e-mail address is bad, please contact ).