TY - JOUR
T1 - Experimenting with measurement error
T2 - Techniques with applications to the caltech cohort study
AU - Gillen, Ben
AU - Snowberg, Erik
AU - Yariv, Leeat
N1 - Funding Information:
Snowberg gratefully acknowledges the support of NSF grants SES-1156154 and SMA-1329195. Yariv gratefully acknowledges the support of NSF grants SES-0963583 and SES-1629613 and Gordon and Betty Moore Foundation grant 1158. We thank Jonathan Bendor, Christopher Blattman, Colin Camerer, Marco Castillo, Gary Charness, Lucas Coffman, Guillaume Frechette, Dan Friedman, Drew Fudenberg, Yoram Halevy, Ori Heffetz, Muriel Nie-derle, Alex Rees-Jones, Shyam Sunder, Roel van Veldhuizen, and Lise Vesterlund, as well
Funding Information:
Snowberg gratefully acknowledges the support of NSF grants SES-1156154 and SMA-1329195. Yariv gratefully acknowledges the support of NSF grants SES-0963583 and SES-1629613 and Gordon and Betty Moore Foundation grant 1158. We thank Jonathan Bendor, Christopher Blattman, Colin Camerer, Marco Castillo, Gary Charness, Lucas Coffman, Guillaume Frechette, Dan Friedman, Drew Fudenberg, Yoram Halevy, Ori Heffetz, Muriel Niederle, Alex Rees-Jones, Shyam Sunder, Roel van Veldhuizen, and Lise Vesterlund, as well as two anonymous reviewers and the editor, Emir Kamenica, for comments and suggestions. We also appreciate the input of seminar audiences at Caltech, Hong Kong University of Science and Technology, the Ifo Institute, Nanyang Technological University, the National University of Singapore, Stanford Institute for Theoretical Economics, the University of Bonn, the University of British Columbia, the University of Southern California, and the University of Zurich. Data are provided as supplementary material online.
Publisher Copyright:
© 2019 by The University of Chicago. All rights reserved.
PY - 2019/8/1
Y1 - 2019/8/1
N2 - Measurement error is ubiquitous in experimental work. It leads to imperfect statistical controls, attenuated estimated effects of elicited be-haviors, and biased correlations between characteristics. We develop statistical techniques for handling experimental measurement error. These techniques are applied to data from the Caltech Cohort Study, which conducts repeated incentivized surveys of the Caltech student body. We replicate three classic experiments, demonstrating that results change substantially when measurement error is accounted for. Collectively, these results show that failing to properly account for measurement error may cause a field-wide bias leading scholars to identify “new” phenomena.
AB - Measurement error is ubiquitous in experimental work. It leads to imperfect statistical controls, attenuated estimated effects of elicited be-haviors, and biased correlations between characteristics. We develop statistical techniques for handling experimental measurement error. These techniques are applied to data from the Caltech Cohort Study, which conducts repeated incentivized surveys of the Caltech student body. We replicate three classic experiments, demonstrating that results change substantially when measurement error is accounted for. Collectively, these results show that failing to properly account for measurement error may cause a field-wide bias leading scholars to identify “new” phenomena.
UR - http://www.scopus.com/inward/record.url?scp=85067567612&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85067567612&partnerID=8YFLogxK
U2 - 10.1086/701681
DO - 10.1086/701681
M3 - Article
AN - SCOPUS:85067567612
SN - 0022-3808
VL - 127
SP - 1826
EP - 1863
JO - Journal of Political Economy
JF - Journal of Political Economy
IS - 4
ER -