Correct acceptance weighs more than correct rejection: A decision bias induced by question framing

Yaakov Kareev, Yaacov Trope

Research output: Contribution to journalArticle

Abstract

We propose that in attempting to detect whether an effect exists or not, people set their decision criterion so as to increase the number of hits and decrease the number of misses, at the cost of increasing false alarms and decreasing correct rejections. As a result, we argue, if one of two complementary events is framed as the positive response to a question and the other as the negative response, people will tend to predict the former more often than the latter. Performance in a prediction task with symmetric payoffs and equal base rates supported our proposal. Positive responses were indeed more prevalent than negative responses, irrespective of the phrasing of the question. The bias, slight but consistent and significant, was evident from early in a session and then remained unchanged to the end. A regression analysis revealed that, in addition, individuals' decision criteria reflected their learning experiences, with the weight of hits being greater than that of correct rejections.

Original languageEnglish (US)
Pages (from-to)103-109
Number of pages7
JournalPsychonomic Bulletin and Review
Volume18
Issue number1
DOIs
StatePublished - Feb 2011

Fingerprint

Regression Analysis
Learning
Weights and Measures
Rejection (Psychology)
Acceptance
Rejection
Phrasing
Prediction

Keywords

  • Decision making
  • Type I and Type II errors

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology

Cite this

Correct acceptance weighs more than correct rejection : A decision bias induced by question framing. / Kareev, Yaakov; Trope, Yaacov.

In: Psychonomic Bulletin and Review, Vol. 18, No. 1, 02.2011, p. 103-109.

Research output: Contribution to journalArticle

@article{03c2d4c54c11419fbf156f6c4332e95e,
title = "Correct acceptance weighs more than correct rejection: A decision bias induced by question framing",
abstract = "We propose that in attempting to detect whether an effect exists or not, people set their decision criterion so as to increase the number of hits and decrease the number of misses, at the cost of increasing false alarms and decreasing correct rejections. As a result, we argue, if one of two complementary events is framed as the positive response to a question and the other as the negative response, people will tend to predict the former more often than the latter. Performance in a prediction task with symmetric payoffs and equal base rates supported our proposal. Positive responses were indeed more prevalent than negative responses, irrespective of the phrasing of the question. The bias, slight but consistent and significant, was evident from early in a session and then remained unchanged to the end. A regression analysis revealed that, in addition, individuals' decision criteria reflected their learning experiences, with the weight of hits being greater than that of correct rejections.",
keywords = "Decision making, Type I and Type II errors",
author = "Yaakov Kareev and Yaacov Trope",
year = "2011",
month = "2",
doi = "10.3758/s13423-010-0019-z",
language = "English (US)",
volume = "18",
pages = "103--109",
journal = "Psychonomic Bulletin and Review",
issn = "1069-9384",
publisher = "Springer New York",
number = "1",

}

TY - JOUR

T1 - Correct acceptance weighs more than correct rejection

T2 - A decision bias induced by question framing

AU - Kareev, Yaakov

AU - Trope, Yaacov

PY - 2011/2

Y1 - 2011/2

N2 - We propose that in attempting to detect whether an effect exists or not, people set their decision criterion so as to increase the number of hits and decrease the number of misses, at the cost of increasing false alarms and decreasing correct rejections. As a result, we argue, if one of two complementary events is framed as the positive response to a question and the other as the negative response, people will tend to predict the former more often than the latter. Performance in a prediction task with symmetric payoffs and equal base rates supported our proposal. Positive responses were indeed more prevalent than negative responses, irrespective of the phrasing of the question. The bias, slight but consistent and significant, was evident from early in a session and then remained unchanged to the end. A regression analysis revealed that, in addition, individuals' decision criteria reflected their learning experiences, with the weight of hits being greater than that of correct rejections.

AB - We propose that in attempting to detect whether an effect exists or not, people set their decision criterion so as to increase the number of hits and decrease the number of misses, at the cost of increasing false alarms and decreasing correct rejections. As a result, we argue, if one of two complementary events is framed as the positive response to a question and the other as the negative response, people will tend to predict the former more often than the latter. Performance in a prediction task with symmetric payoffs and equal base rates supported our proposal. Positive responses were indeed more prevalent than negative responses, irrespective of the phrasing of the question. The bias, slight but consistent and significant, was evident from early in a session and then remained unchanged to the end. A regression analysis revealed that, in addition, individuals' decision criteria reflected their learning experiences, with the weight of hits being greater than that of correct rejections.

KW - Decision making

KW - Type I and Type II errors

UR - http://www.scopus.com/inward/record.url?scp=79951857425&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79951857425&partnerID=8YFLogxK

U2 - 10.3758/s13423-010-0019-z

DO - 10.3758/s13423-010-0019-z

M3 - Article

C2 - 21327363

AN - SCOPUS:79951857425

VL - 18

SP - 103

EP - 109

JO - Psychonomic Bulletin and Review

JF - Psychonomic Bulletin and Review

SN - 1069-9384

IS - 1

ER -