TY - JOUR
T1 - Order-constrained inference to supplement experimental data analytics in behavioral economics
T2 - A motivational case study
AU - Ludwig, Jonas
AU - Cavagnaro, Daniel R.
AU - Regenwetter, Michel
N1 - We gratefully acknowledge National Science Foundation grants SES # 10–62,045 (PI: M. Regenwetter), SES # 14–59,699 (PI: M. Regenwetter), and SES # 20–49,896 (PI: M. Regenwetter, Co-PI: D. Cavagnaro). The development of QTest was further supported by a grant from the Alexander von Humboldt Foundation (Co-PIs: J. Stevens and M. Regenwetter). The first author is supported by a grant from the John Templeton Foundation. We further thank Sylvia E, Emily Neu Line, Alexandra Ortmann, Carrie Dale Shaffer-Morrison, Ori Weisel, and the participants of the 2nd Advances in Behavioral Decision Analytics Summer School held in Champaign in June 2022 for valuable discussions and feedback. Data and materials used in this manuscript are available on the OSF, see https://osf.io/3va2w .
Author Note: Ludwig contributed to this project with the project idea, translation of verbal hypotheses into order constraints, data analytics using QTEST, and nearly all of the writing of the manuscript. The first author also served as the team expert on the substantive research area. Cavagnaro and Regenwetter trained the first author in order-constrained modeling and inference at a summer school, helped translate verbal hypotheses into mathematical models, ran the final the QTEST analyses on powerful computers, helped with the interpretation of the results, and edited various versions of the manuscript. All authors have read and approved the final manuscript. We gratefully acknowledge National Science Foundation grants SES # 10–62,045 (PI: M. Regenwetter), SES # 14–59,699 (PI: M. Regenwetter), and SES # 20–49,896 (PI: M. Regenwetter, Co-PI: D. Cavagnaro). The development of QTEST was further supported by a grant from the Alexander von Humboldt Foundation (Co-PIs: J. Stevens and M. Regenwetter). The first author is supported by a grant from the John Templeton Foundation. We further thank Sylvia E, Emily Neu Line, Alexandra Ortmann, Carrie Dale Shaffer-Morrison, Ori Weisel, and the participants of the 2nd Advances in Behavioral Decision Analytics Summer School held in Champaign in June 2022 for valuable discussions and feedback. Data and materials used in this manuscript are available on the OSF, see https://osf.io/3va2w. This work has been presented at the 65th Conference of Experimental Psychologists (TeaP, Trier, Germany). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of funding agencies or the authors’ home institutions. Jonas Ludwig is now at Technische Universität Berlin (Email: [email protected]).
PY - 2023/12
Y1 - 2023/12
N2 - A common approach to theory testing in behavioral and experimental economics relies on null hypothesis significance testing via (generalized) linear regression models. Here, we showcase order-constrained inference as an alternative route to theory testing. Order-constrained inference can improve the precision and nuance of behavioral decision analytics. For example, the method can be leveraged to quantify the evidence in support of, or against, a given hypothesis. It also offers advanced model selection tools for quantitative competition among multiple theories. To illustrate our case for order-constrained methods, we re-analyze data from a pre-registered experiment on incentives, cognitive reflection, and dishonest behavior. Building on this publicly available dataset, we further highlight the advantages of Bayesian order-constrained inference. We discuss how the method can deliver more convincing and more nuanced evidence than frequentist null hypothesis significance testing, pointing to new research avenues for supplementing and expanding on experimental designs in behavioral economics.
AB - A common approach to theory testing in behavioral and experimental economics relies on null hypothesis significance testing via (generalized) linear regression models. Here, we showcase order-constrained inference as an alternative route to theory testing. Order-constrained inference can improve the precision and nuance of behavioral decision analytics. For example, the method can be leveraged to quantify the evidence in support of, or against, a given hypothesis. It also offers advanced model selection tools for quantitative competition among multiple theories. To illustrate our case for order-constrained methods, we re-analyze data from a pre-registered experiment on incentives, cognitive reflection, and dishonest behavior. Building on this publicly available dataset, we further highlight the advantages of Bayesian order-constrained inference. We discuss how the method can deliver more convincing and more nuanced evidence than frequentist null hypothesis significance testing, pointing to new research avenues for supplementing and expanding on experimental designs in behavioral economics.
KW - Bayesian statistics
KW - Order-constrained inference
KW - Regression analysis
UR - http://www.scopus.com/inward/record.url?scp=85173935507&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85173935507&partnerID=8YFLogxK
U2 - 10.1016/j.socec.2023.102116
DO - 10.1016/j.socec.2023.102116
M3 - Article
AN - SCOPUS:85173935507
SN - 2214-8043
VL - 107
JO - Journal of Behavioral and Experimental Economics
JF - Journal of Behavioral and Experimental Economics
M1 - 102116
ER -