Time Pressure and Regret in Sequential Search

How to cite

Opitz, Timm; Klimm, Felix; Kocher, Martin; Schudy, Simeon. 2019. "Time Pressure and Regret in Sequential Search." AEA RCT Registry. AEARCTR-0004065. https://doi.org/10.1257/rct.4065-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Timm Opitz* , Felix Klimm , Martin Kocher , Simeon Schudy

* Primary investigator

Registry ID
AEARCTR-0004065
DOI
https://doi.org/10.1257/rct.4065-1.0
First registered
Last updated
Status
In development
Trial dates
2019-07-29 – 2019-08-09
Countries
Private
JEL codes
D47, D91
Registry record
http://www.socialscienceregistry.org/trials/4065

Abstract

Globalization has increased the diversity of goods, the numbers of competitors for goods and services, the set of job opportunities, as well as the amount of available knowledge with the potential to create new research opportunities and to stimulate innovation. This comes at the cost of having to acquire and process this decentralized information to make an optimal decision. At the same time, the available time for consideration of each offer is limited. In this study, we investigate (i) how time pressure influences sequential search processes and (ii) whether people forgo material benefits through insufficient search to avoid feelings of self-blame in case the additional search was not fortunate. In a 2x2 design, we modify the available time for making the decision on the on hand, and the informational structure on the other hand. We use a sequential search task in a laboratory setting to explore how time pressure and regret contribute to the common finding that subjects fail to achieve optimal outcomes in sequential decisions.

Experimental design

We use a sequential search task where the subjects try to find the optimal time of purchase in a price search for a fictitious product, while incurring a fixed cost for every offer they elicit. The offers are drawn from a known uniform distribution. After each offer, they decide whether to continue the search for another period or to take the best price observed so far, i.e. they can recall every previously rejected offer. Subjects can search as long as there is a possibility to achieve a positive payoff and are not aware of the (theoretically) finite nature of the search process.

We use a 2x2 between-subject design in which we vary time pressure and feedback for participants in a sequential search task. We induce time pressure by limiting the amount of time an individual can spend on each search step (i.e. buy the product vs. draw another offer). A transgression of this limit is punished by a deduction of points from the initial endowment and no new offer is shown until a decision has been made. Orthogonal to that treatment, we vary feedback (expectations). While in one treatment, the search ends after the offer has been chosen, we introduce a feedback-treatment where subjects see additional realizations of prices after the end of the search. Thus, they see offers they would have received if they had continued. The number of actually displayed (additional) offers is determined by a lottery to avoid that people perceive the search as finite.

The search task is repeated for 10 rounds. We vary the theoretically optimal cutoff value to stop the search by altering the search costs across rounds.

In addition to the repeated search task in the 2x2 between-subject design, participants perform 2 rounds of non-sequential search with pre-commitment. In those rounds, subjects prespecify a price at or below which they are willing to buy the good.

The experiment will take place at the Munich Experimental Laboratory for the Social Sciences (MELESSA).

Design details

Randomization method
All randomizations are done by a computer, either in real time or in the first experimental session.
Randomization unit
Individual
Clusters
192 (8 Sessions*24) Participants
Observations
192
Arms
48 (2 Sessions*24 participants)
Primary outcomes
The primary outcomes are the search length and the stopping offer (i.e. the actually taken offer and the offer after which the search is stopped). (See preanalysis plan for details).
Secondary outcomes
See preanalysis plan for details.

Cited by

No citation data available for this registration yet.

Keywords