Open- versus Closed-Ended Elicitation in Surveys and the Predictive Content of Stated Values: Evidence from Sustainable Retail Investing
How to cite
Wekhof, Tobias. 2026. "Open- versus Closed-Ended Elicitation in Surveys and the Predictive Content of Stated Values: Evidence from Sustainable Retail Investing." AEA RCT Registry. AEARCTR-0017723. https://doi.org/10.1257/rct.17723-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
This study examines how different survey question formats shape what people report as important, and how well those reports predict later decisions. Some surveys ask open-ended questions where respondents describe priorities in their own words, while others use checklists where respondents select items from a list. Using a randomized survey of retail investors in Switzerland, the study compares these two formats and evaluates how strongly stated priorities are reflected in a subsequent hypothetical investment choice that involves a tradeoff between sustainability and management fees.
Experimental design
Participants complete a survey with two formats of value elicitation: an open-ended question (respondents describe investment priorities in their own words) and a closed-ended checklist (respondents select relevant priorities from a list). The order of these two formats is randomized across participants. The treatment group first answered the closed-ended question and then the open-ended question (i.e., the respondents were made aware of the different options from the closed-ended question). The second treatment arm (control) first answered the open-ended question and then the closed-ended question (i.e., these respondents were not aware of the closed-ended options when answering the open-ended question). After the value elicitation, respondents complete a hypothetical investment choice between two portfolios that differ in sustainability characteristics and management fees. The study compares how responses differ across question formats and how well each format predicts the subsequent choice.
Design details
Cited by
No citation data available for this registration yet.