A randomized experiment to investigate the effects of increasing sophistication on stated demand for commitment and time inconsistency in physical activity.

Last registered on November 26, 2020
View Trial History

Pre-Trial

Trial Information

General Information

Title
A randomized experiment to investigate the effects of increasing sophistication on stated demand for commitment and time inconsistency in physical activity.
RCT ID
AEARCTR-0004857
Initial registration date
October 18, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
October 18, 2019, 10:40 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
November 26, 2020, 2:39 AM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country
Netherlands
Region

Primary Investigator

Name
Diarmaid O Ceallaigh
Affiliation
Erasmus University Rotterdam
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Kirsten I.M. Rohde
PI Affiliation
Erasmus University Rotterdam
Contact Investigator
PI Name
Hans van Kippersluis
PI Affiliation
Erasmus University Rotterdam
Contact Investigator

Additional Trial Information

Status
In development
Start date
2019-10-21
End date
2021-07-31
Keywords
Health
Additional Keywords
behavioral economics, physical activity, time inconsistency, sophistication, commitment
JEL code(s)
Secondary IDs
Abstract
Commitment contracts have been shown to be effective to overcome time inconsistent health behavior, including physical activity. However, the take-up of these contracts is usually low. According to several behavioral economic theories, including the quasi-hyperbolic model and dual-self models, the demand for commitment should be increasing in the individual's sophistication, or awareness of their time inconsistency problems. We posit that an intervention to increase sophistication can boost the demand for commitment. We also test the relation between the level of sophistication and the level of time inconsistent preferences (the extent to which an individual's ex-ante ideal level of physical activity exceeds the actual physical activity they carry out). The dual-self model predicts that the level of time inconsistent preferences is decreasing in sophistication, while the quasi-hyperbolic model suggests that this level is independent of sophistication. We carry out these tests in a large multi-wave general population survey, where we measure the time inconsistency, sophistication and stated demand for commitment in physical activity of respondents. We allocate a subset of this population to a treatment to increase their sophistication, and measure the treatment effect on time inconsistency, sophistication and demand for commitment.
External Link(s)

Registration Citation

Citation
O Ceallaigh, Diarmaid, Kirsten I.M. Rohde and Hans van Kippersluis. 2020. "A randomized experiment to investigate the effects of increasing sophistication on stated demand for commitment and time inconsistency in physical activity.." AEA RCT Registry. November 26. https://doi.org/10.1257/rct.4857-3.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2019-10-21
Intervention End Date
2020-07-31

Primary Outcomes

Primary Outcomes (end points)
Physical activity sophistication levels, time inconsistent behavior, time inconsistent preferences and demand for commitment.
Primary Outcomes (explanation)
Note: This is a three-wave longitudinal survey with each survey completed by respondents approximately 2 weeks apart.
Physical activity sophistication levels will be measured by comparing ex-post hours of physical activity actually carried out (as reported by respondents in wave 2(3) of the survey) to ex-ante predictions made by respondents of hours of physical activity for those two weeks (as made in wave 1(2) of the survey).
Time inconsistent behavior will be measured using the same comparison as above, as ex-ante predictions also represent ex-ante plans for physical activity.
Time inconsistent preferences will be measured by comparing ex-post hours of physical activity actually carried out (as reported in wave 2(3) of the survey) to the respondent's ex-ante ideal hours of physical activity for those two weeks (as reported in wave 1(2) of the survey).
Stated demand for commitment will be measured in each of waves 1-3 using hypothetical questions eliciting demand for commitment devices where repayment of a deposit or payment of a reward is contingent on meeting physical activity targets.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We randomly assign respondents to a control group or to one of two different treatment groups, which receive a sophistication intervention. Further details are provided in the Hidden Experimental Design which will be visible after the completion of the study.
Experimental Design Details
Respondents are randomly assigned to either the control group, treatment A or treatment B, with equal probability of being assigned to each group. The sophistication intervention is implemented in wave 2 of this three wave survey (each wave is approximately two weeks apart). The intervention involves providing each respondent with information on their physical activity time inconsistency levels in the two weeks since wave 1 of the survey. This is done by showing them their actual number of hours of physical activity in those two weeks (as reported by them in wave 2) and their ex-ante ideal hours for those two weeks as stated by them in wave 1. Treatment A will receive this information before they answer wave 2 questions eliciting their ideal and predicted number of hours of physical activity for the next two weeks, while Treatment B will receive this information after they have answered these questions.

The treatment effect on measured sophistication levels will serve as a manipulation check, while the treatment effect on stated demand for commitment (in wave 2 and 3) and time inconsistency (in the two weeks after wave 2) serve as our main outcome measures.
Our primary analysis involves estimating treatment effects on a between-individual basis based on differences in outcome measures between treatment groups and control post-intervention (e.g. differences between treatment A and the control in stated demand for commitment in wave 2 (post-intervention) as well as differences between treatment A and the control in the change in stated demand for commitment in wave 2 (post-intervention) compared to wave 1 (pre-intervention)). We will also carry out a secondary analysis of within-individual treatment effects based on changes in outcome measures measured pre- and post-intervention for a single individual over time (i.e. a change in stated demand for commitment for an individual in wave 1 (pre-intervention) compared to wave 2 (post-intervention)).
Note that the intervention can affect time inconsistent preferences for the two weeks after wave 2 (ex-ante ideal level of physical activity for the next two weeks as stated in wave 2 minus the actual level of physical activity carried out in those two weeks as reported in wave 3) through two channels. The first is by affecting the ex-ante ideal level. The second is by affecting the actual level of physical activity carried out. The purpose of treatment B is to allow us to disentangle these two effects. Any differences in time inconsistent preferences between treatment B and the control will operate through the second channel, while differences between treatment A and the control can operate through both channels.
Randomization Method
Randomization carried out automatically in Qualtrics online survey software.
Randomization Unit
Individual
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
N/A
Sample size: planned number of observations
Invitations to take part in this survey will be sent out to approximately 65,000 members of the Lifelines general population cohort study in the Netherlands. We estimate that approximately 15,000 respondents will complete all three waves of the survey.
Sample size (or number of clusters) by treatment arms
5,000 respondents per group (equal allocation of the expected number of respondents of 15,000 between control and two treatment groups).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Medisch Ethische Toetsingscommissie METC UMC Groningen
IRB Approval Date
2019-09-03
IRB Approval Number
METc 2019/464
Analysis Plan

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials