Back to History

Fields Changed

Trial
Field Before After
Trial End Date October 4, 2019 November 8, 2019
Last Published September 16, 2019 02:03 PM October 28, 2019 02:20 PM
Intervention End Date October 4, 2019 November 8, 2019
Experimental Design (Public) Subjects are asked to allocate approximately 30 minutes of work between pairs of work days occurring over the course of two weeks. Subjects allocate work between Monday of Week 1 and Monday of week 2; subjects repeat this procedure for Tuesday, Wednesday, and Thursday. Decisions for Mondays are made on the first Monday. Decisions for Tuesdays are made on the first Tuesday. Decisions for Wednesdays are made both on Monday and Wednesday. Decisions for Thursdays are made on both Tuesday and Thursday. Monday and Wednesday constitute Treatment A; Tuesday and Thursday constitute Treatment B. For each subject, one treatment is randomly assigned the RIS treatment, while the other treatment is randomly assigned the certainty treatment. Thus treatments are randomly ordered between subjects. Both treatments offer five discount rates $R \in \{0.5, 0.75, 1, 1.25, 1.5\}$ at which subjects must allocate a present value of 360 tasks between the pair of days in question. Subjects know that for the first day of the RIS treatment, one of their five decisions made on that same day will be implemented. Subjects also know that for the second day of the RIS treatment, one of their five decisions from that day or one of their five decisions from the previous day in that treatment will be implemented. Each decision is implemented with uniform probability. Subjects know that for the first day of the certainty treatment, the decision corresponding to $R = 1.25$ will be implemented with certainty, while the decisions for the other four rates are merely hypothetical. Subjects also know that for the second day of the certainty treatment, the decision corresponding to $R = 1.25$ made on that same day will be implemented with certainty, while the allocations made for the other four rates on that day and the allocations made for all five rates on the previous day in that treatment are merely hypothetical. Subjects must complete an intake questionnaire that tests comprehension of the experimental instructions; subjects must answer all questions correctly to be enrolled in the trial. On Day 1 (e.g., Monday of Week 1), subjects are asked to allocate approximately 30 minutes of work between Day 2 (e.g., Wednesday of Week 1) and Day 3 (e.g., Wednesday of Week 2). On Day 2, subjects are again asked to allocate work between Day 2 and Day 3. For each subject, an independent fair coin toss determines whether decisions from Day 1 or Day 2 are implemented. If a subject is assigned to the “Uncertain-Day” treatment, she learns the result of this coin toss after her decisions on Day 2. If a subject is assigned to the “Certain-Day” treatment, she learns the result of this coin toss before her decisions on Day 2. On each day that decisions are made, subjects are asked to allocate a fixed present value budget of real-effort tasks between Days 2 and 3. Subjects make allocation decisions for each of five gross interest rates R ∈ {0.5, 0.75, 1, 1.25, 1.5}. If a subject is assigned to the “Uncertain-List” treatment, after she makes all of her decisions, one of her decisions from the selected day is chosen at random with uniform probability to be implemented. If a subject is assigned to the “Certain-List” treatment, her decision from the selected day for the gross interest rate R = 1.25 is chosen with certainty to be implemented. Each subject is assigned to a “Day” treatment and a “List” treatment, resulting in a two-by-two factorial design. Each subject is fully informed of the procedures and mechanisms used in her assigned treatment.
Randomization Method Randomization is done in advance using a computer program. Subjects are enrolled in the order in which they complete the intake questionnaire. Treatments are assigned to subjects using a round-robin algorithm as subjects are enrolled. Randomization of gross interest rates is done in advance using a computer program (using unique seeds to permit replication).
Randomization Unit The two treatments are randomly ordered by subject. The decisions chosen for implementation are randomized in advance. Subjects are individually assigned to one of the four treatments.
Planned Number of Clusters The pilot study will have 12 individual subjects; each subject is clustered. The full study will have 60 individual subjects; each subject is clustered. The trial will have 192 to 208 subjects; each subject represents a cluster.
Planned Number of Observations The pilot study will have 12 individual subjects each making 30 task allocation decisions, yielding 360 observations. The full study will have 60 individual subjects each making 30 task allocation decisions, yielding 1800 observations. Each subject makes 20 allocation decisions. The trial will thus yield 3840 to 4160 observations.
Sample size (or number of clusters) by treatment arms The two treatments are conducted within-subjects. Each treatment in the pilot study has 12 subjects/clusters and 360 observations. Each treatment in the full study has 60 subjects/clusters and 1800 observations. Certain-Day-Certain-List: 64 subjects. Certain-Day-Uncertain-List: 64 subjects. Uncertain-Day-Certain-List: 32 subjects. Uncertain-Day-Uncertain-List: 32 subjects.
Secondary Outcomes (End Points) Subjects are asked to complete daily survey questions regarding their outside labor market participation which are used to help calibrate the structural model. Subjects are asked to complete an exit survey that includes portions of the Behavioral Risk Factor Surveillance System Questionnaire which are used to look for relationships between time and risk preferences and included health survey responses. Subjects must complete an intake survey with demographic and socioeconomic survey questions in order to participate in treatments; these data will be used to predict attrition and may also be used as controls in prediction of primary results. The intake questionnaire includes a short demographic survey; these data will be used to predict attrition and may also be used as controls in prediction of primary results.
Back to top