'Supportive vs. consequences-focused court date reminders' (AsPredicted #124,602)
Author(s) This pre-registration is currently anonymous to enable blind peer-review. It has 4 authors.
Pre-registered on 2023/03/09 12:59 (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? We are sending text message reminders to public defender clients to help these individuals plan for and remember upcoming court dates. These court dates are often mandatory, and failing to attend these court dates can result in the receipt of a bench warrant, resulting in eventual incarceration.
For this experiment, we are comparing two reminder templates and asking two main questions about these templates:
1. Overall treatment effect: Are supportive court date reminders or consequences-focused court date reminders more effective at reducing bench warrants for public defender clients compared to no messages, and compared to each other?
2. Heterogeneous treatment effects: Are supportive or consequences-focused court date reminders more or less effective at reducing bench warrants for certain subpopulations? We are specifically interested in treatment effects for people with a history of bench warrants.
3) Describe the key dependent variable(s) specifying how they will be measured. The dependent variable is issuance of a bench warrant for: (1) the first scheduled court date; and (2) within 28 days of the first scheduled court date. We may additionally examine impacts on incarceration, conditional on data availability.
4) How many and which conditions will participants be assigned to? Three conditions: clients will be randomized with equal probability to one of three treatments:
1. No court date reminders
2. Supportive court date reminders
3. Consequences-focused court date reminders
Reminders will be sent 7-, 3-, and 1-day before each upcoming court date after assignment to one of the two reminder conditions. Translated reminders will be sent instead of English if clients have indicated a preference for a translator in Spanish and Vietnamese. Additional automated responses will be provided for unprompted confirmations, declinations, or other responses. These automated responses will be identical across the two reminder arms.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. Logistic regression predicting the DVs listed above as a function of treatment assignment (control vs. consequences vs. supportive) and other relevant covariates, e.g., demographic variables like age, case information like alleged charges, and known bench warrant history before the reminder.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. We will include all experiment participants in our analysis. We will also continue to run a simultaneous experiment on a random subset of public defender clients, testing the impact of a previous version of a consequences-focused reminder on bench warrant rates. These clients will be excluded from this study.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. We will run the experiment for 12 months at a minimum. At approximately 300 clients per month, we expect this to result in 3,600 clients. Depending on the support of our partner site, we may add up to another 12 months for up to 7,200 clients.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) Co-authors will include Emma Brunskill, Julian Nyarko, and Todd Rogers.
