#59058 | AsPredicted

'video and text messages about COVID vaccination 02 22 2021'
(AsPredicted #59058)


Created:       02/22/2021 04:19 PM (PT)

This is an anonymized version of the pre-registration.  It was created by the author(s) to use during peer-review.
A non-anonymized version (containing author names) should be made available by the authors when the work it supports is made public.

1) Have any data been collected for this study already?
It's complicated. We have already collected some data but explain in Question 8 why readers may consider this a valid pre-registration nevertheless.

2) What's the main question being asked or hypothesis being tested in this study?
We plan to run an online experiment with a sample of Americans to measure the effect of 4 different text messages aimed at encouraging individuals to schedule a COVID-19 vaccination appointment at their local healthcare provider. In particular, we will ask participants to imagine being a hospital patient and receiving a text message about their eligibility to get the COVID vaccine. In a 2x2 design, we vary whether the messages contain the link to a video providing information about COVID-19 (e.g., the prevalence of COVID-19 and the effectiveness of the COVID-19 vaccine) and whether the messages use enhanced language aimed at facilitating people with follow-through. We will investigate the perceived persuasiveness of the messages and the extent to which these messages affect people’s likelihood of scheduling a COVID-19 vaccination appointment. We will explore the underlying mechanisms.

3) Describe the key dependent variable(s) specifying how they will be measured.
1. How persuasive do you think the message is? (1 = not at all persuasive, 7 = very persuasive)
2. How likely would you be to schedule a vaccination appointment after receiving this message from your healthcare provider? (1 = not at all likely, 7= very likely)

4) How many and which conditions will participants be assigned to?
Participants will be randomly assigned to one of four conditions.
1. In the simple text condition, the message is, "[Your first name], you can get the COVID-19 vaccine at our clinics. Make a vaccination appointment here: [link to the appointment website]"
2. In the Simple Text+Video, the message is, "[Your first name], you can get the COVID-19 vaccine at our clinics. Please watch this important 2 min video: [Link to the video]. Make a vaccination appointment here: [Link to the appointment website]"
3. In the Enhanced Text, the message is, "[Your first name], a COVID-19 vaccine has just been made available to you at our clinics. Claim your dose today by making a vaccination appointment here: [link to the appointment website]". People will watch an educational video about covid-19.
4. In the Enhanced Text+Video, the message is, "[Your first name], a COVID-19 vaccine has just been made available to you at our clinics. Please take 2 simple steps: 1. Watch this important 2 min video: [Link to the video]. 2. Claim your dose today by making a vaccination appointment here: [Link to the appointment website]". People will watch an educational video about covid-19.

5) Specify exactly which analyses you will conduct to examine the main question/hypothesis.
For the main analysis, we will run an ordinary least squares regression (OLS) to predict (1) perceived persuasiveness (1-7) and (2) likelihood of scheduling a vaccination appointment (1-7). The predictor variables will include (1) an indicator for random assignment to the message with the Video, (2) an indicator for random assignment to the message with Enhanced Text, and 3) the interaction between these indicators.

We will include the following control variables:
1. Indicators for participant gender (male, female, other/unknown)
2. Participant age
3. Indicators for participant race/ethnicity (Black non-Hispanic, Hispanic, Asian, white non-Hispanic, other/unknown)

6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations.
Participants recruited through MTurk (18 or older) based in the United States and who pass 4 screening questions will take our study. Only those who answer our primary dependent variables (Section 3) will be included in our final sample.

The screening questions will include 1) a captcha question to filter out bots, 2) an attention check question assessing whether participants are reading instructions closely, 3) a question asking whether participants have already received the COVID-19 vaccine or had an appointment (if yes, they will not be able to take the study), and 4) a question asking which political party participants identify the most with (the Democratic party, the Republican party, and independent; those who choose independent will be asked further whether they identify relatively more with the Democratic or Republican party or equally between the two parties). For 4), only participants who identify with either the Democratic party or the Republican party (including those who initially select independent but then choose one of the two parties in the follow-up question) can take the study. Also, to achieve a balanced sample, we will use the quotas in our Qualtrics survey to get half of the sample from people identifying with the Democratic party and half from people identifying with the Republican party. Once the quota is reached for one party, then new participants in that party will not be able to take the 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 already collected one wave of data for 400 people. We plan to collect the second wave of 800 people, which has 80% power to detect the main effects of interest. We plan to combine the first and second waves so we have more data to analyze the heterogeneous treatment effect listed in Section 8.

8) Anything else you would like to pre-register?
(e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?)

We will collect the following measures:
1) Participants’ belief in how likely they would catch the COVID-19 with (1a) versus without (1b) COVID-19 vaccine (ranging from -100% to 100%) and the difference between these measures (1c)
2) Participants’ belief in how likely they would develop symptomatic COVID-19 with (2a) versus without (2b) COVID-19 vaccine (ranging from -100% to 100%) and the difference between these measures (2c)
3) Anticipated regret for not taking the COVID-19 vaccine (1-7 scale);
4) How worried participants are about passing COVID-19 to other people (1-7 scale);
5) Perceptions of trust in the vaccine (concerns about side effects, worries about long term impact, concerns about the vaccine being authorized too quickly, trust in the research and development of the vaccine, safety of the vaccine)
6) Extent to which patients feel that the vaccine is already theirs (1-7 scale)
7) Extent at which the message makes participants feel that getting the COVID-19 vaccine is easy (1-7 scale)

We will use the same OLS regression as described in Section 5 to predict variables 1a-7. We will also look at the average effect of watching the video by regressing variables 1a-5 on an indicator for being randomly assigned to watching the video. We will conduct mediation analyses and see which items predict intentions to get the COVID-19 vaccine.

We will explore the following moderators:
1. Whether the participant received a flu shot either in 2019-20 flu season or the 2020-2021 flu season
2. The political party that the participant identifies with (democratic vs. republican)
3. Education (Below bachelor or bachelor/above bachelor)
4. How often the participant wears masks when going outside (above vs. below median)
5. How often the participant washes hands when coming back home (above vs. below median)
6. Whether the participant has children under 18