Skip to main content

Table 3 A/B Testing—change in clicks and engagements—3 out of 100 as the basis

From: Behaviorally informed digital campaigns and their association with social media engagement and COVID-19 vaccine uptake in Belize

 

Clicks

Engagements

(i)

(ii)

(iii)

(iv)

When vaccinated

Few persons

7.616*

7.819**

6.963

7.430

(4.441)

(3.813)

(5.627)

(4.851)

Majority

9.132**

7.454*

8.643

6.563

(4.472)

(4.087)

(5.474)

(5.021)

Reach

0.008***

0.008***

0.009***

0.009***

(0.001)

(0.001)

(0.001)

(0.001)

Controls

NO

YES

NO

YES

District FE

NO

YES

NO

YES

Observations

372

372

372

372

R2

0.754

0.815

0.742

0.806

Adjusted R2

0.752

0.808

0.740

0.799

Mean

58.393

69.880

68.857

83.614

SD

  1. *p < 0.1, **p < 0.05, ***p < 0.01. OLS estimations with robust standard errors in parentheses. Columns (i) and (iii) represent our reduced form specifications. In columns (ii) and (iv), we also included a set of controls and district fixed effects. In (i) and (ii), the outcome variable of interest is the absolute number of Clicks. In (iii) and (iv), the outcome variable of interest is the absolute number of Engagements. All the coefficients associated with the alternatives of the “When vaccinated” Campaign are to be interpreted with the “3 out of a 100” alternative as the comparison point. Controls included the absolute number of Reaches, an indicator variable that takes the number 1 if the campaign was run in English, the number of new COVID cases, the number of new COVID-related deaths, and the average number of days each campaign ran in each location