# Vaccination experiment: teaching data

Source: Campos-Mercade et al. (2021), *Monetary incentives increase COVID-19 vaccinations*, Science 374:879–882. Original replication archive: https://doi.org/10.5281/zenodo.5529626 (version 1).

This derivative uses `data_analysis.dta`. It retains the 8,286 main-analysis observations and 25 variables needed for the week 1 exercise. The authors' sample restriction is `already_vacced == 0` and observed `intention1`; all rows in this source file meet it. Other upstream exclusions are those in the authors' analysis file, not reconstructed from raw survey records here.

Changes: both `vaccinated` and `intention1` are divided by 100, so their values are 0/1. The precomputed `nudge` variable is omitted. Unneeded variables are omitted. All retained assignment and covariate values are unchanged. No observations are added or imputed.

The main `vaccinated` outcome is vaccination within 30 days after survey completion. `intention1` records intention to vaccinate within 30 days of availability. See supplementary Section 1.2.2 for the distinction and its relation to the plan.

## Variables

| Variable | Meaning / coding |
|:---|:---|
| `treatment` | Arm: 0 incentives; 1 social impact; 2 argument; 3 information; 4 no reminders; 5 control. Do not treat this code as a quantitative regressor. |
| `treat_pay` | Incentives Cond.; 1 yes, 0 no. |
| `treat_soc` | Social Impact Cond.; 1 yes, 0 no. |
| `treat_arg` | Argument Cond.; 1 yes, 0 no. |
| `treat_info` | Information Cond.; 1 yes, 0 no. |
| `treat_min` | No Reminders Cond.; 1 yes, 0 no. |
| `control` | Control Cond.; 1 yes, 0 no. |
| `vaccinated` | Vaccinated within 30 days of survey completion: 1 yes, 0 no. |
| `intention1` | Intends to vaccinate within 30 days of availability: 1 yes, 0 no. Measured after assignment. |
| `age` | Age |
| `female` | Female; 1 yes, 0 no. |
| `single` | Single; 1 yes, 0 no. |
| `haschildren` | Has Children; 1 yes, 0 no. |
| `university` | College Education; 1 yes, 0 no. |
| `income` | Monthly income represented by the authors' income-bin values in SEK. Divide by 10,000 for S3; use C(income) for S5. |
| `foreign` | Immigration Background; 1 yes, 0 no. |
| `unemployed` | Unemployed; 1 yes, 0 no. |
| `age_band` | Authors' age-group code for the main specification (five-year bands); use categorically. |
| `region` | Authors' region code; use categorically. It is not the broader regional classification in S4. |
| `covid1_2` | In a risk group for COVID-19; 1 yes, 0 no. |
| `civilstatus` | Civil-status category; use C(civilstatus). |
| `education` | Education category; use C(education). |
| `occupation` | Employment-status category; use C(occupation). |
| `mother` | Mother's place-of-birth category; use C(mother). |
| `father` | Father's place-of-birth category; use C(father). |

## Category labels from the source

These are the authors' labels, retained here for documentation. Region and age-band value labels were not supplied in the source.

### civilstatus

1: Single; 2: Sarbo; 3: Couple; 4: Married; 5: Other Civil Status.

### education

1: Elementary School or Lower; 2: High-school; 3: Professional Training; 4: In College; 5: College Degree; 6: PhD.

### occupation

1: Employed; 2: Unemployed; 3: In College; 4: Retired; 5: Other Professional Situation.

### mother

1: Mother from Sweden; 6: Mother from Rest of Europe; 7: Mother from North America; 8: Mother from South America; 9: Mother from Africa; 10: Mother from the Middle-east; 11: Mother from the Rest of Asia; 12: Mother from Oceania.

### father

1: Father from Sweden; 6: Father from Rest of Europe; 7: Father from North America; 8: Father from South America; 9: Father from Africa; 10: Father from the Middle-east; 11: Father from the Rest of Asia; 12: Father from Oceania.

### income

2500: Income 0-5000kr; 7500: Income 5001-10000kr; 12500: Income 10001-15000kr; 17500: Income 15001-20000kr; 22500: Income 20001-25000kr; 27500: Income 25001-30000kr; 32500: Income 30001-35000kr; 37500: Income 35001-40000kr; 42500: Income 35001-40000kr; 47500: Income 45001-50000kr; 52500: Income 50000kr-55000kr; 60000: Income more than 55000kr.

The source repeats the 35,001–40,000 label for the 42,500 income code; the numeric codes are preserved. This does not affect the exercise's categorical income controls.