# Replication materials to: The Sibsize Revolution in International Context. Declining Social Disparities in the Number of Siblings in 26 Countries, _Demographic Research_, May 2020

## Files

The replication materials consist of five files:

1) 01_data.do,

2) 02_analyses.do,

3) 03_visualization.R, 

4) monden_et_al_2018.zip, and

5) this "readme" file.

## Data

The data necessary for replicating the analyses is publicly available. Table 1 of the manuscript contains a detailed list of the 111 surveys used. Monden _et al._ (2018) provide a large set of Stata 13 do-files to harmonize the individual data sets into the required format.

## Software versions

The two Stata do-files were written in Stata 13, but final changes were made in Stata 15. The R file was written with version 3.6, but the final version of the figures was created with R 4.0.

## Necessary steps for replication

1) Obtain data sets listed in Table 1 of the manuscript. Web links to the data providers can be found in the reference list of the manuscript. Many of the data sets are publicly available but some require registration or special licenses.

2) Save these data sets in a folder called "raw data".

3) Execute the Stata do-files provided in Monden _et al._ (2018) for each data set in the folder "raw data" to harmonize the survey data sets obtained in Step 1). Each survey corresponds to one do-file which is named after the country and year of data collection. Country codes follow the [ISO 3166-1 alpha-3](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3) convention. As a courtesy, the files from Monden _et al._ (2018) are included in the file monden_et_al_2018.zip.

4) Save the harmonized data sets in a folder on your hard drive.

5) Install the user-written -fastcd- command in Stata (-ssc install fastcd-) and use the -fastcd- command to designate the folder containing the harmonized data sets as "sibsize" (-c cur sibsize-)

6) Execute the 01_data.do file in Stata to combine the data sets into one large combined file.

7) Execute the 02_analyses.do file in Stata to conduct the analyses.

8) Execute the 03_visualization.R in R to create the figures showing the results.

9) Voila.

## Variables

Analyses are based on the following variables:

- country: Country
- birthyr: Birth cohort, yearly
    - bcohort: Birth cohort in 10-year intervals
- nsibs: Sibship size
    - onlyc: Only-child family
    - large: Large family
- father_tertiary: Father has tertiary education
- mother_tertiary: Mother has tertiary education
    - parent_tertiary: At least one of the parents has tertiary education

# Reference

Monden, C., Choi, S., Taiji, R., and Chen, M. (2018). _International Sibsize and Educational Attainment Database (ISEAD)_. Cambridge, MA: Harvard Dataverse. doi:[10.7910/DVN/F9PKG5](http://dx.doi.org/10.7910/DVN/F9PKG5).