jvpoulos / ok-lottery

Code and data for the paper "Gender gaps in wealth and entrepreneurship? Evidence from 1901 Oklahoma land lottery winners" https://arxiv.org/abs/2206.14922

ok-lottery

Code and data for the paper "Gender gaps in wealth and entrepreneurship? Evidence from 1901 Oklahoma land lottery winners"

Please cite the paper if you use this code for academic research:

@article{poulos2023gender,
  title={Gender Gaps in Frontier Entrepreneurship? Evidence from 1901 Oklahoma Land Lottery Winners},
  author={Poulos, Jason},
  journal={Journal of Historical Political Economy},
  volume={2},
  number={4},
  pages={611--634},
  year={2023},
  publisher={Now Publishers, Inc.}
}

Prerequsites

• R >= 3.5.0 (tested on 4.0.2)

Data files

The main data files used for the study are listed below.

• Lawton land lottery records: data/lawton-A.csv and data/lawton-B.csv
• El Reno land lottery records: data/el-reno-homesteader.csv
• Land patents data: data/patents.csv.zip
• 1900 Census sample
• 1910 Census sample

Instructions

1. cd ok-lottery, and download Census samples into data folder: wget -0 data/census-1900-100-sample.csv.zip "https://www.dropbox.com/s/aq7igj6sudaf73o/census-1900-100-sample.csv.zip?dl=1" and wget -0 data/census-1910-100-sample.csv.zip "https://www.dropbox.com/s/tj3pkzjzjx3hqp8/census-1910-100-sample.csv.zip?dl=1"

2. Open code/main.R and ensure libraries are installed. By default, links participants to census and land patent records.

3. chmod +x code/main.sh and run from the command line. code/main.R prepares data and outputs estimates and descriptive plots. By default, it uses trained models for record linkage.