Socioeconomic disparities in mobility behavior and policy response during the COVID-19 pandemic in developing countries

<b>README</b>The SupplementaryInformation.pdf is licensed under Creative Commons Attribution 4.0 International License. Code and Data are licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.This dataset contains aggregated mobility data from 6 different middle-income countries capturing mobility behaviour across different socio-economic groups over time, from January 2020 until the end of 2020. Several mobility metrics are provided, informing different dimensions of the mobility behaviour of individuals. Data were processed from raw GPS data, provided by the Veraset company.Additionally, code to process and plot the aggregated data is provided.<br>Processing includes:1. time-series construction2. analysis of the association between containment policy enactment and behavioural response of the different socio-economic groups3. GPS data preprocessing and individual association with demographic information<br>More specifically, in the compressed file you will find:1. "<i>create_ts.py" </i>and<i> "aux.py"</i>: process data within the "data" directory to aggregate mobility patterns into time-series<i>2. within the "policy_analysis_code":</i> code that processes time series and uses policy enactment data (contained within the "data" directory) and analyses, with a panel regression model, the association between different socio-economic group behaviour and their changes in mobility behaviour.<i>3. "compute_stop_locations"</i>: is a directory containing Databricks notebook used to process and aggregate the individual data, in the form of raw GPS points, into stop locations (with duration information) and associate them to administrative units and local demographical data.