chrisimpacts/Rwanda-CwD-Analysis

Analysis of anonymised & de-identified data for children with disabilities and developmental delays in Rwanda # Rwanda-CwD-Analysis Analysis of anonymised & de-identified data for children with disabilities and developmental delays in Rwanda ## Data Sources Data was collected in Rwanda with two sampling methods: 1. 3 cells (3c): All households in 3 cells in 3 districts were surveyed. 2. 22 districts (22d): Parents in 60 sectors were encouraged to self-identify Two survey tools were used: 1. For under-3s: A newly designed tool asking about family, birth, child health and social risk factors which could be related to developmental delay 2. For 3-8 year olds: Washington Group Child Functioning Module Questionnaire was administered Data was de-identified by removing all names, phone numbers, names of schools, open-input columns and village identifiers. ## Description This dataset allows us to examine a key question in the detection of disability and developmental delays: what are the most important factors that could be used to efficiently identify and support households of children with disabilities or developmental delays. There are 2 steps to this process: 1. Data Wrangling: Recoding questionnaire responses, using WG response logic to calculate prevalance 2. Data Analysis: Summary statistics, Visualisations, Regression & machine learning analysis ##### Set up 1. Clone the repository to your local machine 2. Install the reuired python packages with pip 3. Run the two notebooks #### Rwanda-Mapping-Data-Wrangling: The data wrangling process takes the raw data and produces a CSV file of cleaned and processed data. The following tasks are performed: * clean column names * concatenate data from all surveys and samples * recode variables Details For the developmental delay risk survey, the following tasks are performed: * recode multiple choice options into a boolean, 1 if the response choice presents a risk to child development and 0 if not. * group risk variables For the Washington Group Child Functioning u5 & u9 surveys, the following tasks are performed: * recode mu …