Using quantitative research methods to analyze domestic violence trends in developing countries against women and girls.
Quant Research
UWM Coursework
Dec 2022
As a daughter of a single mother, I learned the importance of women standing up for themselves at a very young age. My mother would tell me that a woman will only lose control of her life if she willingly gives it up. But if a woman decides not to do that then no power in the world can control her. She made sure that her two daughters were well educated and financially independent so that we are always in control of our lives. Being brought up with that mindset Women Empowerment is something very close to my heart and so I try to advocate this in whatever opportunity I get. Even through a Data Science project for Grad school.
🔸 While browsing through thousands of datasets on the internet, I finally came across a dataset of my interest. I sourced my dataset from Kaggle.
🔸This dataset contains data from 70 different developing countries. I decided to focus on developing countries because I feel that these regions require more focus and attention currently.
🔸 The data was collected as part of a DHS survey. “DHS surveys are nationally-representative household surveys that provide data for a wide range of monitoring and impact evaluation indicators in the areas of population, health and nutrition. Standard DHS Surveys have large sample sizes (usually between 5,000 and 30,000 households) and typically are conducted around every 5 years, to allow comparisons over time.”
🔸 I also did not want to mix data from developed nations into this analysis because I feel that the conditions and the psychology behind domestic violence in these countries would differ vastly and it should be dealt with as a separate topic for analysis.
The data collected from female and male correspondents was analyzed and plotted separately in hopes to understand what reasons of domestic violence women relate to the most and what reasons men feel are justified for domestic violence against women.
🔸 Pandas was primarily used here for analyzing most of the data. Fancier libraries like Plotly were used to map geospatial data towards the end of our analysis
🔸 The whole analysis was done using Google Collaboratory
🔸 The full code for the analysis can be found here
🔸 While analyzing the domestic violence trend country wise, we ended up focusing on extreme cases like War/Conflict ridden countries, unstable economies etc. which could seem to be obvious cases. Instead we should have focused on dissecting and analyzing average cases and the reasons for domestic violence in those countries. A war/Conflict ridden country will be bound to have inadequate living conditions with negligible Human Right Laws. The true reasons for violence against women will show up if we had focused more on stable peaceful developing countries. We try to tackle this issue towards the end of our paper but the analysis is still a bit lacking there.
🔸 There are some scenarios we wanted to analyze but couldn't due to the format of our dataset.
