This project presents our statistical analysis of factors influencing low birth weight among newborns, conducted as part of an applied biostatistics exercise.
We used Generalized Linear Models (GLM) with a logistic regression approach to study how demographic, behavioral, and medical characteristics of mothers affect the likelihood of low birth weight.
We began with an exploratory analysis of the dataset, examining relationships between variables such as maternal age, race, smoking status, and socioeconomic background.
Following this stage, we fitted several logistic regression models to identify the most significant predictors of low birth weight and compared them based on fit statistics and interpretability.
Our findings indicated that race, smoking status during pregnancy, and maternal age were among the strongest predictors.
We also performed residual analysis and model diagnostics to evaluate the model’s validity and detect potential outliers or influential points.
GLM-Low-Birth-Weight-Analysis.Rmd– full reproducible analysis in RGLM-Low-Birth-Weight-Analysis.pdf– complete written report including explanations, tables, and visualizations
Created by Eden Malka and Itay Ben Avraham – 2025
© 2025 Eden Malka & Itay Ben Avraham. All rights reserved.