Dataset and R Script: PCA and Firth Regression Analysis in Elderly Quality of Life (UnATI-UFAL Program)

This repository contains anonymized data and analysis scripts from a study investigating quality of life (QoL) in older adults participating in the UnATI-UFAL program (Maceió, Brazil). The dataset includes demographic (e.g., age, gender), clinical (e.g., diabetes, osteoporosis), and psychosocial variables (e.g., functional capacity scores) from 23 elderly participants. The analysis employs Principal Component Analysis (PCA) to identify latent patterns in multidimensional QoL data and Firth Regression to address small sample bias in logistic models.

Key Features:

Data: Anonymized CSV file with 23 rows and 24 variables (e.g., sexo, idade, diabetico, osteoporose, escolaridade, functional scores P_1_DF to P_8_MA).

Code: R script for reproducibility (data preprocessing, KMO test, PCA, Firth regression, visualizations).

Application: Demonstrates advanced statistical methods for gerontological studies in resource-limited settings.

Methodological Relevance:

Addresses challenges of small sample sizes (n=23) using Firth regression.

Provides a template for dimensionality reduction (PCA) in aging-related multidimensional data.

Practical Impact:Supports public health policies targeting active aging in socioeconomically vulnerable regions, such as Northeast Brazil.

Keywords:Aging, Quality of Life, Principal Component Analysis, Firth Regression, Gerontology, Brazil, UnATI-UFAL.