ITClaude/TopicTwo

This a results analysis for topic two titled: Impact of Climate Variability on Crop Selection and Rotation Practices in Semi-Arid Zones: Evidence from Integrated Farmer Surveys and AI-Based Panel Data Analysis in Bugesera, Rwanda # Impact of Climate Variability on Crop Selection and Rotation Practices ## Evidence from Bugesera District, Rwanda ## Overview This repository contains the code, dataset, and summary outputs for the study **“Impact of Climate Variability on Crop Selection and Rotation Practices in Semi-Arid Zones.”** It analyses survey data from smallholder farmers in Bugesera District, Rwanda, to examine how climate variability influences crop selection, crop rotation practices, and broader adaptation behaviour. The workflow combines: - data cleaning and preparation; - descriptive and inferential statistical analysis; and - machine learning models for predictive insights. ## Study Objectives This project aims to: - assess how climate variability affects crop selection and crop rotation practices; - identify socioeconomic and environmental factors associated with farmer decision-making; - evaluate patterns linked to agricultural losses and adaptive responses; and - provide a reproducible analytical workflow for climate-smart agriculture research. ## Dataset The analysis is based on cross-sectional field survey data collected from smallholder farmers in Bugesera District. ### Dataset highlights - **Sample size:** 240 valid observations - **Data type:** Cross-sectional survey data - **File:** `data/FINAL Data collected in Bugesera District-Template (1).csv` ### Variables included The dataset contains information on: - household and demographic characteristics; - farm size and input use; - perceived and geospatial climate risk indicators; - agricultural output and loss measures; and - adaptation-related behaviour, including crop rotation and advisory access. ## Analytical Workflow The project is organised into three main stages. ### 1. Data cleaning and preparation Script: `src/01_data_quality_and_cleaning.py` This stage: - removes blank or invalid rows; - standardises text-based entries; - parses numeric values from mixed-format fields; - converts land area and yield fields …