Basic Sociodemographic Factors for Depression Prediction in Mexico (ENSANUT 2021-2023)

This dataset is derived from the Mexican National Health and Nutrition Survey (ENSANUT 2021–2023). Only a subset of ENSANUT variables is included in this dataset. The full survey data are publicly available at: https://ensanut.insp.mx/ The dataset was constructed to support the analysis and prediction of clinically significant depressive symptoms (CSDS) using sociodemographic variables. It includes data from 32,085 participants aged 20 years and older. Six sociodemographic variables were selected: age, sex, marital status, education level, area of residence (rural/urban), and employment status. One-hot encoding was applied to categorical variables with multiple levels, resulting in a total of 16 input features. Depressive symptoms were measured using the CESD-7 (Center for Epidemiologic Studies Depression Scale). The target variable is binary and indicated the presence of clinically significant depressive syptoms (CSDS). The classififcation criteria are: CESD7 score >= 9 for adults aged 20-59. CESD7 score >= 5 for adults aged 60 and older Instances meeting these thresholds were labeled as positive cases. The dataset contains 6,063 positive cases.