Demographic and health survey: NFHS India Data Analysis in R

Name: Demographic and health survey: NFHS India Data Analysis in R
Rating: 4.8 (32 reviews)

R & RStudio for NFHS-5 Data Analysis | Survey Weights, gtsummary, Regression & Public Health Research

Created byMd Ahshanul Haque

Last updated 3/2026

English

English [Auto],

What you'll learn

Analyze National Family Health Survey (NFHS) data using R and RStudio for public health research
Clean, recode, and label NFHS survey variables using reproducible R workflows
Multilevel models in R using NFHS data
Apply survey weights, strata, and primary sampling units for correct NFHS analysis in R

Course content

7 sections • 29 lectures • 1h 52m total length

Overview of the National Family Health Survey2:36
Introduces the National Family Health Survey, its purpose, scope, and role in generating nationally representative health data.
Demographic and Health Surveys (DHS) Program: Data Download | Stata Format2:09
Learn how to access the DHS Program website, request permission, and download NFHS-5 datasets for research and analysis.
How to get the R code0:28
During the video, you will find the R code file available in the resources section of the corresponding lecture.
Checking the Dataset and Codebook in the Project Folder2:03
In this lecture, you will learn how to systematically check your dataset and codebook before starting any analysis for Poisson regression in cohort studies. Reviewing the dataset structure and accompanying codebook is a critical step to ensure data accuracy, correct variable interpretation, and reproducible research.

Import NFHS Data into RStudio | Stata format data to R Programming and RStudio3:15
Learn how to import NFHS survey datasets into RStudio from Stata formats and check that the data are ready for analysis.
Selecting Variables Based on the Study Objective8:39
Learn how to create a clear codebook for NFHS data by defining variable names, labels, value labels, and missing codes using Excel.
Viewing and Verifying Labels of Factor and Categorical Variables2:57
Learn how to apply meaningful value labels to categorical NFHS variables such as education, wealth index, and residence using R.
Short Descriptive Summary Using gtsummary (tbl_summary)2:26
In this lecture, you will learn how to create a short, clear descriptive summary table using the gtsummary package in R. This table provides a quick overview of cohort study variables and is commonly used as a baseline summary before Poisson regression analysis.
Identifying Extreme and / Values in NFHS Data Using RStudio3:15
Learn how to detect extreme and flag values in NFHS survey data using summary statistics and visual tools in R.
Replacing Unusual Values and Setting Decimal Points for Analysis in R9:29
In this lecture, you will learn how to identify and replace unusual or invalid values in cohort study data and how to set appropriate decimal points for numeric variables before Poisson regression analysis. Cleaning extreme and miscoded values is a crucial step to ensure valid incidence rate estimation and stable model results.
Generating Nutrition Indicator in R | stunting, wasting, and underweight7:32
In this lecture, you will learn how to generate standard child nutrition indicators—stunting, wasting, and underweight—using R. These indicators are fundamental outcomes in nutrition, epidemiology, and public health research, and are widely used in DHS, NFHS, and WHO-based analyses.
Applying Variable Labels to NFHS Survey Data in R2:49
Learn how to assign clear and meaningful variable labels to NFHS survey variables in R to improve data readability, interpretation, and reporting.
Renaming NFHS Variables for Clear and Reproducible Analysis in R5:38
Learn how to rename NFHS variables using R to create clear, consistent, and reproducible variable names for analysis and reporting.
Finalizing the Dataset and Saving in RData Format3:31
In this lecture, you will learn how to finalize your cleaned and prepared cohort dataset and save it in RData format for efficient reuse, reproducible analysis, and future modeling. Saving data in RData format preserves variable types, factor levels, and labels, which is especially important for Poisson regression and publication-ready outputs.

Summarising All NFHS Variables Using gtsummary in R4:31
Learn how to summarise all NFHS variables using the gtsummary package in R to produce clear, publication-ready descriptive tables.
Bivariate Analysis of NFHS Variables Using gtsummary in R6:05
Learn how to perform bivariate analysis of NFHS data by comparing outcomes across groups using gtsummary tables and appropriate statistical tests in R.
State-Specific NFHS Analysis Using gtsummary in R: West Bengal3:21
Learn how to perform state-specific analysis of NFHS data for West Bengal using the gtsummary package in R to generate clear, publication-ready tables.

When and Why to Adjust Sampling Weights in NFHS Analysis Using R1:43
Learn when sampling weights should be applied in NFHS data analysis, why they matter, and how weighting affects estimates and inference in R.
Univariate Analysis of NFHS Variables Using R | Adjust Sampling Weights | svy7:23
Summarise individual NFHS variables using suitable descriptive measures to understand data distribution and key characteristics.
Bivariate Analysis of NFHS Variables with P-Values Using R | Sampling Weights2:20
Learn how to conduct bivariate analysis of NFHS data by comparing outcomes across groups and calculating appropriate p-values using R.
Fixing Decimal Places in NFHS Tables and Results Using R2:43
Calculating P-Values with Cluster, Strata, and Weights in NFHS Using R4:28
Learn how to correctly calculate p-values for NFHS data by accounting for clustering, stratification, and sampling weights using survey methods in R.

Simple Survey-Weighted Logistic Regression (svyglm) for NFHS Data5:08
Learn how to fit a simple (unadjusted) survey-weighted logistic regression model using svyglm() while accounting for clusters, strata, and sampling weights in NFHS data.
Multiple Survey-Weighted Logistic Regression (svyglm) for NFHS Data1:49
Learn how to fit multiple survey-weighted logistic regression models using svyglm() and interpret adjusted odds ratios while accounting for clusters, strata, and sampling weights.
Changing Reference Categories in Logistic Regression in Rstudio2:20
Learn how to change reference categories for categorical variables and correctly interpret odds ratios in NFHS logistic regression models using R.

Simple Multilevel Logistic Regression Analysis for NFHS Data in R6:01
Fit and interpret a simple multilevel logistic regression model with random intercepts to account for clustering in NFHS data using R.
Multiple Multilevel Logistic Regression Analysis for NFHS Data in R3:11
Apply multiple multilevel logistic regression models to NFHS data to estimate adjusted effects while accounting for hierarchical structure in R.

Requirements

Basic knowledge of R programming and familiarity with RStudio
Introductory understanding of statistics or public health concepts is helpful
A computer with R and RStudio installed

Description

This course provides a hands-on, practical guide to analyzing Indian National Family Health Survey (NFHS) data using R and RStudio, following best practices for DHS-style complex survey analysis. It is designed for public health students, researchers, and analysts who want to work confidently with nationally representative survey data.

You will start by understanding the NFHS-5 survey design, its connection with the Demographic and Health Surveys (DHS) Program, and how to access and download NFHS datasets. The course then walks you through data management in R, including variable selection, recoding, labeling, handling missing values, and preparing clean, analysis-ready datasets.

A key focus is on descriptive and bivariate analysis, where you will create publication-ready tables using the gtsummary package. You will learn to generate univariate and bivariate summaries, conduct state-level analysis, and correctly interpret p-values.

The course also emphasizes complex survey methods, explaining when and how to apply sampling weights, clusters, and strata using survey design objects in R. You will produce survey-weighted results with valid statistical inference.

Finally, you will be introduced to multilevel regression modeling to handle hierarchical data structures common in NFHS data.

By the end of this course, you will be able to conduct reproducible, policy-relevant data analysis in R, suitable for theses, dissertations, and public health research.

Who this course is for:

Public health, epidemiology, and biostatistics students who want hands-on experience analyzing NFHS survey data in R
Researchers and analysts working with population-based health survey data who want to apply reproducible R workflows
MPH, MSc, and PhD students seeking practical training in survey-weighted and multilevel analysis using NFHS data

Demographic and health survey: NFHS India Data Analysis in R

What you'll learn

Explore related topics

Course content

Introduction to the National Family Health Survey (NFHS-5) – India (DHS Program)4 lectures • 7min

Data Management of NFHS Survey Data in R I Demographic and Health Surveys10 lectures • 50min

Summary Tables with P-Values in R (Without cluster and sample weight )3 lectures • 14min

Survey-Weighted Summary Tables with P-Values in R (NFHS)5 lectures • 19min

Logistic Regression Analysis in R (Without Survey Design) R programming2 lectures • 4min

Survey-Weighted Logistic Regression Analysis in R (NFHS) svyglm3 lectures • 9min

Multilevel Logistic Regression Analysis for NFHS Data in R2 lectures • 9min

Requirements

Description

Who this course is for: