Chapter 4: Infrastructure, Governance & Schemes¶

Introduction¶

India's governance datasets represent some of the richest sources of policy-relevant data anywhere in the world. Tens of thousands of government programmes — from rural road construction to school meal distribution to direct cash transfers — generate detailed administrative records that are updated daily, monthly, or annually. These records, when properly accessed and analysed, can reveal where public money is flowing, who is benefiting, and where gaps persist.

Unlike the census or satellite imagery, most datasets in this chapter require no GIS software at all. The primary format is CSV or Excel, downloaded directly from programme portals or data.gov.in. This makes them among the most accessible datasets in this guidebook — a spreadsheet and a basic knowledge of Python (or even Excel) is enough to begin meaningful analysis.

Why Infrastructure Datasets Matter

These datasets sit at the heart of development economics and policy evaluation research in India. Studies on the impact of rural roads (PMGSY) on agricultural wages, or MGNREGA on out-migration, or mid-day meals on school enrolment — all depend on the administrative data described in this chapter. Learning to access and analyse these datasets opens up a rich terrain of research questions.

This chapter covers six flagship government data systems spanning rural employment, education, health, roads, composite district analytics, and direct cash transfers. Together they offer a near-complete picture of the social infrastructure of rural India.

Datasets in This Chapter¶

Dataset	Provider	Website	Access	Format	Best For
PMGSY	MoRD / NIC	omms.nic.in	🟢 No Login	Reports, GIS	Rural road connectivity, habitation coverage
UDISE+	Ministry of Education	udiseplus.gov.in	🟡 Login (school-level)	CSV, Excel	School enrollment, infrastructure, PTR
HMIS	MoHFW	hmis.nhp.gov.in	🟢 No Login	CSV (monthly)	Health facility data, immunisation, deliveries
MGNREGA	MoRD	nregs.nic.in	🟢 No Login	CSV, Excel	Rural employment, wage payments, works data
NDAP	NITI Aayog	ndap.niti.gov.in	🟡 Login to download	CSV, API	Composite district indicators, SDG scores
DBT	DBT Mission	dbtbharat.gov.in	🟢 No Login	Dashboard, CSV	Benefit transfer tracking, scheme coverage

Key Themes¶

1. Following the Money¶

All six datasets are fundamentally about public expenditure and programme delivery. Together they answer the question: is the government reaching the people it intends to reach?

MGNREGA tracks whether poor rural households are getting the 100 days of work they are entitled to, and whether wages are being paid on time.
PMGSY tracks which villages have been connected to all-weather roads and how much has been spent.
HMIS tracks whether health services are actually being delivered at primary health centres every month.
UDISE+ tracks whether schools have the basic infrastructure that pupils need to learn.
DBT tracks whether cash and benefits are reaching beneficiaries' bank accounts directly.
NDAP provides composite indicators to compare districts on multiple development dimensions simultaneously.

2. Linking Datasets for Research¶

The real power of these datasets emerges when they are cross-linked with each other and with census data:

MGNREGA + Census → Do the poorest districts (by poverty headcount) get proportionally more MGNREGA work?
PMGSY + HMIS → Does road connectivity improve health facility utilisation?
UDISE+ + HMIS → Are districts with better school infrastructure also performing better on child immunisation?
NDAP + All others → NDAP already does this aggregation for you at the district level.

These linkages are the foundation of policy evaluation research. Chapter 8 (Combining Datasets) covers spatial joins and data merging in detail.

3. Time Series Potential¶

Unlike the census (which is a decennial snapshot), several datasets here are updated monthly:

Dataset	Update Frequency	Time Series Depth
HMIS	Monthly	April 2008 onwards
MGNREGA	Daily (monthly summary available)	2006-07 onwards
NDAP	Annual (mostly)	Varies by indicator
PMGSY	As projects update	2000 onwards
UDISE+	Annual	2012-13 onwards
DBT	Real-time dashboard	2013 onwards

This time depth makes it possible to study the impact of policy changes, identify seasonal patterns, and assess COVID-19's disruption to public services.

4. Administrative Hierarchy¶

Nearly all these datasets follow India's administrative hierarchy:

National
  └── State (36 states/UTs)
        └── District (~640)
              └── Block / Taluka (~5,500)
                    └── Gram Panchayat (~250,000)
                          └── Village (~600,000)

MGNREGA and PMGSY data go down to Gram Panchayat level. HMIS and UDISE+ go down to facility/school level. NDAP standardises to district level.

Skills Required¶

Dataset	Minimum Skill	Ideal Skill
MGNREGA	Spreadsheet (Excel)	Python (pandas)
UDISE+	Spreadsheet	Python (pandas + matplotlib)
HMIS	Spreadsheet	Python (time series, matplotlib)
PMGSY (reports)	Browser	Spreadsheet
PMGSY (GIS data)	QGIS basics	Python (geopandas)
NDAP	Browser	Python (requests + pandas)
DBT	Browser	Spreadsheet

Start Here

If you are new to government datasets, start with MGNREGA or DBT. Both have clean, browsable dashboards and no login requirement. You can get meaningful numbers within 10 minutes.

What You'll Learn in This Chapter¶

By the end of Chapter 4, you will be able to:

✅ Navigate major government MIS portals (nregs.nic.in, hmis.nhp.gov.in, udiseplus.gov.in, omms.nic.in)
✅ Download CSV/Excel data for your state, district, or block
✅ Load and clean government CSV files in Python with pandas
✅ Calculate key indicators: Pupil-Teacher Ratio, immunisation coverage, MGNREGA wage expenditure
✅ Create time-series plots for monthly health or employment data
✅ Use NDAP API for programmatic district-level data access
✅ Cross-link two or more datasets to generate research insights
✅ Identify data quality issues and handle them appropriately

Chapter Exercise 4.0¶

Chapter Exercise 4.0 — Your First Policy Dataset

This exercise introduces you to the MGNREGA dataset — the most granular, high-frequency welfare dataset available for rural India. Complete the following steps before moving to individual dataset pages.

Objective: Download MGNREGA work data for your block and identify the top 5 Gram Panchayats by expenditure.

Steps:

Go to nregs.nic.in and click "Reports" in the top navigation.
Navigate to: R14 → Financial Progress report.
Select your State → District → Block and the current financial year.
The table will show all Gram Panchayats in your block with funds released, expenditure, and works completed.
Click "Download" or use browser Ctrl+S to save the page as CSV (or copy the table to Excel).
Open the file in Excel or Python. Sort by "Amount Spent" in descending order.
Note the top 5 GPs — what types of works (water conservation? roads? individual assets?) are listed?

Questions to answer: - Which GP had the highest MGNREGA expenditure in your block this year? - What % of the total funds were spent (utilisation rate)? - Are SC/ST worker days proportional to their population share?

Extension: Compare the current year data with the previous year. Did expenditure increase or decrease? What might explain this?

Dataset Pages¶

PMGSY — Rural Road Data

Habitation connectivity, road lengths, and project expenditure from the Pradhan Mantri Gram Sadak Yojana portal.
UDISE+ — School Data

Enrollment, teacher counts, infrastructure, and dropout rates for every school in India.
HMIS — Health Facility Data

Monthly OPD, deliveries, immunisation, and disease data from primary health centres across India.
MGNREGA — Rural Employment

Household employment demand and provision, person-days, wages paid, and works data to GP level.
NDAP — National Data Platform

Curated, cleaned composite indicators from 70+ government sources, all standardised to district level.
DBT — Direct Benefit Transfer

Scheme-wise beneficiary counts and amounts transferred across 300+ central government schemes.

Data Access Badge Legend¶

Badge	Meaning
🟢 No Login	Data freely downloadable without any account
🟡 Login	Free registration required to access full data
🔵 GIS/Coding	Requires GIS software or Python to use effectively
🔴 Advanced	API access, spatial analysis, or special approval needed

A Note on Data Quality¶

Government administrative datasets in India vary considerably in quality. Some general guidance:

Data Quality Considerations

HMIS data is self-reported by health facilities. Coverage and accuracy vary by state — Kerala and Tamil Nadu tend to have better reporting than some northern states.
UDISE+ is filled in by schools themselves. Infrastructure indicators may be optimistically reported.
MGNREGA data is transactional (wage payments are computer-generated), so financial data is generally reliable. Worker counts can sometimes be manipulated at local level.
NDAP data is cleaned but may lag by 1-2 years in some indicators.
Always compare critical findings against a second source before drawing firm conclusions.

Cross-Validation Strategy

Cross-validate administrative data with survey data (NFHS, ASER, NSS) where possible. Large discrepancies between administrative figures and survey figures are themselves a finding worth reporting.