Navigating the World of Secondary Data: Hands-on Experience

 

Global Research & Training

New Delhi

Contact us: info@grtedu.com                                         Website: www.grtedu.com

 

GRT is strengthening academia, advancing research, and empowering researchers globally for a better tomorrow. 

Navigating the World of Secondary Data: Hands-on Experience

Introduction to Data

Ø  Generally, data refers to raw facts, figures, and statistics collected via observation, survey or interview for reference or analysis.

Ø  Some time, some people understand 'data' means 'number,' but in research, data is much more than numbers.

Ø  It can take many forms, including numbers, words, images, audio, and video.

Examples: Survey responses, transaction records, GDP of the country, temperature at a specific time and place, weight of individuals, population of the country, map of any place, interview transcript…etc.  

Some formal definitions include:

• The United Nations defines data as “characteristics or information, usually numerical, that are collected through observation.”

• Burns and Bush (2010): Secondary data is data that was originally collected for a different research question or objective but can be used for new analysis.

• Babbie (2013): Secondary data are "Data collected by someone other than the user.“

• Creswell (2014): Secondary data are data that were collected previously by other researchers or organizations for different research objectives.

Types of Data: Based on Collection Method

When we talk about data collection, there are two broad categories: primary data and secondary data.

Primary Data:

Collected directly by the researcher for a specific purpose.

Examples: Surveys you design yourself, interviews you conduct, field observations... etc. 

Pros: Tailored to your needs; you control the quality.

Cons: Time-consuming, costly, and resource-intensive.

 

Secondary Data

Collected by someone else for a different purpose but available for other use by other researchers.

Examples: Census of India, World Bank Indicators, NSSO surveys, WHO health statistics... etc. 

Pros: Saves time and cost; often covers large populations over many years.

Cons: May not match your exact needs; limited control over how it was collected.

Introduction to Secondary Data

Secondary data refers to information that was collected by someone else for a different purpose, but now it will be used by another researcher.

In other words, researchers do not collect it directly, but he or she uses it for their research purpose.

Example: Number of students in a class (e.g., 60), temperature readings (e.g., 28°C, 30°C), survey responses (e.g., yes/no, male/female)

Sources of secondary data 

Government Data: Data collected from government reports, surveys, or census data. (like MoSPI, RBI, NCRB, NFHS, AISHE, UDISE+ ... etc.)  

Academic Research: Published/unpublished articles, conference papers, books, theses, dissertations, reports from organizations… etc.

Industry Reports: Market research, financial statements, company profiles… etc.

Media Reports: News articles, blogs, social media posts... etc.

Databases: Pre-existing databases available on data-sharing platforms or government websites. (Like the RBI database, the MoSPI database, the UN database, the World Bank’s database, the university or institute database… etc.)

Unit-Level Data vs. Aggregate Data

Unit-Level data:

Contains detailed, disaggregated information at the establishment or unit level.

Data at the firm or household level, e.g., fixed capital, working capital, output, employment… etc. at a factory level.

 

Aggregate data:

Summarized data across establishments, e.g., total employment in a region or sector.

For example, aggregate data might show the total number of workers in the manufacturing sector in a given region without showing data for individual factories.

 

Types of data collection:

Census Data

Data collected from every unit in a population (e.g., all households in a country).

Sample Data

Data collected from a sample of the population, often with a specialized focus (e.g., health surveys, labor force surveys).

Administrative Data

Information gathered through government records and databases (e.g., tax records, school enrollments).

Type of Data: Based on Nature: 

Quantitative Data

Numerical in nature; can be measured and analyzed statistically.

Examples: GDP growth rate, literacy rate, rainfall in mm... etc. 

Qualitative Data

Descriptive, categorical, or non-numerical information.

Examples: gender, occupation type, political affiliation... etc. 

Often coded into numbers for analysis.

Data Types: Based on Time and Units

1.     Time-Series Data:

Data collected and recorded over a specific period at regular intervals (e.g., annually, quarterly, monthly), decade.

Example: Annual GDP data for a country over several years, population census over decades

 

2.     Cross-sectional Data:

Data collected at a single point in time across multiple entities (population, literacy rate, unemployment rate, etc.).

Example: Education levels across different regions in 2024, HDI value across Asian counties in 2024.

 

3.     Panel Data (Longitudinal Data)

A combination of time series and cross-sectional data, where data is collected for the same entities (individuals, regions, countries, etc.) over multiple time periods.

Example:  HDI values for India, Pakistan, Bangladesh, Nepal, and Sri Lanka from 2007–2022.

Comparison of Types
Feature	Time Series	Cross-Sectional	Panel Data
Dimension	Single entity, multiple times	Multiple entities, single time	Multiple entities, multiple times

Two categories of panel data

1. Balanced Panel Data

A panel dataset is called balanced when each cross-sectional unit is observed for the same number of time periods.

Example: 

Firm	2021	2022	2023	2024
Firm A	✓	✓	✓	✓
Firm B	✓	✓	✓	✓
Firm C	✓	✓	✓	✓

 

2. Unbalanced Panel Data

A panel dataset is called unbalanced when each cross-sectional unit is not observed for the same number of time periods.

 Example: 

Firm	2021	2022	2023	2024
Firm A	✓	✓	✓	✓
Firm B	✓	✓	—	✓
Firm C	✓	—	—	✓

 

Types of Data: Based on Structure

1. Structured Data

Data organized in a predefined format, usually in rows and columns.

Each variable has a clearly defined field or column.

Easy to store, search, manage, and analyse using statistical software.

 Examples:

Survey datasets, census data, Excel/CSV files, administrative records, database tables... etc.

 

2. Unstructured Data

Data that does not follow a predefined tabular structure.

Information may exist in different formats and requires additional processing before analysis.

Often requires qualitative analysis, text mining, image processing, or other specialized techniques.

Examples: Interview transcripts, audio recordings, video recordings, photographs & images, social media posts, emails, newspaper articles, text documents... etc. 

 

Advantages of Secondary Data

1. Cost-Effective:

Secondary data is cost-saving and often freely available or comes at a low cost. It is more affordable option, especially for researchers working with limited budgets.

2. Time-Saving:

Secondary data is already available and can be used immediately, which can save substantial time.

3. Large Scale and Comprehensive:

Secondary data often provides access to large datasets, offering broad coverage across multiple regions, time periods, or demographic groups.

4. Cross-Disciplinary Research:

Secondary data can be used across different research disciplines.

 

Limitations of Secondary Data

May not exactly match your research question.

Possible issues with outdated data.

Quality depends on original collection methods.

Limited variables

Sometimes incomplete or missing variables you need.

How to Choose the Right Dataset

Relevance: Does it match your topic, geography, and time period? What variables and indicators are available?

Coverage: Is the population/sample adequate for your analysis? And Who is the population represented?

Accuracy & Credibility: Who collected the data? Was it collected by a trusted source? Why were the data collected?

Level of detail: Unit-level vs. aggregated—which do you need?

Format & Accessibility: Can you easily open and process it?

Licensing: Are there usage restrictions?

 

The Database Discovery Process: Research Question to Data

Example: How has women’s labour force participation changed across Indian states?

What do we need?

Concept: labour force participation

Population: working-age population

Unit: individual/person

Variables: sex, age, labour-force status, state, education, sector, weights

Potential source: PLFS

 

Ethical Considerations: 

1. Data Privacy and Confidentiality:

Researchers must ensure that privacy and confidentiality are maintained. The use of such data should comply with ethical standards and data protection regulations (e.g., PDPA (Personal Data Protection Act) and General Data Protection Regulation (GDPR) etc.

2. Acknowledging Sources and Copyright:

Using secondary data ethically means giving proper credit to the original creators or collectors of the data. This shows respect for their work and helps avoid plagiarism.

Indian database for secondary data: Key sources

India offers several open data platforms through its government websites and data portal. There are some institutions and private databases too. These data span multiple sectors, including demographics, economics, health, education, and more. Key sources include the Census of India, MoSPI (ASI, PLFS, NSS, IIP, Economic Census…etc.), NFHS, RBI, AISHE, UDISE+, NDAP (NITI Aayog’s database), Data.Gov etc.

 

International database for secondary data: Key sources

Several international organizations and universities offer open data platforms through its databases or portal.

These data span multiple sectors, including demographics, economics, social, health, education, and more.

Key sources for international secondary data include the United Nations database, World Bank database, International Labour Organization database, World Health Organization database International Monetary Fund database and FAOStat databse ...etc.

Literature Database: Key Sources

Key literature data sources include peer-reviewed journals, books, and conference proceedings, as well as online databases that provide comprehensive access to scholarly material.

These platforms offer diverse, reliable materials essential for robust research.

 All of which serve as rich repositories for academic literature across disciplines. These platforms offer diverse, reliable materials essential for robust research.

Notable databases include

SodhGanga (a reservoir of theses),

SodhGangotri (a repository of synopses),

Google Scholar

JSTOR

Science Direct

ResearchGate

Academia.edu

Pubmed … etc.

 

AI Tools for Literature Discovery

R Discovery

Semantic Scholar

Undermind

Lit map

Elicit

Consensus

Perplexity

ResearchRabbit

Connected Papers

 

Beware of AI Hallucinations

Possible problems: fabricated references, incorrect DOI, wrong year, misrepresented findings

Verify author, title, journal, year, DOI and original paper Check that the paper actually supports the claim

Never cite a paper you have not verified

Golden Rule: Never cite a paper you have not verified.

For more details on the data sources and its links, please visit this article.

https://grtedu.blogspot.com/2025/01/name-of-secondary-data-sources-and-its.html

 

Thank You and Best Wishes

Raghavendra Yadav

Global Research & Training, New Delhi

Email: info@grtedu.com | Web: www.grtedu.com

 

Connect with us on social media:

WhatsApp                                             LinkedIn

X (Twitter)                                            Facebook

Instagram                          Blog               Youtube