Sources and Acquisition of Data

Data raw facts, figures, observations, symbols or responses हैं जिनका analysis करके meaningful conclusions निकाले जाते हैं।

Key Point: Data स्वयं information नहीं है; processing के बाद यह information और knowledge में बदलता है।

Data Classification:

• By Source: Primary Data, Secondary Data
• By Nature: Quantitative (numerical), Qualitative (categorical)
• By Collection Method: Census, Sample
• By Time: Cross-sectional (one time), Time-series (over time)

Data Acquisition विभिन्न sources से data को systematically obtain करने की process है।

Steps in Data Acquisition:

• Define Objective: What data is needed and why
• Identify Sources: Primary or secondary
• Choose Collection Method: Survey, observation, experiment, etc.
• Collect Data: Census or sampling
• Validate Data: Check accuracy, completeness, consistency
• Store Data: Organised for analysis
• Process and Analyse: Convert raw data to information

Primary Data researcher द्वारा first-hand किसी specific research purpose के लिए directly collected किया गया data है।

Methods of Collection:

• Observation: Directly observing behaviour/events
• Interview: Face-to-face or telephonic conversation
• Questionnaire/Survey: Structured questions
• Experiment: Controlled research setting
• Focus Group Discussion: Group interaction
• Field Survey: Direct data collection from population

Advantages:

• Purpose-specific and relevant
• Researcher controls methodology
• Current and original
• Detailed information possible

Disadvantages:

• Time-consuming
• Expensive
• Training required
• Non-response and bias possible

Secondary Data वह data है जो already collected किया जा चुका है और researcher उसे अपने study के लिए reuse करता है।

Sources:

• Government Publications: Census, NSSO reports, RBI reports
• Institutional Records: UGC/AISHE reports, university records
• Research Publications: Journals, books, theses
• Databases: Shodhganga, JSTOR, Scopus, Google Scholar
• Organisational Data: Company reports, NGO surveys
• Historical Records: Archives, government documents

Advantages:

• Economical and time-saving
• Large geographical/historical coverage
• Comparative and trend analysis possible

Disadvantages:

• May be outdated
• Definitions may differ
• Quality/accuracy questionable
• May not fit research objectives perfectly

Exam Trap: Data का primary/secondary distinction उसके स्वरूप पर नहीं, बल्कि किसने और किस उद्देश्य से एकत्र किया है — इस पर निर्भर करता है।

Census population की every unit से data collect करने की method है।

Characteristics:

• All units of population are included
• Complete and comprehensive
• No sampling error
• Highly reliable (if done properly)
• Time-consuming and expensive

Examples:

• Indian Census (conducted every 10 years)
• Population count of a university
• Annual student enrolment data

When to Use:

• Small population
• When high accuracy is required
• Legal/administrative requirements
• When resources are available

Sampling population के representative subset (sample) से data collect करने की method है।

Characteristics:

• Only selected units are included
• Less time and cost
• Sampling error possible
• Results can be generalised
• Practical for large populations

Examples:

• Survey of 500 students from a university
• Polling for election predictions
• Quality checking of manufactured goods

When to Use:

• Large population
• Limited time/resources
• When detailed study of every unit is impractical
• When destructive testing is involved

• Probability Sampling — प्रायिकता प्रतिदर्श: Every unit has a known probability of selection.
• Non-Probability Sampling — अप्रायिकता प्रतिदर्श: Selection probability is not known.

• Simple Random Sampling: Every unit has equal chance of selection. Lottery method or random number table.
• Systematic Sampling: Selecting every kth unit from a list. (k = N/n)
• Stratified Sampling: Population divided into strata (groups), then random sampling from each.
• Cluster Sampling: Population divided into clusters, random selection of clusters, then all units in selected clusters.
• Multistage Sampling: Multiple stages of sampling (e.g., state → district → village → household).
• Multiphase Sampling: Different phases of data collection on different units.

• Convenience Sampling: Selecting units that are easily available. (e.g., nearby respondents)
• Purposive/Judgement Sampling: Researcher selects units based on judgement/purpose.
• Quota Sampling: Population divided into quotas, then convenient selection within each quota.
• Snowball Sampling: Initial respondents refer other respondents. Used in hard-to-reach populations.
• Voluntary Sampling: Self-selection by respondents (e.g., online polls).

• Questionnaire: Structured set of questions. Self-administered or online (Google Forms).
• Interview Schedule: Face-to-face or telephonic interview with structured/unstructured questions.
• Observation: Directly watching and recording behaviour/events.
• Experiment: Controlled setting to study cause-effect relationships.
• Focus Group Discussion: Group of people discussing a topic with a moderator.
• Case Study: In-depth study of a single unit/individual.
• Document Analysis: Analysis of existing documents and records.

• Census Data: Indian Census (2011, 2021), NSSO surveys
• Government Reports: RBI, UGC/AISHE, Ministry of Education
• International Sources: UNESCO, WHO, World Bank, IMF
• Research Repositories: Shodhganga (theses), Google Scholar, JSTOR
• Institutional Data: Admission records, exam results, attendance
• Public Data: Open Government Data platform (data.gov.in)

• Trap 1: Primary data = always accurate → ❌ Accuracy depends on methodology and execution.
• Trap 2: Secondary data = always unreliable → ❌ Many secondary sources are highly reliable (Census, RBI).
• Trap 3: Census = always better than sampling → ❌ Sampling is often more practical and cost-effective.
• Trap 4: Sample = always representative → ❌ Representative depends on sampling method and sample size.
• Trap 5: Probability sampling = no errors → ❌ Sampling errors still exist.
• Trap 6: Convenience sampling = most reliable → ❌ It is the least reliable due to bias.
• Trap 7: Data source = data type → ❌ Source and type are different concepts.

A researcher collects data directly from students using a questionnaire. This is an example of:

• A. Secondary data
• B. Primary data
• C. Census data
• D. Published data

Explanation: The researcher collected first-hand data for a specific purpose — this is primary data.

Using Census 2011 data for a research study is an example of:

• A. Primary data
• B. Secondary data
• C. Experimental data
• D. Survey data

Explanation: Census data was already collected by the government — researcher is reusing it (secondary data).

Which sampling method ensures that every unit has an equal chance of selection?

• A. Convenience sampling
• B. Purposive sampling
• C. Simple random sampling
• D. Quota sampling

Explanation: In simple random sampling, every unit has an equal and independent chance of selection.

What is the main limitation of census method?

• A. Sampling error
• B. Time and cost
• C. Bias
• D. Lack of accuracy

Explanation: Census covers all units, which makes it time-consuming and expensive.

Which type of sampling is most appropriate for a large, geographically dispersed population?

• A. Simple random sampling
• B. Convenience sampling
• C. Cluster sampling
• D. Purposive sampling

Explanation: Cluster sampling is cost-effective for large, dispersed populations by sampling clusters (e.g., villages, districts).

Primary Data: First-hand, original, collected by researcher

• Methods: Observation, Interview, Questionnaire, Experiment

Secondary Data: Already collected, reused by researcher

• Sources: Government reports, Research publications, Databases

Census Method:

• All units, accurate, time-consuming, expensive

Sampling Method:

• Selected units, economical, sampling error possible

Probability Sampling:

• Simple Random, Systematic, Stratified, Cluster, Multistage

Non-Probability Sampling:

• Convenience, Purposive, Quota, Snowball

Exam Formula:

UGC NET में Sources and Acquisition of Data के questions primary/secondary classification, census vs sampling, और sampling types पर based होते हैं।