NLNET Law-VerseThe Universe of Law
UGC NET PAPER 1 · UNIT 07 · MODULE 2: UNDERSTANDING DATA · TOPIC 2

Sources and Acquisition of Data

Primary and secondary data, collection methods and the pairs examiners contrast: questionnaire and schedule, census and sample.

  • 6 sub-topics
  • 4 min read
  • Topic 2 of 11

What this topic covers

6 sub-topics, in the order to study them. Tap one to jump to it.

  1. §OverviewAcquisition of data means the process of collecting data from the field or from existing records.
  2. 2.1Primary data and methods of collectionPrimary data are original, collected first-hand and specific to the study.
    • Observation
    • Interview
    • Questionnaire
    • Schedule
    • Experiment
  3. 2.2Secondary data and their sourcesSecondary data come from published or unpublished records.
    • Published, government
    • Published, international
    • Published, other
    • Unpublished
  4. 2.3Primary versus secondary dataBasis Primary data Secondary data Originality Fresh and original Already available Cost and time High Low Fit with problem Tailor-made May not fit exactly Quality control…
  5. 2.4Questionnaire versus scheduleBasis Questionnaire Schedule Filled by Respondent Investigator Respondent must be Literate Can be illiterate Response rate Low High Cost Low High Area covered Wide Limited…
  6. 2.5Census versus sample surveyA census is complete enumeration of every unit in the population.
    • Probability sampling
    • Non-probability sampling
  7. 2.6Worked examplesExample 1.
    • Example 1
    • Example 2
    • Example 3

Revision mode hides the full explanations and keeps each section's key terms, lists and tables. Use it from your second read onward.

§Overview#

Acquisition of data means the process of collecting data from the field or from existing records. The first decision a researcher makes is whether to collect fresh data (primary) or to use data that already exist (secondary). The choice depends on cost, time, the exact fit with the research problem, and the need for control over quality.

Back to topic map ↑

2.1Primary data and methods of collection#

In brief

Primary data are original, collected first-hand and specific to the study.

  • Observation
  • Interview
  • Questionnaire
  • Schedule
  • Experiment
  • Survey

Primary data are original, collected first-hand and specific to the study. The main methods are:

  • Observation: the researcher watches and records behaviour. It can be structured or unstructured, and participant (the observer joins the group) or non-participant.

  • Interview: a face-to-face or telephone conversation. It can be structured (fixed questions), semi-structured, or unstructured (open conversation).

  • Questionnaire: a set of written questions that the respondent fills in personally. It is cheap, reaches a wide area and avoids interviewer bias, but response rates are low and it needs literate respondents.

  • Schedule: a set of questions filled in by the investigator while meeting the respondent. It suits illiterate respondents and gives a higher response rate, but costs more.

  • Experiment: data generated by manipulating one variable and observing the effect on another.

  • Survey: collecting data from a large group, either by census or by sample.

  • Focus group discussion and case study: in-depth qualitative methods that explore a topic with a small group or a single case.

Back to topic map ↑

2.2Secondary data and their sources#

In brief

Secondary data come from published or unpublished records.

  • Published, government
  • Published, international
  • Published, other
  • Unpublished

Secondary data come from published or unpublished records.

  • Published, government: Census of India, National Statistical Office (NSO, formerly NSSO) surveys, Economic Survey, National Family Health Survey (NFHS), UDISE+, All India Survey on Higher Education (AISHE), and Reserve Bank of India reports.

  • Published, international: UN, World Bank, WHO, UNESCO, IMF.

  • Published, other: research journals, books, newspapers, trade and business reports.

  • Unpublished: diaries, letters, institutional files, office records.

Before using secondary data, check the compiler's purpose, definitions used, date of collection, and possible bias. This checking step is called assessing the suitability and reliability of the data.

Back to topic map ↑

2.3Primary versus secondary data#

In brief

Basis Primary data Secondary data Originality Fresh and original Already available Cost and time High Low Fit with problem Tailor-made May not fit exactly Quality control…

Basis Primary data Secondary data
Originality Fresh and original Already available
Cost and time High Low
Fit with problem Tailor-made May not fit exactly
Quality control With the researcher Depends on the original compiler
Example Your own field survey Census tables

Back to topic map ↑

2.4Questionnaire versus schedule#

In brief

Basis Questionnaire Schedule Filled by Respondent Investigator Respondent must be Literate Can be illiterate Response rate Low High Cost Low High Area covered Wide Limited…

Basis Questionnaire Schedule
Filled by Respondent Investigator
Respondent must be Literate Can be illiterate
Response rate Low High
Cost Low High
Area covered Wide Limited
Sent by Post or online Carried by the investigator

Back to topic map ↑

2.5Census versus sample survey#

In brief

A census is complete enumeration of every unit in the population.

  • Probability sampling
  • Non-probability sampling

A census is complete enumeration of every unit in the population. India's decadal Population Census is the standard example. It gives full detail but costs a great deal and takes years. A sample survey studies a selected part of the population and infers about the whole. It is faster and cheaper, but results carry sampling error.

  • Probability sampling (every unit has a known chance of selection): simple random, systematic, stratified, cluster and multistage sampling.

  • Non-probability sampling (selection is not by chance): convenience, purposive (judgement), quota and snowball sampling.

Only probability sampling allows a sampling error to be estimated. Stratified sampling divides the population into homogeneous groups (strata) and samples from each. Cluster sampling divides it into natural groups (clusters) and samples whole clusters. Snowball sampling is used for hard-to-reach groups where existing respondents refer the next ones.

Back to topic map ↑

2.6Worked examples#

In brief

Example 1.

  • Example 1
  • Example 2
  • Example 3

Example 1. A researcher visits 300 farmers and records their answers on a form by speaking to each. Which tool is used? Answer: a schedule, because the investigator fills it.

Example 2. A student uses NFHS tables to study child nutrition without meeting any child. Which data are these? Answer: secondary data.

Example 3. A university wants the opinion of every one of its 120 faculty members. Which approach is this? Answer: census (complete enumeration), because every unit is covered.

Back to topic map ↑

Test yourself on Sources and Acquisition of Data

Practise with free MCQs and answer keys, then return here to fix what you got wrong.

↑ Module 2: Understanding Data · All 12 topics in Data Interpretation · Unit overview and sample MCQs

Last updated: . Notes are a revision aid; check official UGC sources for any rule or scheme that changes over time. Spotted a mistake? Write to netlawverse@gmail.com.