C HA P TE R 3
Census and Sample Methods of Collection of Data
Comprehensive, Simplified Revision Notes with Visual Frameworks
1. Introduction & Key Concepts
When an investigator conducts a statistical enquiry, they need to gather data from the specific field of study.
To understand how data is collected, we must first learn about two foundational terms in Statistics: Universe
(Population) and Sample.
Universe or Population
In statistics, the terms Universe or Population refer to the aggregate (total sum) of all items about
which we want to obtain information for our investigation.
Example: If we are studying the economic background of 2,000 students in a college, then all 2,000 students
together form our population. Each individual student is called an "Item".
Sample
A Sample refers to a smaller section, group, or a fraction selected from the total population that is
studied to represent the characteristics of the whole universe.
Example: Instead of examining all 2,000 students, if we pick 150 students randomly to study their backgrounds,
those 150 students constitute a Sample.
Population / Universe
Sample
Selection
(All 2,000 Items) (Selected Reps)
Figure 3.1: Visualizing the relationship between Population (Universe) and Sample.
2. Census Method (Complete Enumeration)
The Census Method is that technique in which data is collected from every single item belonging to the
universe or population relating to the problem under investigation.
Statistics for Economics — Chapter 3 Study Notes 1
Suitability:
• When the field of investigation is limited/small in size.
• When items in the population have widely diverse (different) characteristics.
• When a very high degree of accuracy and reliability is required.
• When intensive, in-depth evaluation of each item is necessary.
Merits & Demerits:
Merits Demerits
• Reliable & Accurate: Results are highly precise • Highly Costly: Requires an immense financial
because each item is individually examined. budget, making it tough for ordinary studies.
• Less Biased: Completely eliminates personal bias • Huge Manpower: Demands a massive army of
of the investigator regarding sample choice. trained enumerators and supervisors.
• Extensive Information: Collects deep, detailed • Not practical for large fields: Becomes nearly
data (e.g., Indian Census gathers age, sex, jobs, impossible to execute if the population size is
literacy together). infinitely huge.
3. Sample Method (Partial Study)
The Sample Method is that technique in which data is collected only from a representative group (called a
sample) extracted from the universe, and conclusions are drawn for the entire population.
Everyday Example: A cook checks just 2 or 3 grains of rice from a boiling pot to decide whether the entire batch of rice
is cooked.
Suitability:
• When the size of the population is very large or infinite.
• When items are mostly homogeneous (similar in nature).
• When an ordinary level of accuracy is sufficient.
• When testing involves the complete destruction of the item (e.g., testing blood or a car's crash safety).
Statistics for Economics — Chapter 3 Study Notes 2
Merits & Demerits:
Merits Demerits
• Economical: Saves substantial money and • Partial & Biased: If the sample is picked poorly or
resource budgets compared to a full census. based on personal favoritism, conclusions will be
• Time-Saving: Data processing is very fast totally wrong.
because the number of items is tiny. • Difficulty in Framing: Choosing a sample that
• Easy Administration: Managing a small team of perfectly mimics the whole population is
data collectors is much more convenient. mathematically difficult.
• Needs Special Knowledge: Requires
understanding of advanced sampling techniques
and formulas.
Essentials of a Good Sample:
1. Representative: The sample must possess all the traits of the underlying population.
2. Independent: The inclusion of one item in the sample must not force or prevent the inclusion of another.
3. Homogeneous: If multiple samples are chosen, they must be consistent and uniform with each other.
4. Adequate: The number of selected items must be large enough to generate safe, stable conclusions.
Statistics for Economics — Chapter 3 Study Notes 3
4. Methods of Sampling
Sampling techniques are broadly categorized into two major classes based on whether items have a
scientifically equal chance of selection: Random Sampling and Non-Random Sampling.
METHODS OF SAMPLING
1. Random Sampling 2. Non-Random Sampling
• Purposive / Deliberate
• Lottery Method
• Stratified / Mixed
• Random Number Tables
• Systematic Sampling
• Quota Sampling
• Convenience Sampling
Figure 3.2: Classification tree of various sampling methods.
A. Random Sampling
Random sampling is a method where each and every item of the universe has an equal chance of being
selected in the sample. The selection is completely free from the personal bias of the investigator.
• Lottery Method: Paper slips are made for all items, shuffled thoroughly in a container, and slips are drawn
out blindly to build the sample.
• Tables of Random Numbers: Standardized numerical tables (like Tippett's Tables) are referenced to
extract values impartially.
B. Non-Random Sampling
In this approach, items do not have an equal probability of selection; instead, the investigator’s choices or
structural restrictions guide the selection.
1. Purposive / Deliberate Sampling: The investigator explicitly uses their own judgment to pick items that
they believe are the most critical representatives of the universe. (High risk of bias).
2. Stratified or Mixed Sampling: Used when items are highly diverse. The population is split into groups
called Strata based on characteristics (like income or streams). Then items are picked randomly from each
stratum.
3. Systematic Sampling: Items are listed in a clear numerical, alphabetical, or geographical sequence.
Then, every nth item is picked as part of the sample (e.g., choosing every 10th name on a roster).
4. Quota Sampling: The population is categorized into groups, and a set proportion (quota) of items is
assigned to the field enumerators. The enumerator then uses personal judgment to fill that specific quota.
5. Convenience Sampling: The investigator ignores scientific criteria and selects items solely because they
are closest, simplest, or most convenient to gather data from.
Statistics for Economics — Chapter 3 Study Notes 4
5. Comparative Look: Census vs. Sample Method
To conclude, here is a quick-glance comparison summarizing the principal differences between both collection
strategies:
Basis Census Method Sample Method
Coverage Data is gathered from every single item of Data is gathered from only a limited few
the population. representative items.
Suitability Best for small, limited fields with diverse Best for vast, extensive fields with
items. homogeneous items.
Cost & Time Requires massive expenditure, huge Highly cost-effective, requires small
manpower, and long duration. teams, and saves massive time.
Accuracy Yields absolute accuracy and precision. Yields relative accuracy (errors can
happen but can be estimated).
Nature of Items Essential if items are highly complex and Best when items are largely uniform or if
varied. tests destroy items.
Statistics for Economics — Chapter 3 Study Notes 5