UNIT-4

BRM
Census Survey
• A complete enumeration of all items or study objects in the ‘Population' is known as a census
inquiry.

• An examination of each and every element of the population is called census.

• This method is practically beyond the reach of ordinary researchers.

• Government is the only institution which can get the complete enumeration carried out such as
population census conducted once in a decade.

• Involves a great deal of time, money and energy.

Sample Survey
• When we Select some of the elements in the population and draw
conclusions about the total population It is known as sample survey

• The selected respondents constitute what is technically called a ‘Sample’.

• The selection process is called the ‘sampling technique’.

• The survey so conducted is known as a ‘sample survey’

• Saves cost, labor, and time.

Sample Terminology
Population: The population refers to the entire group of individuals or items that are of interest to the
researcher. It is the larger group from which a sample is drawn, and the results of the study are often generalized
back to this population.
• For example, if you're studying the average income of adults in a country, the population would be all the
adults in that country.
Sampling frame: A sampling frame is a list, database, or representation of the individuals or items that make up
the population. It serves as a reference or source from which a sample can be drawn.
Sampling: Sampling is the process of selecting a subset of individuals or items from the population to be
included in the study or analysis. It is a practical way to gather information about the entire population without
having to examine each and every member. The goal of sampling is to obtain a representative sample that
accurately reflects the characteristics of the population.
Sampling unit: A sampling unit is an individual element or item in the population that is eligible for selection in
the sampling process. It is the basic unit from which a sample is drawn. The sampling unit depends on the nature
of the study and the level of analysis.
• For example, in a survey of households, the sampling unit would be the individual households, whereas in a
study of students' performance, the sampling unit would be the individual students.
 Example
• Population: A company wants to conduct market
research on the purchasing behavior of EV owners
in a specific region. The population would consist
of all EV owners in that region.
• Sampling Frame: The sampling frame for this
study could be a list of registered electric vehicles
in that region, which contains information such as
the vehicle owners' names, addresses, and contact
details. This list would serve as the source from
which the sample is drawn.
• Sample: The company may decide to randomly
select a sample of 500 car owners from the
sampling frame. These 500 car owners would then
be contacted and surveyed to gather information
about their purchasing behavior, preferences, and
satisfaction with their vehicles.
The population can be Finite or Infinite

• Finite Population: A finite population refers to a population that consists of a specific and countable
number of individuals or elements. In other words, the size of a finite population is fixed and can be
determined.

• For example, if you are conducting a survey on the students of a particular school, the population would
be considered finite because you can count the total number of students in that school.

• Infinite Population: An infinite population, on the other hand, refers to a population that is so large or
theoretically infinite in size that it is not possible to count or observe every individual or element within
it. Infinite populations are often encountered when studying phenomena at a broader scale

• For Example: All social media users in the world: If you want to study online behavior patterns or
trends among social media users globally, the population would be considered infinite. As the number of
social media users worldwide is immense and continuously changing, it is not feasible to capture data
from every individual user.
Sample design:

• Sample design refers to the process of determining how a sample will be selected from a
larger population for a research study or survey.

• It involves making decisions about the sampling method, sample size, and other considerations that
will ensure the sample is representative of the target population.

• The sample design is a crucial step in research as it directly impacts the validity and
generalizability of the study findings.

• The chosen sample design should aim to minimize bias and maximize the accuracy of the
inferences made about the population based on the sample.
Steps in Sample Design
Characteristics of a Good Sample Design:

• Sample design must result in a truly representative sample.

• Sample design must be such which results in a small sampling error.

• Sample design must be viable in the context of funds available for the research study.

• Sample design must be such so that systematic bias (results from errors in the sampling
procedures) can be controlled in a better way.

• Sample should be such that the results of the sample study can be applied, in general, for
the universe with a reasonable level of confidence.
Sampling Error
Sampling Error Example
Sampling Error • Frame Error: Frame error, also known as coverage error or
population specification error, occurs when there is a mismatch
between the target population and the population included in the
study or survey. It arises when the sampling frame or the
defined population for the study does not accurately represent
the intended population of interest.
• Chance Error: Chance error, also referred to as random error or
sampling error, is the variability in measurements or
observations that occurs due to random fluctuations or inherent
variability in the data. arise due to factors such as natural
variability, limitations of measuring instruments, or human
error.

Non-Sampling Error
• Response error, also known as measurement error or non-
sampling error, refers to errors that arise from respondents'
behavior or characteristics that affect their responses in a survey
or study. Response errors can occur due to various reasons, such
as misinterpretation of questions, memory recall issues, social
desirability bias, or deliberate misreporting.
Probability Sampling
• A lottery method in which individual units are picked up from the whole group not
deliberately but by some mechanical process.

• Each individual in the population has an equal chance of being chosen.

• This randomness helps to minimize selection bias and increase the representativeness of
the sample.

• Methods of drawing a random sample:

• 1.The fishbowl draw
• 2.A table of randomly generated numbers
 Step-2
Step-1

Step-3
Dr. Pallavi Chaturvedi, Associate Professor, Department of
Business Administration
Non-probability sampling
• Non-probability sampling refers to a category of sampling techniques where the
selection of individuals or items from a population is based on non-random
methods.

• The probability of any particular member of the population being included in the
sample is unknown and cannot be accurately determined.

• These are commonly used when it is not feasible or practical to use probability
sampling techniques, such as in situations with limited resources, time
constraints, or specific research objectives.
Types of Non Probability Sampling
Convenience (Accidental) sampling
• Individuals are selected based on their availability and accessibility to the
researcher. The selection of participants is based on convenience and ease of
access, rather than randomization or specific criteria.

• With convenience sampling, the researcher chooses individuals who are

conveniently available and willing to participate in the study. This often involves
selecting individuals who are easily accessible or within close proximity to the
researcher.
Convenience sampling
• Convenience sampling is frequently used in situations where time, cost, or
practical constraints make it challenging to employ other sampling techniques.

• It is a convenient and quick way to gather data, but it may introduce biases and
limit the generalizability of the findings to the larger population.

• The findings from convenience sampling should be interpreted with caution and
may not be representative of the entire population.
Judgmental (Purposive) sampling
• Judgmental sampling is a non-probability sampling technique where the researcher selects
participants based on their own judgment or expertise.

• Selection of participants is subjective and relies on the researcher's judgment. The researcher
handpicks individuals who they believe possess the necessary knowledge, experience, or
characteristics to provide valuable insights or represent the population of interest.
Quota sampling
• Involves selecting individuals to be included in the sample based on specific quotas or predetermined
characteristics.

• It aims to ensure that the sample reflects the composition of the population in terms of certain key
variables.

• Steps:

• The population is divided into subgroups or strata based on relevant characteristics such as age, gender,
income level, or geographic location.

• The researcher then sets quotas for each subgroup to ensure that the sample represents the proportions
found in the population.

• Participants within each subgroup are selected using convenience or judgment sampling until the quotas are
met.
Quota Sampling
Snowball sampling
• Snowball sampling is a non-probability sampling technique where participants
are initially selected based on the researcher's or participant's knowledge, and
then additional participants are recruited through referrals from those initial
participants. In snowball sampling, the sample grows like a snowball rolling
downhill, as each participant refers or introduces others who meet the study's
criteria.
Types of Probability Sampling
Simple Random Sampling
• In simple random sampling, every individual or element in the population has an
equal probability of being chosen as part of the sample, and the selection of each
element is independent of the others.

• Researchers typically assign a unique identification number to each member of the

population and use a random number generator or a table of random numbers to
select the sample. This ensures that the selection process is unbiased and follows
the principles of random sampling.
Simple Random Sampling
Systematic Sampling

• Systematic sampling is a technique used in research where elements

from a larger population are selected at regular intervals.

• In systematic sampling, the population is ordered in some way, and a

fixed interval is used to select the sample. It involves selecting every
nth element from the population after determining a random starting
point.
Systematic Sampling
Stratified Random Sampling
• Stratified random sampling is a sampling technique used in research where the population is divided
into distinct subgroups or strata, and a random sample is selected from each stratum.

• The goal of stratified random sampling is to ensure that the sample is representative of the entire
population while allowing for comparisons and analysis within each subgroup.

• Stratified random sampling is practical when the population has distinct subgroups, and the goal is to
make inferences about each subgroup or the population as a whole.
Stratified Random Sampling
Proportionate Stratified Sampling
• Proportionate Stratified Sampling: In proportionate stratified sampling, the sample size selected
from each stratum is proportionate to the size or proportion of that stratum in the overall
population.

• The goal is to ensure that each stratum is adequately represented in the sample based on its
relative size or importance.

• The sample size within each stratum is determined by multiplying the desired overall sample size
by the proportion of the population belonging to that stratum.
EXAMPLE:

• A Researcher wants to survey a college campus to understand students' preferences regarding

extracurricular activities. The campus has three main colleges: the College of Arts, the College of
Sciences, and the College of Business. The researcher decides to use proportionate stratified
sampling to ensure representation from each college.
• Define the Population: The population is all the students on the college campus.
• Identify Strata: The strata are the three colleges: Arts, Sciences, and Business.
• Determine Proportions: The Arts College has 1,000 students, the Sciences College has 800
students, and the Business College has 600 students. She wants to ensure that each college is
proportionately represented in the sample, so she decides to select 40% of the sample from the
Arts College, 30% from the Sciences College, and 30% from the Business College.
• Calculate Sample Sizes: Multiply the proportion for each stratum by the desired sample size. Let's
assume she wants a total sample size of 200 students.
n1=nxN1/N

n1=200x 1000/2400= 84

n2=200x 800/2400= 66

N3=200x600/2400=50
Numerical

• If a bank wants to conduct a survey to understand the problem that its

customers are facing.
• If we have 10,000 customers of a bank in such a way that 1500 of them are
big account holders, 3500 of them are medium-sized account holders, 5000
are small account holders. Suppose the total budget for sampling is fixed at
20,000 Rs. And the cost of sampling a unit is Rs. 200. Find out samples
from strata 1,2,3
n=20000/200= 100
Disproportionate Stratified Sampling
• Disproportionate sampling, also known as non-proportional sampling, is a
sampling technique used in research and statistics where the selection of
individuals or elements for a sample is intentionally biased or skewed towards
certain characteristics or groups within the population.

• The size of the sample drawn from a particular stratum is not proportional to the
relative size of that stratum.
Example

• Customer Satisfaction Surveys: In a survey conducted by a hotel

chain, if the goal is to assess the satisfaction levels of their elite
loyalty program members, the company may disproportionately
sample this segment of customers. This approach allows the hotel
chain to obtain more detailed feedback from their most valued
customers and tailor their services accordingly.
Cluster Sampling
• Cluster sampling is a sampling technique used in statistics and research to select a
subset of individuals or groups (clusters) from a larger population. In cluster
sampling, the population is divided into groups or clusters, and a random
selection of clusters is chosen to be included in the sample.

• Clusters are typically formed based on some natural grouping within the
population, such as geographical regions, schools, households, or organizational
units. The clusters should ideally be heterogeneous internally but similar to each
other in terms of the characteristics being studied. This allows for a
representative sample that reflects the diversity of the population.
Cluster Sampling
• Step 1: Define population
• Step 1: Divide the Population into clusters
• Step 3: Randomly select clusters to use as a sample
• Step 4: Collect data from the sample
• One-stage Cluster Sampling: In one-stage cluster sampling, the researcher selects a random
sample of clusters from the population, and then all individuals within the selected clusters are
included in the sample. This means that only one level of sampling is involved.

• Two-stage Cluster Sampling: In two-stage cluster sampling, there are two levels of sampling. In
the first stage, a random sample of clusters is selected from the population. Then, in the second
stage, a random sample of individuals is selected from each of the selected clusters.

• Multistage Cluster Sampling: Multistage cluster sampling involves more than two levels of
sampling. It is used when the population is organized into a hierarchical structure with multiple
levels of clustering. In each stage, a subset of clusters is selected, and then a subset of units
within the selected clusters is chosen. This approach is useful when the population is large and
dispersed.
• One-stage Cluster Sampling:

• Suppose a marketing researcher wants to study consumer preferences for a particular brand of smartphones in a large metropolitan area. In one-stage cluster
sampling, the researcher may randomly select a sample of retail stores as clusters from different neighborhoods within the city. Then, within each selected store, the
researcher can survey all customers who enter the store during a specified period. This approach saves time and resources by collecting data from all individuals
within the selected clusters (stores) without the need to individually sample customers from each store.

• Two-stage Cluster Sampling:

• Consider a market research study on consumer buying behavior for a specific product category, such as home appliances. In two-stage cluster sampling, the
researcher may first randomly select a sample of cities or towns as clusters from a list of all cities in a region. Then, within each selected city, a random sample of
households can be selected for data collection. The researcher can survey or interview individuals within the selected households to gather information on their
appliance purchasing habits. This method allows for sampling diverse locations (clusters) while still obtaining representative data from individual households.

• Multistage Cluster Sampling:

• Imagine a marketing research project aiming to understand consumer preferences for a new food product at a national level. In multistage cluster sampling, the
researcher may first randomly select a sample of states as clusters from a list of all states in the country. Then, within each selected state, a random sample of cities
or towns can be chosen. Finally, within each selected city or town, the researcher can randomly select supermarkets or grocery stores as the final clusters. The
researcher can then survey customers visiting these supermarkets or grocery stores to gather data on their preferences for the new food product. This method allows
for capturing geographical diversity and consumer variation across multiple stages of sampling.

• In marketing research, cluster sampling methods can be valuable when the population is large and geographically dispersed. By selecting clusters and sampling
within those clusters, researchers can efficiently obtain data from a representative subset of the target population while reducing costs and logistical challenges. The
choice of sampling method depends on the research objectives, the structure of the population, and practical considerations such as time and budget constraints.
Numerical
• The following are the number of departmental stores in 15 cities: 35,
17, 10, 32, 70, 28, 26, 19, 26, 66, 37, 44, 33, 29 and 28. If we want to
select a sample of 10 stores, using cities as clusters and selecting
within clusters proportional to size, how many stores from each city
should be chosen?
• (Use a starting point of 10).
 Selection
Cumulative
City No. of Stores Points Falling in Stores Selected
Total
this Range
1 35 35 10 1
2 17 52 – 0
3 10 62 – 0
4 32 94 63 1 Sampling interval (I)=
5 70 164 116, 169 2 Total Stores/Sample
6 28 192 – 0 Size
7 26 218 – 0
8 19 237 222 1
500/10= 50
9 26 263 275 1
Random start: 10
10 66 329 328 1
11 37 366 381 1
Selection points: (Store
12 44 410 – 0
Positions)
13 33 443 434 1 10, 60, 110, 160, 210,
14 29 472 – 0 260, 310, 360, 410, 460
15 28 500 487 1
Sample Size Determination
• The appropriate sample size depends on several factors, including the
Research objectives,
• study design,
• statistical methods,
• desired level of precision,
• expected effect size, and
• available resources.
Sample Size for Infinite Population

n = required sample size

Z = Z-score (based on the desired confidence level)
𝑝 = estimated proportion of the population (if unknown, use
0.5 for maximum variability)
𝑒 = margin of error (in decimal form, e.g., 5% = 0.05)
 Example Infinite Population
• z = 1.96 (Based on a 5% margin of
error. Data are assumed two-tailed
(i.e., a margin of error of 2.5% on
each end of a normal distribution
curve), thus, a value of 0.9750 will
be looked up within the z-score
table.)
• p̂ = 50% or 0.50 (This value is often
pulled from previous research/
literature. If unsure, use 50%.)
• ε = 5% or 0.05 (Same value used to
get the z-score estimate, but
provided as a decimal/
percentage.)
Sample Size for Finite Population
 Example Finite Population
• n = 385
• z = 1.96 (Based on a 5% margin of error. Data
are assumed two-tailed (i.e., a margin of error
of 2.5% on each end of a normal distribution
curve), thus, a value of 0.9750 will be looked
up within the z-score table.)
• P̂ = 50% or 0.50 (This value is often pulled
from previous research/ literature. If unsure,
use 50%.)
• Ε = 5% or 0.05 (Same value used to get the z-
score estimate, but provided as a decimal/
percentage.)
• N = 1000 (This value is inserted if known and
is often pulled from research/ literature or
some prior background knowledge about the
population of interest.)
Factors affecting Sample Size
• Population Size
• Level of confidence
• Margin of Error
• Variability of data
• Research design
• Resources
Factors affecting Sample Size
• Population size: The size of the target population plays a role in determining the sample size. Generally, a larger
population requires a larger sample size to achieve a representative sample.

• Margin of error: The margin of error is the maximum amount of sampling error allowed in the estimation of a
population parameter. A smaller margin of error necessitates a larger sample size. It measures the precision with
which an estimate from a sample reflects the population. For example - In a national voting poll the Margin of Error
might be + or – 3%.

• Desired level of confidence: The level of confidence or margin of error you want to achieve affects the sample size.
A higher confidence level or lower margin of error requires a larger sample size.

• Population variability: The variability or dispersion of the characteristic being measured within the population
affects the sample size. If the population is more heterogeneous, a larger sample size is needed to account for the
increased variability.

• Time and resources: Practical considerations, such as time constraints, available resources, and budget limitations,
can also influence the sample size determination. Researchers often need to strike a balance between statistical
requirements and practical feasibility.