Business Research Methodology

Chapter 16:
Sampling Design and Procedures

1
 Sampling Terminology
• A sample is a representative subset of a larger
population.
– The purpose of sampling is to estimate an unknown
characteristic of a population.
• A population is any complete group—for example,
people, sales territories, stores, or college students.
– Shares some common set of characteristics.
• Population element refers to an individual member
of the population.
• A census is an investigation of all the individual
elements that make up the population.
2
 Reasons for Sampling?
• Pragmatic Reasons:
– Budget and time constraints; For example, census of past
purchasers’ reactions of Ford motors.
– Sampling cuts costs, reduces labor requirements, and gathers
vital information quickly.
• Accurate and Reliable Results: Sample may sometimes be
more accurate than a census since non-sampling errors
may increase because of the increased volume of work.
• Destruction of Test Units: Especially those in quality-
control testing, require the destruction of the items being
tested. For example, testing bulbs, blood count.
• The physical impossibility of checking all items in the
population. Ex: population of fish, insects, animals etc.

3
Stages in the Selection of a Sample

4
 Stages in the Selection of a Sample (cont’d)

1. Defining the Target Population:

– The question to answer is, “Whom do we want to
talk to?”
– All students of Dhaka University
2. The Sampling Frame: A list of elements from
which a sample may be drawn; also called
working population.
– A list of all current students of all departments
– In practice, almost every list excludes some
members of the population.

5
 Stages in the Selection of a Sample (cont’d)

3. Sampling Units: A single element or group of

elements subject to selection in the sample.
– For example, if an airline wishes to sample
passengers, it may take every 25th name on a
complete list of passengers.
– Alternatively, the airline could first select certain
flights as the sampling unit and then select certain
passengers on each flight. In this case the
sampling unit would contain many elements.

6
 Sampling errors: Less Than Perfectly
Representative Samples
• Sample considering both errors produce less than perfectly
representative samples.

7
 Random Sampling Error
• The difference between the sample results and
the result of a census conducted using identical
procedures
• Statistical fluctuation due to chance variations
means the sampling units, even if properly
selected according to sampling theory, may not
perfectly represent the population, but generally
they are reliable estimates.
• As sample size increases, random sampling error
decreases.

8
 Systematic Errors
• Errors result from nonsampling factors such as
unrepresentative sample results
• Not due to chance
• Due to study design or imperfections in
execution

9
 Two Major Categories of Sampling Methods

• Probability sampling:
– every member of the population has a known,
nonzero probability of selection.
• Non-probability sampling
– Probability of selecting any particular
member is unknown.
– Researchers rely heavily on personal judgment.
– No appropriate statistical techniques exist for
measuring random sampling error from a
nonprobability sample.

10
 Nonprobability Sampling Methods

• Convenience sampling
• Judgment sampling
• Quota sampling
• Snowball sampling

11
12
 Convenience Sampling
• The sampling procedure of obtaining the people
or units that are most conveniently available.
– For example, television stations often present person-
on-the-street interviews that are presumed to reflect
public opinion
• Used to obtain a large number of completed
questionnaires quickly and economically.
• Used when obtaining a sample through other
means is impractical.
– Also called haphazard or accidental sampling
• Are best used for exploratory research when
additional research will subsequently be
conducted with a probability sample.
13
 Judgment Sampling
• An experienced individual selects the sample
based on his or her judgment about some
appropriate characteristics required of the
sample member.
– Also called purposive sampling.
– For example the consumer price index (CPI) is based
on a judgment sample of market-basket items,
housing costs, and other selected goods and services.
• Researchers select samples that satisfy their
specific purposes, even if they are not fully
representative.
– Judgment sampling often is used in attempts to
forecast election results.

14
 Quota Sampling
• Ensures that the various subgroups in a
population are represented on relevant sample
characteristics to the exact extent that the
investigators desire.
• It should not be confused with stratified sampling
since the interviewer has a quota to achieve.
– For example, an interviewer in a particular city may be
assigned 100 interviews, 35 with owners of Sony TVs,
30 with owners of Samsung TVs, 18 with owners of
Panasonic TVs, and the rest with owners of other
brands.
15
 Quota Sampling
• Possible bias:
– Quota samples tend to include people who are
easily found, willing to be interviewed, and middle
class.
• However offers speed of data collection, lower
costs, and convenience.
• Appropriate when the researcher knows that a
certain demographic group is more likely to
refuse to cooperate with a survey.
– For instance, if older men are more likely to
refuse, a higher quota can be set for this group.
16
 Snowball Sampling
• Initial respondents are selected by probability
methods. Additional respondents are
obtained from information provided by the
initial respondents.
• Used to locate members of rare populations
by referrals.
• Reduced sample sizes and costs are clear-cut
advantages of snowball sampling.
• Bias is likely to enter into the study because
one is suggesting another.
• Appropriate for focus groups.
17
 Probability Sampling methods

• Simple random sample

• Systematic sample
• Stratified sample
• Cluster sample

18
 Simple Random Sampling

• A sampling procedure that ensures that each

element in the population will have an equal
chance of being included in the sample.
• A table of random numbers can be used.
• Widely used.

19
 Systematic Sampling
• A starting point is selected by a random process
and then every nth number on the list is selected.
– Suppose a researcher wants to take a sample of 1,000
from a list of 200,000 names. With systematic
sampling, every 200th name from the list would be
drawn.
• The problem of periodicity (the tendency to recur
at intervals) occurs if a list has a systematic
pattern—that is, if it is not random in character.
– Collecting retail sales information every seventh day
would result in a distorted sample because there
would be a systematic pattern.
20
 Stratified Sampling
• Characteristics of the samples are homogenous
within the group but heterogeneous across the
group.
• Strata: Dividing the population into subgroups.
• Stratum: Each subgroup is known as stratum.
• Choosing strata on the basis of existing information.
– For example, classifying retail outlets based on
annual sales volume.
• Then a subsample is drawn using simple random
sampling within each stratum.

21
 Stratified Sampling
• Used to obtain a more efficient sample than a simple
random sampling.
– For example, data are collected both from urban and rural
customers.
• Random sampling error will be reduced and sample will
accurately reflect the population.
• In a proportional stratified sample the number of
sampling units drawn from each stratum is in proportion
to the relative population size of the stratum.
• In a disproportional stratified sample the sample size for
each stratum is not allocated in proportion to the
population size.
22
Stratified Sampling

23
 Cluster Sampling
• The primary sampling unit is not the individual
element in the population but a large cluster of
elements; clusters are selected randomly.
• Economical while retains the characteristics of a
probability sample.
– Consider a researcher who must conduct five hundred
personal interviews with consumers scattered
throughout Bangladesh. Travel costs and time are
likely to be enormous. If the product equally appeals
dwellers of Dhaka and Chittagong then only one
cluster can be interviewed.
• Ideally a cluster should be heterogeneous.
24
 Examples of Clusters

Population Element Possible Clusters in the United States

Airline travelers Airports

Planes

Sports fans Football stadiums

Basketball arenas
Baseball parks

25
Stratified vs Cluster Sampling
 What is the Appropriate Sample Design?

• Degree of accuracy
• Resources
• Time
• Advanced knowledge of the population
• National versus local
• Need for statistical analysis

27
 Internet Sampling is Unique

• Internet surveys allow researchers to rapidly

reach a large sample.
• Sample size requirements can be met
overnight or almost instantaneously.
• Survey should be kept open long enough so all
sample units can participate.

28
 Internet Sampling

• Major disadvantage
– Lack of computer ownership and Internet access
among certain segments of the population.
• Yet Internet samples may be representative of
a target populations.
– Target population - visitors to a particular Web
site.
• Hard to reach subjects may participate.

29
 Z-Value
▪ For Two-tailed test, Z-value for 90%, 95%, and 99%
confidence interval are 1.645, 1.96, and 2.58 respectively.
▪ For One-tailed test, Z-value for 90%, 95%, and 99%
confidence interval are 1.28, 1.645, and 2.33 respectively.
▪ For Two-tailed test, Z-value for 10%, 5%, and 1%
significance level (α) are 1.645, 1.96, and 2.58
respectively.
▪ For One-tailed test, Z-value for 10%, 5%, and 1%
significance level (α) are 1.28, 1.645, and 2.33
respectively.
 Sampling Math Q1
• Historically, the standard deviation of the
mortgage values tends to be around $20000.
To estimate the average amount of these
mortgages, the bank wishes to construct
a 95% confidence interval for the mean using
a random sample. If the bank wants the
margin of error to be no more than $2000,
what is the sample size?
Sampling Math Q1- Ans
 Sampling Math Q2
• In the past, 19% of all homes with a stay-at-home
parent, the father is the stay-at-home parent. An
independent research firm has been charged with
conducting a sample survey to obtain more current
information.
(a) What sample size is needed if the research firm's
goal is to estimate the current proportion of homes
with a stay-at-home parent in which the father is the
stay-at-home parent with a margin of error of 0.03?
Use a 95% confidence level.
(b) Repeat part (a) using a 99% confidence level.
Sampling Math Q2-Ans
 Sampling math Q3
• A political pollster wants to estimate the proportion of voters
who will vote for the Democratic candidate in a presidential
campaign. The pollster wants to have 90% confidence that
their prediction is correct to within plus or minus 0.04 of the
population proportion (i.e., margin of error is 4%).
• a. What sample size is needed? (Hints: Assume sample
proportion (p) = 50% if not given)
• b. If the pollster wants to have 95% confidence what sample
size is needed?
• c. If she wants to have 95% confidence and a sampling error
of plus or minus 0.03, what sample size is needed?
• d. On the basis of your answers to a, b, and c, what general
conclusions can be reached about the effects of the
confidence level desired and the acceptance sampling error
on the sampling size needed?
Sampling math Q3-Ans
 Sampling math Q3-Ans (cont’d)

The level of Confidence and sample size required are positively related.
Sampling error and sample size required are negatively related.
 Sampling math Q4
• Suppose X̄ = 21.2 results from the random sampling
of a N(μ, 64) population. Obtain a 95% confidence
interval estimate for the population mean in the
following cases:
• (i) n = 25
• (ii) n = 100
 Sampling math Q4- Answer
(i) n = 25
95% Confidence interval = X̄ ± z* σ / (√n)
=21.2 ± 1.96*[64/√25]
=21.2 ±25.09
=-3.89 to 46.29
(ii) n = 100
95% Confidence interval = X̄ ± z* σ / (√n)
=21.2 ± 1.96*[64/√100]
=21.2 ±12.54
= 8.66 to 33.74
As sample size increases the confidence interval for the
population mean decreases.
 Sampling Maths (H.W.)
• Q5: A survey of shoppers is planned to see what percentage
use credit cards. Prior surveys suggest 53% of shoppers use
credit cards. How many randomly selected shoppers must we
survey in order to estimate the proportion of shoppers who
use credit cards to within 3% with 95% confidence?

• Q6: In a study of government financial aid for college students

it becomes necessary to estimate the percentage of full time
college students who earn a bachelor's degree in four years or
less. Find the sample size needed to estimate that percentage.
Use a 0.4 margin of error and use a confidence level of 99%.
Thank You!!!

41