MKT 0019 1

Marketing
MKT0019 AlgonquinResearch
College
 MKT 0019
Market Research

SAMPLING | THEORY AND

METHODS

CHAPTER 6

MKT0019 Algonquin College 2

 CHAPTER 6 | HIGHLIGHTS
 The necessity for
sampling.
 Sampling definitions.
 Sampling techniques.
 Reliable vs valid.
 Sample errors.
 Sampling process steps.

MKT0019 Algonquin College

 SAMPLING
 Sampling involves
selecting a relatively
small number of
elements from a larger
defined group of
elements.
 Expecting that the sample
will be representative of
the larger group.
 Sampling is used when a
census is not possible.

MKT0019 Algonquin College

 NECESSITY FOR SAMPLING | AN EXAMPLE
An insurance company wants to carry out
research among new car owners in Canada

Statistics Canada (2023): new motor vehicle sales

MKT0019 Algonquin College
between July to November 2022
 6

BENEFITS OF
SAMPLING
 Cost-
Effectiveness.
 Time Efficiency.
 Practicality and
Accessibility.
 Manageability.
 Less Disruptive.
 Detailed Data.

MKT0019 Algonquin College

 SAMPLING THEORY BASICS | DEFINITIONS

 The Population is the entire which the actual sample is

group of individuals or entities drawn.
that are relevant to your  Sampling Unit: A sampling unit
research objectives. is a non-overlapping collection of
 Sampling Element: A sampling one or more elements from the
element is the individual target population.
member of the target  Sampling Time Frame: This
population. term refers to the specific period
 Sampling Frame: The sampling during which the data is
frame is a list or database from collected. 7

MKT0019 Algonquin College

 SAMPLING FRAME
| DEFINITION
 Essentially the source
material or database from
which a sample is drawn.
 The sampling frame
should ideally encompass
the entire target
population of the study.
 Telephone Directory.
 Membership List.
 Subscription List.
 Census Data.

MKT0019 Algonquin College

 EXAMPLE| AUTO

Mazda executives need to understand the profile and

purchasing patterns of new Mazda buyers to make strategic
decisions.

 Population: This is the whole group Mazda is interested in studying. In

This Photo by Unknown Author is licensed under CC BY-SA this case, the target population would be all adult purchasers of
automobiles.
 Sampling Frame: This is a list or database from which Mazda's market
researchers can draw their sample. The sampling frame should be as
comprehensive as possible to ensure all new Mazda purchasers within
the nine-month period are represented. It could be a list of all adults who
have purchased a Mazda at all dealerships during this time.
 Sampling Unit: The sampling unit refers to the basic level of
investigation. For Mazda, each new Mazda purchaser is a sampling unit -
that is, individual buyers are the ones from whom data will be collected.
 Sampling Time Frame: This is the period during which the sampling is
conducted. Mazda's sampling time frame is the first nine months of the
year. This means the study will only consider the purchases made within
this period.
 Sampling Element: This is the subject on which the measurement is
taken. In Mazda's research, the sampling element would be the
individual adult purchaser of a new Mazda. This is the entity from which 9
MKT0019 Algonquin College
This Photo by Unknown Author is licensed under CC BY-SA
data will be collected for the study.
 UNIT OR
ELEMENT?
 Sampling elements are
what or who is being
studied.
 Sampling unit is how the
study breaks down the
population for sampling.
 If Mazda is looking at
individuals, the two are
the same.
 If Mazda is looking at
something broader (like
households, dealerships,
or transactions), the
sampling units may Sampling Sampling Unit
encompass multiple
elements.
Element (e.g.
MKT0019 Algonquin College
Households) 10
 EXAMPLE | BANKS
The executives at the retail banking services are tasked with
developing a strategy to enhance the banking experience for
households within a specific geographical area.
 Target Population: Here, the target population would be all
households with checking accounts.
 Sampling Frame: For this retail banking study, the sampling
frame would be a list or database of all households with
checking accounts within the bank's service area, which in
this case is specified as a 10-mile radius of Charlotte, NC.
 Sampling Unit: In the given example, each household within
a 10-mile radius of Charlotte, NC, that has a checking account
with the bank would be a sampling unit.
 Sampling Time Frame: For this study, the sampling time
frame is the first four months of the year.
 Sampling Element: The sampling element is the smallest
unit about which or from which data could be collected. In this
example, if data is collected about each individual checking
account (rather than the household as a whole), then each
11
MKT0019 Algonquin College checking account would be considered a sampling element.
This Photo by Unknown Author is licensed under CC BY-NC
 THE CENSUS

 A census is a research study that

includes data about every member of the
defined target population.
 Sampling is quicker and less costly than
a census.
 In market research, a census provides a
complete and accurate picture of the
market or target group, allowing for more
precise analysis and decision-making.
This Photo by Unknown Author is licensed under CC BY-SA
 The Canadian Census is administered by
Statistics Canada every five years., and
covers various topics, including age, sex,
marital status, language, ethnicity, labor,
education, income, and more.

MKT0019 Algonquin College

 SAMPLING | PROBABILITY

 Each sampling unit in the defined target

population has a known probability of The sample should accurately reflect the
being selected for the sample. larger group it's drawn from. For example,
 Rules ensure the following: if 50% of the population is female, then
 Unbiased selection of the sampling units. roughly 50% of the sample should be
 Proper sample representation of the defined female too.
target population.
 Results using probability sampling designs
can be generalized to the target population.

MKT0019 Algonquin College

 WHAT DOES KNOWN PROBABILITY MEAN?

 Equal Probability  Proportional Probability

 In simple random sampling, every individual  In some cases, the probability is not equal
has the same probability of being selected. but is known and is based on the design of
 For example, if there are 100 people in a the sampling method.
population and you want a sample of 10,  For example, in stratified sampling, the
each person has a 10% chance of being population is divided into strata (e.g., by age
chosen. groups), and individuals are randomly
selected in proportion to the strata's size in
the population.

14

MKT0019 Algonquin College

 PROBABILITY SAMPLING | TYPES

• Simple Random Sampling:  Cluster Sampling: The

Every individual has an equal population is divided into
chance of being selected. E.g., clusters (usually
drawing names from a hat. geographically), and a random
• Stratified Random Sampling: sample of clusters is chosen.
Population is divided into Then, all or a random sample of
subgroups (strata) based on a members from selected clusters
characteristic, and samples are are surveyed.
drawn proportionally from each
stratum.

MKT0019 Algonquin College

 PROBABILITY SAMPLING | TYPES

Simple
Random
16
Sampling Cluster
Stratified
Sampling Sampling

MKT0019 Algonquin College

 17

SAMPLING |
NON-
PROBABILITY
 The probability of each
sampling unit in the
defined target population
to be selected is NOT
known.
 Selection of sampling unit
is based on researcher
judgment and may or may
not represent the target
population.
 Advantages: cheaper and
faster to execute than
probability sampling.

MKT0019 Algonquin College

 t
ec
NON-PROBABILITY SAMPLING | TYPES

oj
Pr
r
Fo
Convenience Sampling

 Selected groups by personal judgment Judgement Sampling

 Decreases chance of error Quota Sampling
 Reliability and Validity increases, but data is not
statistically representative of defined target Snowball Sampling
population

MKT0019 Algonquin College

 Draws samples at the convenience of the
researcher.

Enables many respondents to be interviewed in a

short time.

Commonly used in the early stages of research and

pretesting of questionnaires.

The data is not generalizable to the defined target

population.

NON-PROBABILITY SAMPLING | CONVENIENCE

SAMPLING
MKT0019 Algonquin College
 Respondents are selected because the researcher believes
they meet the requirements of the study.

If researcher judgment is correct, the sample generated by

judgment sampling will be better than one generated by
convenience sampling.

NON-PROBABILITY SAMPLING | JUDGEMENT SAMPLING

MKT0019 Algonquin College
 Demographic characteristics,
Involves the
selection of
prospective
participants
according to Specific attitudes,
prespecifie
d quotas for
either:
Behaviours.

NON-PROBABILITY SAMPLING | QUOTA SAMPLING

MKT0019 Algonquin College
 TYPES OF POTENTIAL ERRORS IN SAMPLING

• Sampling Error: Any difference between the sample value and the true values of a population.

• Non-Sampling Error: Errors not related to sampling, like respondent errors or survey design

errors etc.

• Non-Coverage Error: Missing a region or characteristic (see types of behaviours)

• Non-response Error: Not having people take your survey/questionnaire/focus group request, or

• Respondent Errors: Responses or non-responses. Respondents not available or either by intention or

without intention provide incorrect response.

22
• Data Entry or Office Error: Survey design, incorrect tabulation, not following guidelines of market research.
MKT0019 Algonquin College
 AL
GO
Reliable BUT Valid BUT Neither Valid Both Valid &
Not Valid Not Reliable Nor Reliable Reliable

RELIABLE VS VALID

MKT0019 Algonquin College

 24

VALIDITY

VALID: Red bullseye

 Accurately
measuring what
you want to
measure.
 Decision Problem
and Research
Problems

MKT0019 Algonquin College

 25

RELIABILITY

Reliable: Consistent
black dots
 Same results over
and over .
 Asking the RIGHT
questions, the
BEST way to the
proper sample.

MKT0019 Algonquin College

 SAMPLING AND SECONDARY DATA
• Secondary Data refers to data that was collected by
someone else or for a different purpose.
• It includes sources like government reports, previous
market research, historical sales data, etc. Secondary
data can be valuable, saving time and resources.
• However, it must be scrutinized for relevance and
reliability before being used.
• Is the data collected in the right way for our needs?
• Can we trust the people who gathered this data
before? Are they competent and reputable?
• Do we have enough details about how they did their
research?
• When was the data collected, and did anything special
happen at that time which might affect it? 26

MKT0019 Algonquin College

 1 2 3 4 5
Define Select the Identify Select Calculate
your Data Sampling Sampling Sampling
Collection Frame Method/Tech Size.
Target Method. sources. niques.
Population.

STEPS FOR DRAWING A SAMPLING PLAN

A Sampling Plan Helps Ensure That Data Is Representative Of The Total Target Population 27
MKT0019 Algonquin College
 28

MKT0019 Algonquin College

 KEY TAKEAWAYS!

1. 6.

2. 7.

3. 8.

4. 9.

5. 10.

MKT0019 Algonquin College

MKT0019 Algonquin College