Transcript Chapter 16 - Exploring Marketing Research

MR2300: MARKETING RESEARCH
PAUL TILLEY
Unit 9: Sampling Designs, Sampling
Procedures & Sample Size.
IN THIS VIDEO WE WILL:
1.
Define a sample; a population; a population element and a census
2.
Explain why researchers use samples.
3.
Design an appropriate sample.
4.
Use appropriate statistical tools to extract a useful sample from a population.
5.
Identify the key concepts in a sampling plan
6.
Control for errors that can occur in sampling
7.
Illustrate the distinctive features of probability and non-probability samples
8.
Calculate and interpret the Mean, Median, Mode and Standard Deviation of data.
9.
Develop frequency distributions for data
10. Calculate
sample size and the sample size of a proportion.
POPULATION

Any complete group

Usually people
Target Population: Canada? NL?
Target Population
Cda Population= 35,540,400
NL Population= 526,977
CENSUS

Investigation of all individual elements that make up a population

difficult, slow and very expensive to measure
SAMPLE

A sample is a subset of a larger target population

The Sampling process involves drawing conclusions about an entire
population by taking a measurement from only a portion of all the
population elements

Taking samples of populations is easier, faster and cheaper than
taking a census of the population. Sample size relative to the
population size will determine how accurately the sample results will
mirror the population results. The difference is known as Error.

Samples may have to be used if testing results in destruction of the
test unit
Stages in the
Selection
of a Sample
Define the target population
Select a sampling frame
Determine if a probability or nonprobability
sampling method will be chosen
Plan procedure
for selecting sampling units
Determine sample size
Select actual sampling units
Conduct fieldwork
SAMPLING FRAME

A list of elements from which the sample may be drawn

Working population

Mailing lists - data base marketers

Sampling frame error
SAMPLING UNITS

Group selected for the sample

Primary Sampling Units (PSU)

Secondary Sampling Units

Tertiary Sampling Units
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
SYSTEMATIC ERRORS

Nonsampling errors

Unrepresentative sample results

Not due to chance

Due to study design or imperfections in execution
ERRORS ASSOCIATED WITH SAMPLING

Sampling frame error

Random sampling error

Nonresponse error
TWO MAJOR CATEGORIES OF
SAMPLING

Nonprobability sampling

Probability of selecting any particular member is
unknown
 Convenience
 Judgment
 Quota
Sample
Sample
 Snowball

Sample
Sample
Probability sampling

Known, nonzero probability for every element
 Simple
Random Sample
 Stratified
 Cluster
Sample
Sample
 Multistage
Area Sample
NONPROBABILITY SAMPLING

Convenience Sampling - (also called haphazard or accidental sampling) refers to
the sampling procedure of obtaining the people who are most conveniently
available.

Judgment - is a nonprobability technique in which an experienced individual
selects the sample upon his or her judgment about some appropriate characteristic
required of the sample members

Quota - In quota sampling, the interviewer has a quota to achieve. to ensure that
the various subgroups in a population are represented on pertinent sample
characteristics to the exact extent that the investigators desire.

Snowball - refers to a variety of procedures in which initial respondents are selected
by probability methods, but additional respondents are then obtained from
information provided by the initial respondents. This technique is used to locate
members of rare populations by referrals.
PROBABILITY SAMPLING

Simple random sample

Systematic sample

Stratified sample

Cluster sample

Multistage area sample
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 simple random sample of 10 students is to be selected from a
class of 50 students. Using a list of all 50 students, each
student is given a number (1 to 50), and these numbers are
written on small pieces of paper. All the 50 papers are put in a
box, after which the box is shaken vigorously to ensure
randomisation. Then, 10 papers are taken out of the box, and
the numbers are recorded. The students belonging to these
numbers will constitute the simple random sample.
SYSTEMATIC SAMPLING

A simple process

Every nth name from the list will be drawn
Systematic sampling works well when the
individuals are already lined up in order. In the
past, students have often used this method when
asked to survey a random sample of CNA
students. Since we don't have access to the
complete list, just stand at a corner and pick
every 3rd person walking by.
STRATIFIED SAMPLING

Probability sample

Subsamples are drawn within different strata

Each stratum is more or less equal on some characteristic

Do not confuse with quota sample
One easy example using a stratified technique
would be a sampling of people at CNA. To
make sure that a sufficient number of students,
faculty, and staff are selected, we would stratify
all individuals by their status - students, faculty,
or staff. (These are the strata.) Then, a
proportional number of individuals would be
selected from each group.
CLUSTER SAMPLING

The purpose of cluster sampling is to sample economically while retaining the
characteristics of a probability sample.

The primary sampling unit is no longer the individual element in the population

The primary sampling unit is a larger cluster of elements located in proximity to
one another
Suppose your company makes light bulbs, and you'd
like to test the effectiveness of the packaging. You
don't have a complete list, so simple random
sampling doesn't apply, and the bulbs are already in
boxes, so you can't order them to use systematic. And
all the bulbs are essentially the same, so there aren't
any characteristics with which to stratify them.
To use cluster sampling, a quality control inspector
might select a certain number of entire boxes of bulbs
and test each bulb within those boxes. In this case,
the boxes are the clusters.
WHAT IS THE
APPROPRIATE SAMPLE DESIGN?

Degree of accuracy

Resources

Time

Advanced knowledge of the population

National versus local

Need for statistical analysis
AFTER THE SAMPLE DESIGN
IS SELECTED

Determine sample size

Select actual sample units

Conduct fieldwork
SAMPLE STATISTICS

Variables in a sample

Measures computed from data

English letters for notation
FREQUENCY DISTRIBUTION OF
DEPOSITS
Amount
less than $3,000
$3,000 - $4,999
$5,000 - $9,999
$10,000 - $14,999
$15,000 or more
Frequency (number of
people making deposits
in each range)
499
530
562
718
811
3,120
PERCENTAGE DISTRIBUTION OF
AMOUNTS OF DEPOSITS
Amount
less than $3,000
$3,000 - $4,999
$5,000 - $9,999
$10,000 - $14,999
$15,000 or more
Percent
16
17
18
23
26
100
MEASURES OF CENTRAL TENDENCY

Mean - arithmetic average

µ, Population;
, sample

Median - midpoint of the distribution
X

Mode - the value that occurs most often
NUMBER OF SALES CALLS PER DAY BY
SALESPERSONS
Salesperson
Mike
Patty
Billie
Bob
John
Frank
Chuck
Samantha
Number of
Sales calls
4
3
2
5
3
3
1
5
26
MEASURES OF DISPERSION
OR SPREAD
Range
- the distance between the smallest and the
largest value in the set.
Variance
- measures how far a set of numbers is
Standard
variance
deviation - square root of the
spread out.
THE NORMAL DISTRIBUTION

Normal curve

Bell shaped

Almost all of its values are within plus or minus 3 standard
deviations

I.Q. is an example
NORMAL DISTRIBUTION
13.59%
2.14%
34.13%
34.13%
13.59%
2.14%
INGREDIENTS IN DETERMINING SAMPLE SIZE
 Estimated
standard
deviation of population
 Magnitude
of acceptable
sample error
 Confidence
level
SAMPLE SIZE CALCULATION FOR
QUESTIONS INVOLVING MEANS
Where:
n = Number of items in samples
Z = Standard Deviation Confidence interval
S = Standard Deviation Estimate for Population
E = Acceptable error
Z
S
n=
E
2
SAMPLE SIZE CALCULATION FOR A
PROPORTION
Where:
n = Number of items in samples
Z2 = The square of the confidence interval
in standard error units.
p = Estimated proportion of success
q = (1-p) or estimated the proportion of failures
E2 = The square of the maximum allowance for error
between the true proportion and sample proportion
or zsp squared.
z2pq
n=
2
E