Sampling Designs & Sampling

Procedures
 • A sample is a subset, or some part, of a larger
population. The purpose of sampling is to estimate an
unknown characteristic of a population.
• A population (universe) is any complete group—for
example, of people, sales territories, stores, or college
students—that shares some common set of
characteristics.

Sampling • The term population element refers to an individual

member of the population.
Terminology • A census is an investigation of all the individual
elements that make up the population—a total
enumeration rather than a sample
• Sampling cuts costs, reduces labor requirements and
gathers vital information quickly
• Most properly selected samples give results that are
reasonably accurate
 Photograph 1 Photograph
Portrait of young 2
man 2,000 dots

Photograph Photograph
3 4
1,000 dots 250 dots
Source: Adapted with permission from A. D. Fletcher and T. A. Bowers, Fundamentals of
Advertising Research
(Columbus, OH: Grid Publishing, 1983), pp. 60–61.
 • Defining the Target Population
• The crucial characteristics of the population
• The question to answer is, “Whom do we want to talk to?” The answer may be users,
nonusers, recent adopters, or brand switchers
• Tangible characteristics should be used to define the population. Child bearing ladies

Practical • The Sampling Frame

• A list of elements from which the sample may be drawn – also called working
Sampling population – Medical Association, Bar Council
• Some firms, called sampling services or list brokers, specialize in providing lists or

Concepts databases that include the names, addresses, phone numbers, and e-mail addresses of
specific populations
• Lists are compiled from subscriptions to professional journals, credit card
applications, warranty card registrations, and a variety of other sources.
• A reverse directory provides, in a different format, the same information contained
in a telephone directory. Listings may be by city and street address or by phone
number, rather than alphabetical by last name - useful when a research wishes to
survey only a certain geographical area
 • A sampling frame error occurs when certain
sample elements are excluded or when the entire
population is not accurately represented in the
sampling frame
• Example - Election polling that used a telephone
Practical directory as a sampling frame would be contacting
households with listed phone numbers, not
Sampling households whose members are likely to vote - A
better sampling frame might be voter registration
Concepts records
• Another potential sampling frame error involving
phone records is the possibility that a phone survey
could underrepresent people with disabilities -
hearing and speech impairments, might make
telephone use impossible.
Practical Sampling Concepts
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
 • The sampling unit is a single element or
group of elements subject to selection in the
sample.
Sampling • A unit selected in the first stage of sampling
is called a primary sampling unit (PSU).
Units • A unit selected in a successive stages of
sampling is called a secondary sampling
unit or (if three stages are necessary)
tertiary sampling unit
 • Random sampling error is the difference between the sample
result and the result of a census conducted using identical
procedures.
• Occurs because of chance variation in the scientific selection of
sampling units.
Random • A function of sample size. As sample size increases, random

sampling &
sampling error decreases
• Systematic Sampling Error - result from non-sampling

non- factors, primarily the nature of a study’s design and the

correctness of execution. These errors are not due to chance

sampling
fluctuations
• Example - highly educated respondents are more likely to

errors cooperate with mail surveys than poorly educated ones

• Less Than Perfectly Representative Samples
• Random sampling errors and systematic errors associated with
the sampling process may combine to yield a sample that is less
than perfectly representative of the population
Errors associated with Sampling

Planned Respondent
Total population Sampling frame s
sample
(actual
sample)

Nonrespon
se
Random error
sampling
Sampling error
frame error

Source: Adapted from Cox, Keith K. and Ben M. Enis, The Marketing Research Process (Pacific Palisades, CA: Goodyear, 1972); and Bellenger, Danny
N. and Barnet A. Greenberg, Marketing Research: A Management Information Approach (Homewood, IL: Richard D. Irwin, 1978), pp. 154–155.
 • Probability sampling - every element in the
population has a known, nonzero probability
of selection.
• Nonprobability sampling, the probability of
any particular member of the population
Probability vs being chosen is unknown - units of the
Nonprobability sample are selected on the basis of personal
judgment or convenience
Sampling
• Technically, no appropriate statistical
techniques exist for measuring random
sampling error from a nonprobability sample
• Nonprobability samples best suited for a
specific researcher purpose
 • Convenience sampling refers to sampling by obtaining
people or units that are conveniently available – shopping
malls, persons on streets during elections, internet surveys.
• Respondents may not be representative
• Judgment (purposive) sampling is a nonprobability sampling
technique in which an experienced individual selects the
sample based on his or her judgment about some appropriate
Nonprobability characteristics required of the sample member - consumer price
index (CPI) is based on a judgment sample of market-basket
Sampling items, housing costs, and other selected goods and services
expected
• Test-market cities - viewed as typical cities whose demographic
profiles closely match the national profile. A fashion
manufacturer regularly selects a sample of key accounts that it
believes are capable of providing information needed to predict
what may sell in the fall/spring
• Used in elections forecast results
 • Quota Sampling - A nonprobability sampling procedure that
ensures that various subgroups of a population will be represented
on pertinent characteristics to the exact extent that the investigator
desires
• 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. The interviewer is responsible for
finding enough people to meet the quota
• The haphazard selection of subjects may introduce bias - prefer to
Nonprobability interview people who are similar to themselves.

Sampling • Tend to include people who are easily found, willing to be

interviewed, and middle class
• Concentrate their interviewing in areas with heavy pedestrian
traffic such as downtowns, shopping malls, and college campuses
• Advantages - 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 – set a
higher quota for that group
 • Snowball Sampling - A sampling procedure
in which initial respondents are selected by
probability methods and additional
respondents are obtained from information
provided by the initial respondents
Nonprobability • Used to locate members of rare populations
Sampling by referrals – players of Golf with specialist
equipment
• Small sample size and costs are less
• Bias can be an issue
 Nonprobability Samples Cost and Degree of Use
Advantages
Description Disadvantages

1. Convenience: The researcher Very low cost, extensively No need for list of Unrepresentative samples likely;
uses the most convenient sample used population random sampling error estimates
or economical sample units. cannot be made; projecting data
beyond sample is relatively risky
2. Judgment: An expert or Moderate cost, average use Useful for certain types Bias due to expert’s beliefs may
experienced researcher selects the of forecasting; sample make sample unrepresentative;
sample to fulfill a purpose, such as guaranteed to meet a projecting data beyond sample
ensuring that all members have a specific objective is risky
certain characteristic.
3. Quota: The researcher classifies Moderate cost, very Introduces some stratification Introduces bias in researcher’s
the population by pertinent extensively used of population; requires no list classification of subjects;
properties, determines the desired of population nonrandom selection within classes
proportion to sample from each means error from population
class, and fixes quotas for each cannot be estimated; projecting
interviewer. data beyond sample is risky
4. Snowball: Initial respondents are Low cost, used in special Useful in locating members High bias because sample units
selected by probability samples; situations of rare populations are not independent; projecting
additional respondents are data beyond sample is risky
obtained by referral from initial
respondents.
 • Simple Random Sampling - ensures each element in the
population will have an equal chance of being included in the
sample
• Systematic Sampling - A sampling procedure in which a
starting point is selected by a random process and then
every nth number on the list is selected.
• While systematic sampling is not actually a random
selection procedure, it does yield random results if the
arrangement of the items in the list is random in character

Probability • Stratified Sampling - A probability sampling procedure in

which simple random subsamples that are more or less equal

Sampling
on some characteristic are drawn from within each stratum
of the population
• In this, simple random subsamples that are more or less
equal on some characteristic are drawn from within each
stratum
• Urban and rural groups have widely different attitudes
toward energy conservation, but members within each
group hold very similar attitudes - Random sampling error
will be reduced
• Will accurately reflect the population on the basis of the
criterion or criteria used for stratification
Stratified
Sampling
• Proportional stratified sample - Percentage Proportion Disproportio
the number of sampling units drawn in
Population
al
Sample
nal
Sample

from each stratum is in proportion Warehouse

Clubs
20% 20%

50%

to the relative population size of the Chain 57% 57%

Stores
stratum 38%

• Disproportional stratified sample Small

Independents 23% 23%
12%

- the sample size for each stratum is

not allocated in proportion to the
population size but is dictated by
analytical considerations, such as
variability in store sales volume
 Percentage Proportional Disproportional
in Sample Sample
Population

20% 20%
Warehouse
Clubs
50%

Chain 57% 57%

Stores

38%
Small
Independents 23% 23%
12%
 • Cluster Sampling - An economically efficient sampling
technique in which the primary sampling unit is not the
individual element in the population but a large cluster of
elements; clusters are selected randomly
• Used when lists of the sample population are not
available
• Researchers investigating employees and self-employed
Probability workers for a downtown revitalization project found that
a comprehensive list of these people was not available,
Sampling they decided to take a cluster sample, selecting
organizations (business and government) as the clusters. A
sample of firms within the central business district was
developed, using stratified probability sampling to identify
clusters. Next, individual workers within the firms
(clusters) were randomly selected and interviewed
concerning the central business district.
• A cluster should be as heterogeneous as the population
itself—a mirror image of the population.
Population Element Possible Clusters in the United States

U.S. adult population States

Counties
Metropolitan Statistical Areas
Census Tracts
Blocks
Households
College seniors Colleges
Manufacturing firms Counties
Metropolitan Statistical Areas
Localities
Plants
Airline travelers Airports
Planes
Sports fans Football Stadiums
Basketball Arenas
Baseball Parks
 • Multistage Area Sampling - involves two or more
steps that combine some of the probability
techniques
• Geographic areas are randomly selected in
progressively smaller (lower-population) units
• A political pollster investigating an election in MP
Probability might first choose Districts within the state to

Sampling
ensure that the different areas are represented in the
sample. In the second step, precincts within the
selected districts may be chosen. As a final step, the
pollster may select blocks (or households) within the
precincts, then interview all the blocks (or
households) within the geographic area.
• Researchers may take as many steps as necessary to
achieve a representative sample
 Probability Samples Cost and Degree of Use
Advantages
Description Disadvantages

1. Simple random: The researcher High cost, moderately used Only minimal advance knowledge Requires sampling frame to work
assigns each member of the in practice (most common in of population needed; easy to from; does not use knowledge
sampling frame a number, then random digit dialing and analyze data and compute error of population that researcher
selects sample units by random with computerized sampling may have; larger errors for same
method. frames) sampling size than in stratified
sampling; respondents may be
widely dispersed, hence cost may
be higher
2. Systematic: The researcher uses Moderate cost, moderately Simple to draw sample; easy If sampling interval is related
natural ordering or the order used to check to periodic ordering of the
of the sampling frame, selects population, may introduce
an arbitrary starting point, then increased variability
selects items at a preselected
interval.
3. Stratified: The researcher divides High cost, moderately used Ensures representation of Requires accurate information
the population into groups and all groups in sample; on proportion in each stratum;
randomly selects subsamples from characteristics of each stratum can if stratified lists are not already
each group. Variations include be estimated and comparisons available, they can be costly to
proportional, disproportional, and made; reduces variability for same prepare
optimal allocation of subsample sample size
sizes.
4. Cluster: The researcher selects Low cost, frequently used If clusters geographically defined, Larger error for comparable
sampling units at random, then yields lowest field cost; requires size than with other probability
does a complete observation of listing of all clusters, but of samples; researcher must be able
all units or draws a probability individuals only within clusters; to assign population members
sample in the group. can estimate characteristics of to unique cluster or else
clusters as well as of population duplication or omission of
individuals will result
5. Multistage: Progressively smaller High cost, frequently used, Depends on techniques Depends on techniques
areas are selected in each stage by especially in nationwide combined combined
some combination of the first four surveys
techniques.
Determination of Sample Size
• Random Error and Sample Size
• When the standard deviation of the population is unknown, a confidence interval is calculated
using the following formula:

= sample mean
• z = confidence level value
• S = sample standard deviation
• n = sample size
Random Error
& Sample Size

• The equation for the plus or

minus error factor in the
confidence interval includes n,
the sample size:
• E = Z *S/√n
• If n increases, E is reduced
• that increases in sample size
reduce sampling error at a
decreasing rate.
 Factors in Determining
Sample Size for
Questions Involving
Variable Symbol Typical Source of
Information
Means
Standard S Pilot study or rule of
deviation thumb • Three factors are required to specify sample
size:
Magnitude of E Managerial judgment or • (1) the heterogeneity (i.e., variance) of the
error population; Standard Deviation of the
calculation (Z S_)
X population
Confidence Zc.l. Managerial judgment
level
• (2) the magnitude of acceptable error,
confidence interval (i.e., ± some amount);
and
• (3) the confidence level (i.e., 90 percent, 95
percent, 99 percent).
 • Follow three steps:

1. Estimate the standard deviation of the population. – Pilot study -

Thumb Rule – One sixth of the range
Estimating 2. Make a judgment about the allowable magnitude of error. –
Manager’s decision
Sample 3. Determine a confidence level. – Manager’s decision

Size for 4. Sample size formula

Questions n = (Z S/E) 2
Z = standardized value that corresponds to the confidence level
Involving S = sample standard deviation or estimate of the population standard
deviation
Means • E = acceptable magnitude of error, plus or minus error factor (range is
one-half of the total confidence interval)
 • In most cases the size of the population does
not have an effect on the sample size.
• The variance of the population has the largest
The effect on sample size.
Influence of • A finite correction factor may be needed to
adjust a sample size that is more than 5
Population percent of a finite population
Size on • √(N —n)/(N —1)
Sample
• N = population size and n = sample
Size size
 • For estimation of a proportion, the researcher
requires some knowledge of the logic for
determining a confidence interval around a
sample proportion estimation (p) of the
Factors in population proportion (π).
Determining • For a confidence interval to be constructed
Sample Size around the sample proportion (p), an estimate
of the standard error of the proportion (Sp) must
for be calculated and a confidence level specified.
Proportions • The precision of the estimate is indicated by the
value Zc.l.Sp. Thus, the plus-or-minus estimate of
the population proportion is
• Confidence interval = p ± Zc.l.Sp
 • If the researcher selects a 95 percent
probability for the confidence interval, Zc.l.
will equal 1.96.
Factors in • The formula for Sp is
Determining • Sp=√ pq/n or Sp= √p(1- p)/n
• where
Sample Size • Sp = estimate of the standard
for error of the proportion
Proportions • p = proportion of successes
• q = 1 – p, or proportion of
failures
 • To determine sample size for a proportion, the researcher
must make a judgment about confidence level and the
maximum allowance for random sampling error. Furthermore,
the size of the proportion influences random sampling error, so
an estimate of the expected proportion of successes must be
Factors in made, based on intuition or prior information. The formula is
Determining • n = Z 2 c.l. pq/ E 2

Sample Size • n = number of items in sample

• Z 2 c.l. square of the confidence level in standard error units
for • p = estimated proportion of successes
Proportions • q = 1 – p, or estimated proportion of failures
• E2 = square of the maximum allowance for error between the
true proportion and the sample proportion, or Zc.l.Sp squared
 • Selection of the appropriate item, question, or
characteristic to be used for the sample size calculations
• The item that will produce the largest sample size will be
used to determine the ultimate sample size
• The cost of data collection becomes a major
Determining consideration, and judgment must be exercised regarding
the importance of such information
Sample Size
• Another consideration stems from most researchers’ need
on the Basis to analyze various subgroups within the sample
of Judgment • There is a judgmental rule of thumb for selecting
minimum subgroup sample size: Each subgroup to be
separately analyzed should have a minimum of 100 units
in each category of the major breakdowns.
• With this procedure, the total sample size is computed by
totaling the sample sizes necessary for these subgroups
Fieldwork
• Making Initial Contact and Securing the Interview
• PERSONAL INTERVIEWS - carry a letter of identification or an ID card to indicate that the study is a bona fide research project
and not a sales pitch
• TELEPHONE INTERVIEWS
• INTERNET SURVEYS
• GAINING PARTICIPATION
• Asking the Questions
• There are five major rules for asking questions:

1. Ask questions exactly as they are worded in the questionnaire.

2. Read each question very carefully and clearly.
3. Ask the questions in the specified order.
4. Ask every question specified in the questionnaire.
5. Repeat questions that are misunderstood or misinterpreted
 • Probing When No Response Is Given
• Repeating the question
• Using a silent probe
• Repeating the respondent’s reply
• Asking a neutral question
• Recording the Responses
• Some suggestions for recording
Fieldwork open-ended answers include
• Record responses during the interview.
• Use the respondent’s own words.
• Do not summarize or paraphrase the respondent’s answer.
• Include everything that pertains to the question objectives.
• Include all of your probes.
 • Terminating the Interview
• The fieldworker should also answer any
respondent questions concerning the nature
and purpose of the study to the best of his
or her ability
Fieldwork • Extremely important to thank the
respondent for his or her time and
cooperation.
• The fieldworker may be required to
reinterview the respondent at some future
time
 • The Basics
• Have integrity and be honest
• Have patience and tact
• Pay attention to accuracy and detail
Principles of
• Exhibit a real interest in the inquiry at hand,
Good but keep your own opinions to yourself
Interviewing • Be a good listener
• Keep the inquiry and respondents’ responses
confidential
• Respect others’ rights
 • Required Practices
• Complete the number of interviews according to the
sampling plan assigned to you
• Follow the directions provided
• Make every effort to keep schedules
Principles of • Keep control of each interview you do
Good • Complete the questionnaires meticulously
Interviewing • Check over each questionnaire you have completed
• Compare your sample execution and assigned quota
with the total number of questionnaires you have
completed.
• Clear up any questions with the research agency
 • Briefing Session for Experienced
Interviewers
• Training to Avoid Procedural Errors in
Sample Selection
Fieldwork • Supervision of Fieldworkers
Management • Sampling Verification
• Interviewer Cheating
• Verification by Reinterviewing