McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
CHAPTER 13
Basic Sampling Issues
LEARNING OBJECTIVES
1. To understand the concept of sampling.
3. To understand the concepts of sampling error and nonsampling error.
5. To understand sampling implications of surveying over the Internet.
KEY TERMS
Sampling
Population
Census
Sample
Sampling frame
Random-digit dialing
Probability samples
Simple random sample
Systematic sampling
Stratified sample
Proportional allocation
Disproportional, or optimal, allocation
Cluster sample
Multistage area sampling
CHAPTER SCAN
This chapter introduces the basic issues of sampling. The vocabulary of sampling is discussed,
and an understanding of the terms population, sample, and census are explained. The steps in
developing a sampling plan are introduced next. The first step is to define the population of
interest. This step can cause the project to be invalid if determined improperly. Next a data
collection method is chosen. These were discussed in chapter six. A sampling frame is chosen
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
CHAPTER OUTLINE
1. Concept of Sampling
I. Sampling
2. Developing a Sampling Plan
I. Step One: Define the Population of Interest
A. Basic Issues in Developing a Sampling Plan
II. Step Two: Choose Data Collection Method
3. Sampling and Nonsampling Errors
4. Probability Sampling Methods
I. Simple Random Sampling
A. Simple Random Sampling Defined
II. Systematic Sampling
5. Nonprobability Sampling Methods
6. Internet Sampling
7. Summary
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
CHAPTER SUMMARY
1. CONCEPT OF SAMPLING
I. Sampling
A. Sampling Defined–process of obtaining information from a subset (a sample) of a larger
group (the universe or population).
1. Motivation for sampling –to make these estimates more quickly and at a lower cost than
would be possible by any other means
3. Critical–selected in a scientific manner, which ensures that the sample is representative– a
1. Population of interest is the total group of people from whom we need to obtain information
a. Defining the population of interest is the first step in the sampling process
b. No specific rules–requires good logic and judgment
III. Sample versus Census
A. Census Defined–data are obtained from or about every member of the population of interest.
1. Not often employed in marketing research
3. May be appropriate and feasible in settings where a firm has only a small number of customers
5. Census does not necessarily provide more accurate results than a sample
B. Sample Defined–a subset of all the members of a population.
2. DEVELOPING A SAMPLING PLAN
See Exhibit 13.1 Developing a Sampling Plan (p 381)
I. Step One: Define the Population of Interest
A. Basic Issues in Developing a Sampling Plan
1. Specified in terms of some combination of the following characteristics:
a. Geography characteristics
b. Demographic characteristics
c. Product or service use characteristics
d. Awareness measures
See Exhibit 13.2 Some Bases for Defining the Population of Interest (p 383)
See Practicing Marketing Research: Driver’s Licenses and Voter Registration Lists as
Sampling Frames (p 382)
Medical researchers at the University of North Carolina at Chapel Hill wanted to provide the
most representative sampling frame for a population-based study of the spread of HIV among
heterosexual African Americans living in eight rural North Carolina counties. They found that
the list of driver’s licenses for men and women aged 18 to 59 gave them the “best coverage” and
Questions:
1. What kinds of usable data could a statistical analysis of driver’s license lists generate, and how
would you go about the study?
2. Identify two other market research categories in which driver’s license lists would excel in
3. Define the characteristics of individual who should be excluded
a. Security Question–people who work in the industries in question are viewed as security risks
See Exhibit 13.3 Example of Screening Question Sequence to Determine Population
Membership (p 383)
II. Step Two: Choose a Data Collection Method
A. Data Collection Methods
1. Selection of a data-collection method has implications for the sampling process
3. Telephone surveys have a less significant problem with no response, but suffer from call
screening technologies and reduced access to individuals with mobile phones only.
4. Internet surveys have problems with professional respondents (discussed in Chapter 7) and
web respondents may not be representative of the target population.
III. Step Three: Identify a Sampling Frame
A. Sampling Frame
1. Sampling frame defined–a list of the members or elements of the population from which
1) Telephone survey sample frame might be telephone book.
a) Listed numbers and those with no listing are significantly different in several very important
ways
b) Unlisted numbers are more prevalent in the West, metropolitan areas, and among nonwhites
and those in the 18-to-34 age group
c) Random digit dialing–a method of generating lists of telephone numbers at random
d) Details on the way such companies draw their samples can be found at
www.sssisamples.com/random_digit.html
See Practicing Marketing Research: How to Achieve Near Full Coverage for Your Sample
Using Address-Based Sampling (p 385)
Address-Based Sampling (ABS) offers potential benefits in comparison to a strictly telephone
based method of contact. Landlines offer access to only about 75 percent of U.S. households, and
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
contacting people via wireless devices can be a complicated process. Market research firm
Survey Sampling International (SSI), however, has found that using an ABS approach can
almost completely fill that access gap.
Questions:
1. Can you think of any demographic segments that might still be difficult to reach via ABS?
2. What are some ways researchers could use to mitigate the increased costs of mail surveys?
IV. Step Four: Select a Sampling Method
A. Selection of Sampling Method
1. The selection will depend on:
a. Objectives of the study
See Exhibit 13.4 Classification of Sampling Methods (p 388)
2. Major alternative sampling methods
a. Probability Samples–every element of the population has a known, nonzero probability of
selection
1) Simple random sample is the best known of probability sample approaches
a) Researcher must adhere to precise selection procedures.
2) Nonprobability Samples–specific elements from the population are selected in a nonrandom
manner.
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
a) Nonrandomness–when population elements are selected on the basis of convenience–because
they are easy or inexpensive to reach
b) Purposeful nonrandomness–occurs when a sampling plan systematically excludes or over-
represents certain subsets of the population
3) Advantages of Probability Samples
4) Disadvantages of Probability Samples
6) Representativeness of the sample is not known
7) Results cannot and should not be projected to population
b. Advantages of Nonprobability Samples
1) Cost less than probability samples
3) Are reasonably representative if executed in a reasonable manner
See Practicing Marketing Research: Can a Single Online Respondent Pool Offer a Truly
Representative Sample? (p 387)
Online research programs can often benefit by building samples from multiple respondent pools.
Achieving a truly representative sample is a difficult process for many reasons. When drawing
from a single source, even if researchers were to use various verification methods, demographic
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
Questions:
1. If one respondent pool is not sufficient, how many do you think you would have to draw from
to get a truly representative sample? Why do you think that?
2. When creating a sample, how would you propose accounting for the types of extrinsic
characteristics mentioned?
V. Step Five: Determine Sample Size
A. Determine the Appropriate Sample Size (Discussed more in depth in Chapter 14)
1. Nonprobability Samples–rely on such factors as:
2. Probability Samples–use formulas to calculate the sample size required
a. Acceptable error (the difference between sample result and population value)
b. Levels of confidence (the likelihood that the confidence interval–sample result plus or minus
the acceptable error–will take in the true population value).
c. The ability to make statistical inferences about population values based on sample results is a
major advantage of probability samples
VI. Step Six: Develop Operational Procedures for Selecting Sample Elements
A. Operational Procedures
1. Selecting elements in the data-collection—developed and specified whether a probability or
nonprobability sample is used.
3. Failure to develop a proper operational plan–jeopardize the entire sampling process.
1. Requires adequate checking to ensure that specified procedures are followed
See Exhibit 13.5 Example of Operational Sampling Plan (p 389)
3. SAMPLING AND NONSAMPLING ERRORS
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
I. Errors
A. Population Parameter–a value that defines a true characteristic of a total population (e.g.,
the average gross income of the population).
2. Instead we take a sample, which is an estimate of the population, to make inferences about the
population.
B. Accuracy of Sample–affected by two general types of error. The following formula
1. Sampling error–when the sample selected is not perfectly representative of the population.
a. Administrative error–relates to the problems in the execution of the sample, such as flaws in
design or execution.
b. Random sampling error—due to chance and cannot be avoided. Note: under the rarest of
cases will the sample estimate exactly equal the population parameter—hence there is always
2. Measurement or nonsampling error–includes everything other than sampling error that can
cause inaccuracy and bias.
See From the Front Line: Sampling and Data Collection with Internet Panels (p 390)
1. First you start with getting bids.2. The next item to consider is your previous experience with
each panel.
4. If you are fortunate enough to have more than one panel that can meet your quota
5. At the completion of the data collection phase, you may need to get data from a third party
4. PROBABILITY SAMPLING METHODS
I. Simple Random Sampling
A. Simple Random Sampling Defined–the purest form of probability sampling.
For a simple random sample, the known and equal probability is computed as follows:
Formula for the known and equal probability:
Probability of selection = Sample size ÷ Population size
1. If a sampling frame (listing of all the elements of the population) is available, the research can
select a simple random sample as follows:
See Exhibit 1, Appendix 3 “Statistical Tables” (p A-26)
2. Simple random samples can be obtained through the use of:
a. Random digit dialing
b. Computer files such as customer lists
II. Systematic Sampling
A. Systematic Sampling–probability sample in which the entire population is numbered and
elements are selected using a skip interval
B. Steps:
1. Obtain a listing of the population and number the entire population
3. Select names based on skip interval, using a random starting point.
5. Greatest danger is in listing of population. Some populations may contain hidden patterns that
the researcher may inadvertently pull into the sample.
McDaniel & Gates – Marketing Research, 9th Edition Instructor’s Manual
III. Stratified Samples
A. Stratified samples–probability samples that are distinguished by the following procedural
steps.
1. The original or parent population is divided into two or more mutually exclusive and
exhaustive subsets.
3. Subsets–should be related to the characteristics of the population we are really interested in
measuring.
4. Stratified samples–used over simple random samples because of their potential for greater
statistical efficiency. This exists because one source of variation has been eliminated.
B. Three Steps in Implementing a Proper Stratified Sample
1. Identify salient demographic or classification factors–that are correlated with the behavior
of interest.
2. Determine what proportions of the population fall into the various subgroups under each
stratum
a. Proportional allocation–number selected is directly proportional to the size of the stratum in
relation to the size of the population. The proportion of elements from each stratum is
b. Disproportional or optimal allocation–double weighting based on relative size of the
stratum and the standard deviation of the distribution of the characteristic under consideration for
all elements in the stratum.
3. Select separate simple random samples from each stratum
C. Why Stratified Samples Are Not Used All the Time
2. The time or costs of stratification may not be warranted from a cost versus value of
information perspective
IV. Cluster Sampling
A. Cluster Sampling Defined–probability sample in which the sampling units are selected from
a number of small geographic areas to reduce data collection cost