Chapter 12 Sampling Designs and Sampling Procedures 2007

Chapter 12 Sampling Designs and Sampling Procedures © 2007 Thomson/South-Western. All rights reserved.

LEARNING OUTCOMES After studying this chapter, you should be able to 1. Define sample, population element, and census 2. Explain reasons for taking a sample rather than a complete census 3. Describe the process of identifying a target population and selecting a sampling frame 4. Compare random sampling and systematic (nonsampling) errors 5. Identify the types of nonprobability sampling, including their advantages and disadvantages © 2007 Thomson/South-Western. All rights reserved. 2

LEARNING OUTCOMES (cont’d) After studying this chapter, you should be able to 6. Summarize the advantages and disadvantages of the various types of probability samples 7. Discuss how to choose an appropriate sample design, as well as challenges for Internet sampling © 2007 Thomson/South-Western. All rights reserved. 3

Sampling Terminology • Sample Ø A subset, or some part, of a larger population. • Population (universe) Ø Any complete group of entities that share some common set of characteristics. • Population Element Ø An individual member of a population. • Census Ø An investigation of all the individual elements that make up a population. © 2007 Thomson/South-Western. All rights reserved. 4

Why Sample? • Pragmatic Reasons Ø Budget and time constraints Ø Limited access to total population • Accurate and Reliable Results Ø Samples can yield reasonably accurate information Ø Strong similarities in population elements makes sampling possible Ø Sampling may be more accurate than a census • Destruction of Test Units Ø Sampling reduces the costs of research in finite populations. © 2007 Thomson/South-Western. All rights reserved. 5

Practical Sampling Concepts • Defining the Target Population Ø What is the relevant population? Ø Whom do we want to talk to? v Population is operationally defined by specific and explicit tangible characteristics. • The Sampling Frame (Working Population) Ø A list of target population elements from which a sample may be drawn; also called working population. © 2007 Thomson/South-Western. All rights reserved. 6

Practical Sampling Concepts (cont’d) • Sampling Frame Sources Ø Sampling services (list brokers) v Provide lists or databases of the names, addresses, phone numbers, and e-mail addresses of specific populations. Ø Reverse directory v. A directory similar to a telephone directory except that listings are by city and street address or by phone number rather than alphabetical by last name. • Sampling Frame Error Ø Occurs when certain sample elements are not listed or are not accurately represented in a sampling frame. © 2007 Thomson/South-Western. All rights reserved. 7

EXHIBIT 12. 1 Stages in the Selection of a Sample © 2007 Thomson/South-Western. All rights reserved. 8

EXHIBIT 12. 2 Mailing List Directory Page © 2007 Thomson/South-Western. All rights reserved. 9

Sampling Units • Sampling Unit Ø A single element or group of elements subject to selection in the sample. • Primary Sampling Unit (PSU) Ø A unit selected in the first stage of sampling. • Secondary Sampling Unit Ø A unit selected in the second stage of sampling. • Tertiary Sampling Unit Ø A unit selected in the third stage of sampling. © 2007 Thomson/South-Western. All rights reserved. 10

Random Sampling and Nonsampling Errors • Random Sampling Error Ø The difference between the sample result and the result of a census conducted using identical procedures. Ø A statistical fluctuation that occurs because of chance variations in the elements selected for a sample. • Systematic Sampling Error Ø Systematic (nonsampling) error results from nonsampling factors, primarily the nature of a study’s design and the correctness of execution. v It is not due to chance fluctuation. © 2007 Thomson/South-Western. All rights reserved. 11

Random Sampling and Nonsampling Errors (cont’d) • Less than Perfectly Representative Samples Ø 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. © 2007 Thomson/South-Western. All rights reserved. 12

EXHIBIT 12. 3 Errors Associated with Sampling Source: Adapted from Keith K. Cox and Ben M. Enis, The Marketing Research Process (Pacific Palisades, CA: Goodyear, 1972); and Danny N. Bellenger and Barnet A. Greenberg, Marketing Research: A Management Information Approach (Homewood, IL: Richard D. Irwin, 1978), pp. 154– 155. © 2007 Thomson/South-Western. All rights reserved. 13

Probability versus Nonprobability Sampling • Probability Sampling Ø A sampling technique in which every member of the population has a known, nonzero probability of selection. • Nonprobability Sampling Ø A sampling technique in which units of the sample are selected on the basis of personal judgment or convenience; the probability of any particular member of the population being chosen is unknown. © 2007 Thomson/South-Western. All rights reserved. 14

Nonprobability Sampling • Convenience Sampling Ø Obtaining those people or units that are most conveniently available • Judgment (Purposive) Sampling Ø An experienced individual selects the sample based on personal judgment about some appropriate characteristic of the sample member. • Quota Sampling Ø Ensures that various subgroups of a population will be represented on pertinent characteristics to the exact extent that the investigator desires. © 2007 Thomson/South-Western. All rights reserved. 15

Nonprobability Sampling (cont’d) • Possible Sources Of Bias Ø Respondents chosen because they were: v Similar to interviewer v Easily found v Willing to be interviewed v Middle-class v Needed to fill out the sample quota • Advantages of Quota Sampling Ø Speed of data collection Ø Lower costs Ø Convenience © 2007 Thomson/South-Western. All rights reserved. 16

Nonprobability Sampling (cont’d) • Snowball Sampling Ø A sampling procedure in which initial respondents are selected by probability methods and additional respondents are obtained from information provided by the initial respondents. © 2007 Thomson/South-Western. All rights reserved. 17

Probability Sampling • Simple Random Sampling Ø Assures each element in the population of an equal chance of being included in the sample. • Systematic Sampling Ø A starting point is selected by a random process and then every nth number on the list is selected. • Stratified Sampling Ø Simple random subsamples that are more or less equal on some characteristic are drawn from within each stratum of the population. © 2007 Thomson/South-Western. All rights reserved. 18

Proportional versus Disproportional Sampling • Proportional Stratified Sample Ø The number of sampling units drawn from each stratum is in proportion to the population size of that stratum. • Disproportional Stratified Sample Ø The sample size for each stratum is allocated according to analytical considerations. © 2007 Thomson/South-Western. All rights reserved. 19

Cluster Sampling • Cluster Sampling Ø 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. © 2007 Thomson/South-Western. All rights reserved. 20

EXHIBIT 12. 6 Examples of Clusters © 2007 Thomson/South-Western. All rights reserved. 21

Multistage Area Sampling • Multistage Area Sampling Ø Involves using a combination of two or more probability sampling techniques. v Typically, geographic areas are randomly selected in progressively smaller (lower-population) units. v Researchers may take as many steps as necessary to achieve a representative sample. v Progressively smaller geographic areas are chosen until a single housing unit is selected for interviewing. © 2007 Thomson/South-Western. All rights reserved. 22

EXHIBIT 12. 5 Illustration of Multistage Area Sampling Source: From Interviewer’s Manual, Revised Edition (Ann Arbor, MI: Survey Research Center, Institute for Social Research, University of Michigan, 1976), p. 36. Reprinted by permission. © 2007 Thomson/South-Western. All rights reserved. 23

What Is the Appropriate Sample Design? Degree of Accuracy Adaptation Knowledge of Population © 2007 Thomson/South-Western. All rights reserved. Appropriate Sample Design Resources Time 24

EXHIBIT 12. 6 Comparison of Sampling Techniques: Nonprobability Samples © 2007 Thomson/South-Western. All rights reserved. 25

EXHIBIT 12. 7 Comparison of Sampling Techniques: Probability Samples © 2007 Thomson/South-Western. All rights reserved. 26

EXHIBIT 12. 10 Comparison of Sampling Techniques: Probability Samples (cont’d) © 2007 Thomson/South-Western. All rights reserved. 27

Internet Sampling is Unique • Advantages Ø Internet surveys can rapidly reach a large sample. v Speed is both an advantage and a disadvantage: – Sample size requirements met almost instantaneously. – Survey must be kept open long enough for all sample units to participate. Ø Hard-to-reach subjects may participate v Internet samples may be representative of a target population if the target population is visitors to a particular Web site • Disadvantage Ø Lack of computer ownership and Internet access among certain segments of the population © 2007 Thomson/South-Western. All rights reserved. 28

Internet Sampling Sources Website Visitors Panel Samples Recruited Ad Hoc Samples Opt-in Lists © 2007 Thomson/South-Western. All rights reserved. 29

Internet Sampling Is Unique (cont’d) • Web Site Visitors Ø Unrestricted samples are clearly convenience samples Ø Randomly selecting visitors for a questionnaire request randomly "pops up" Ø Samples may over-represent more frequent visitors Ø Filters (e. g. , cookies, passwords, and sign-ins) can identify unique visitors © 2007 Thomson/South-Western. All rights reserved. 30

Internet Sampling Is Unique (cont’d) • Panel Samples Ø Typically yield a high response rate v Members may be compensated for their time with a sweepstake or a small, cash incentive. • Database on Members Ø Demographic and other information from previous questionnaires • Select quota samples based on: Ø Product ownership Ø Lifestyle Ø Other characteristics. © 2007 Thomson/South-Western. All rights reserved. 31

Internet Sampling Is Unique (cont’d) • Recruited Ad Hoc Samples Ø Sample is recruited from databases of e-mail addresses compiled from customer/client lists, advertising banners, online sweepstakes, and registration forms. • Opt-in Lists Ø Subjects give permission to receive selected e-mail, such as questionnaires, from a company with an Internet presence. © 2007 Thomson/South-Western. All rights reserved. 32

Key Terms and Concepts • Sample • Population (universe) • Population element • Systematic (nonsampling) error • Nonresponse error • Probability sampling • Census • Sampling frame • Reverse directory • Nonprobability sampling • Convenience sampling • Judgment (purposive) sampling • Sampling frame error • Sampling unit • Primary sampling unit (PSU) • Quota sampling • Snowball sampling • Simple random sampling • Secondary sampling unit • Tertiary sampling unit • Random sampling error • Systematic sampling • Periodicity • Stratified sampling © 2007 Thomson/South-Western. All rights reserved. 33

Key Terms and Concepts (cont’d) • Proportional stratified sample • Disproportional stratified sample • Cluster sampling • Multistage area sampling • Opt in © 2007 Thomson/South-Western. All rights reserved. 34