Representative Sampling

1
Representative Sampling

• Presented at the AWDS Task Forces Marketing
  WorkshopBig Sky, Montana Friday, September 20,
  2002
• Len Singel, AWDS Coordinator

2
Overview

• Definitions
• Reducing Error
• Sample Creation
• Examples

3
Definitions

• Population consists of all the units
  (individuals, households, organizations) to which
  you desire to generalize survey results
• Sample Frame list from which your sample is
  drawn

4
Definitions

• Sample the set of respondents selected from a
  larger population for the purpose of a survey
• Completed Sample consists of all units that
  return completed questionnaires (instruments)

5
Definitions

• Coverage Error results from every unit in the
  Survey Population not having a known, non-zero
  chance of being included in the sample
• Sampling Error the result of collecting data
  from a subset, rather than all of the members of
  the sampling frame

6
Avoiding Coverage Error
7
How to Reduce Coverage Error?

• Make certain your list contains everyone in the
  Survey Population
• Unique, non-repeating Sampling Units
• Update Properly maintain your list
• Have your list contain other information that can
  be used to improve your survey

8
How Large Should Your Sample be?

• Sample does not have to be arbitrarily selected
• ALL THAT IS REQUIRED
• Knowing about your study population
• AND
• Knowing your survey objectives

9
How Large Should Your Sample Be?

• Sample Size Depends On
• How much sampling error can be tolerated
• Population size
• Variation of Population (50/50 or 80/20)
• Amount of Confidence

10
Does Size Matter
11
Small Populations
Population Size Sample Size for 95 Percent Confidence Sample Size for 95 Percent Confidence Sample Size for 95 Percent Confidence Sample Size for 95 Percent Confidence
Population Size 3 Sampling Error 3 Sampling Error 5 Sampling Error 5 Sampling Error
Population Size 50/50 Split 80/20 Split 50/50 Split 80/20 Split
100 92 87 80 71
500 341 289 217 165
1,000 516 406 278 198
5,000 880 601 357 234

12
Larger Populations
Population Size Sample Size for 95 Percent Confidence Sample Size for 95 Percent Confidence Sample Size for 95 Percent Confidence Sample Size for 95 Percent Confidence
Population Size 3 Sampling Error 3 Sampling Error 5 Sampling Error 5 Sampling Error
Population Size 50/50 Split 80/20 Split 50/50 Split 80/20 Split
10,000 964 639 370 240
50,000 1,045 674 381 245
100,000 1,056 678 383 245
1,000,000 1,066 682 384 246

13
3 Easy Steps to Sampling

1. Identify your Target Population
2. Put together a Population List
3. Select the Sample

14
Hey, Wait a Minute..

• I DONT KNOW HOW TO
• SELECT THE SAMPLE!!!

15
Sampling Methods

• Probability
• Simple Random Sampling (SRS)
• Systematic Sampling
• Nonprobability (Purposeful)

16
Probability Sampling Method

• Simple Random Sampling (SRS)
• Each member of the target population has an equal
  chance of being selected
• All elements are selected randomly CAN BE
  CUMBERSOME!!!
• Systematic Sampling
• Members of the target population have an equal
  chance of being selected
• Only the first element is selected randomly rest
  selected systematically (every 5th, 10th, )

17
Purposeful Sampling Method

• Depends on subjective judgment
• Selected because it is convenient or typical
• All members do not have the same chance of being
  selected

18
Purposeful Sampling Method

• Appropriate for
• Exploratory Research
• Focus Groups
• Organizing Committees
• Building Networks
• Do not use if your goal is to learn about the
  larger population
• Results are not
  generalizable!!

19
Example 1 Successful Hunting Experience

• Target Population 1 million Deer Hunters
• Survey Objectives Learn about Hunting
  Success
• Acceptable Sampling Error 3
• Confidence Level 95
• Population Variation 80/20

20
Example 1Successful Hunting Experience

• Population List Automated Licensing
  Database
• Sample Size 683
• Sampling Method Systematic Sampling

21
Example 2Female Hunter Perceptions

• You know you want to study female waterfowl
  hunters and their satisfaction levels.
• But
• You do not know much about them
• AND
• You do not know your survey objectives

22
Example 2Female Hunter Perceptions

• Target Population Female Waterfowl
  Hunters (n 100?)
• Survey Objectives Explore their satisfaction
  levels
• Acceptable Sampling Error 3
• Confidence Level 95
• Population Variation 50/50 or 80/20 ???

23
Example 2Female Hunter Perceptions

• Source Automated Licensing Database
• Sample Method ??
• Sample Size??

24
Example 2Female Hunter Perceptions

• The Solution!
• Conduct Exploratory Research
• Increase Sampling Error to 10
• Hold a series of focus groups (n 50)

25
Example 2Female Hunter Perceptions

• Enables you to
• Learn about Population Variation
• Learn about General Characteristics
• Define Survey Objectives
• Select appropriate sample size to be surveyed.

26
Review

• What does sample size depend on?
• What are the 3 steps to sampling?
• Does size matter?

27
Sources

• Dillman, D.A. 2000. Mail and Internet Surveys
  The Tailored Design Method. John Wiley Sons,
  Inc. 464pp.
• Salant, P. and D.A. Dillman. 1994. How to Conduct
  your Own Survey. John Wiley Sons, Inc. 232pp.