Subjects, Participants, and Sampling

1
Subjects, Participants,and Sampling
2
The proof of the pudding is in the eating. By
a small sample we may judge of the whole
piece.Miguel de Cervantes Saavedra Spanish
Writer, 1547-1616
3
Definitions

• Subject or participant A person from whom data
  are collected.
• Subject is the term often used in a
  quantitative context participant is used in a
  qualitative context.
• Sample The collective group of subjects or
  participants from whom data are collected
• Population A large group of individuals to
  whom the results of a study are to be generalized.

4
Two Types of Sampling Procedures

• Probability Statistically-driven sampling
  techniques where the probability of being
  selected is known. The purpose is to select a
  group of subjects representative of the
  population. (Think quantitative)
• Non-Probability Pragmatically-driven sampling
  techniques where the probability of being
  selected is known. The purpose is to select
  particularly knowledgeable participants. (Think
  qualitative)

5
Sampling for Quantitative Research Studies
6
Goals for Quantitative Sampling

• To select a sample that is representative of
  the population you will generalize your results
  to.
• To reduce sampling error and bias
• Sampling error The difference between the
  true result and the observed result that can
  be attributed to using samples rather than
  populations.
• Sampling bias The difference between the
  observed and true results that can be
  attributed to errors made by the researcher.

7
Strategies for Quantitative Sampling

• Simple Random
• Systematic Sampling
• Stratified Sampling
• Cluster Sampling
• Convenience Sampling

8
Simple Random

• A number is assigned to each subject in the
  population and a table of random numbers or a
  computer is used to select subjects randomly from
  the population.

9
1970 Viet Nam War Draft Lottery
Last Called
Call Number
Birth Date
10
Systematic Sampling

• Proportional Stratified Sampling The
  proportion of subjects in each strata in the
  population are reflected in the proportions of
  subjects in each strata of the sample.
• Disproportional Stratified Sampling The
  proportions of subjects in each strata in the
  sample are the same regardless of the proportions
  of subjects in the strata of the population.

11
Example Stratified Sample
12
Cluster Sampling

• Similar to random sampling except that
  naturally occurring groups are randomly selected
  first, then subjects are randomly selected from
  the sampled groups.
• Typical educational clusters are districts,
  schools, or classrooms.

13
Convenience Sampling

• Typical of much educational (and other)
  research given the constraints under which it is
  conducted.
• The major concern is the limited ability to
  generalize the results from the sample to a
  population the audience cares about.

14
Steps in Quantitative Sampling
Key first step Define the target population. Who
do you want to generalize your results to?
15
Sampling for QualitativeResearch Studies
16
Goals for Qualitative Sampling

• To select participants that are particularly
  knowledgeable about the topic/phenomenon you are
  researching.
• Who does an investigator or reporter interview?
  People who are most knowledgeable or have the
  closest experience with the issue.

17
Strategies for Qualitative Sampling

• Typical Case Selecting a representative
  participant.
• Extreme Case Selecting a unique or atypical
  participant.
• Maximum Variation Selecting at least two
  participants who represent extreme cases.
• Snowball (aka Network) Selecting participants
  from recommendations of other participants.
• Critical Case Selecting the most important
  participants related to the phenomenon.

18
Evaluating Sampling in Research Studies
19
Criteria for Evaluating Sampling

• The subjects or participants were clearly
  described.
• The population was clearly defined.
• The sampling procedure was clearly described.
• The sampling procedure was appropriate for the
  problem being investigated.
• The selection of subjects was free of bias.
• Adequate sample sizes were used.
• The return rate was reported and analyzed.
• The qualitative study had knowledgeable
  participants.

20
Risks Associated with Volunteers

• Different characteristics between volunteers
  and non-volunteers can lead to non-representative
  responses.
• Educational level
• Socio-economic status
• Need for social approval
• Ability to socialize
• Conformity
• Commonly used due to availability and
  convenience.

21
Sample Sizes for Experiments

• For experimental designs, sample size is a
  function of level of significance, effect size,
  and power. Change one of these, and the minimum
  sample size will change.
• For a t-test at the 0.05 level of significance,
  a power of 0.80, and a small effect size, the
  minimum sample to produce a statistically
  significant result is 393 in each group for a
  total of 786 participants. With a large effect
  size, the minimum sample would be 26 in each
  group for a total of 52 participants.

22
Sample Sizes for Experiments
ES 0.6
1.0
ES 0.5
minimum power
0.8
ES 0.4
0.6
Power
0.4
0.2
0.0
50
100
150
200
0
Sample Size Per Group
23
Sample Sizes for Correlations

• For correlational designs, sample size is also
  a function of level of significance, effect size,
  and power.
• For a statistically significant Pearson
  product-moment correlation at a 0.05 level of
  significance, a power of 0.80, and a medium
  effect size, you need 85 people.

24
Sample Sizes for Surveys

• The minimum sample size for surveys is a
  function of
• The margin of error you are willing to accept.
  This is usually set at 5.
• The confidence interval you set. Typical
  choices are 90, 95, or 99. Go with 95.
• The population size you are targeting. Usually
  it is quite large, and any number over 20,000 has
  little effect on sample size.
• For the assumptions above you would need a
  sample of 384.
• A sample size of 1,000 will give a margin of
  error of 3.
• A sample size of 100 will give a margin of
  error of 10.