Chapter 11
Sampling
Outline

Concepts

I. The Role of Sampling in Quantitative Research 

A. Relating Sampling to Other Concepts 
sampling: selecting events from a
(sample, population, statistic, parameter) population 
B. Sampling Begins by Defining the Population 

C. Careful Sampling Is a Key to Eliminate Bias 
bias: in sampling, bias means a tendency for the sample to err so that it fails to represent the population 
II. Essentials of Sampling A. Representative Sampling: The Goal of Effective Sampling 
representative sample: one that accurately reflects characteristics of the population from which it was drawn 
B. Sample Size 1. The Notion of Sampling Error 
sampling error: the degree to which a sample differs from population characteristics on some measure (in general, as the sample size gets to be a larger and larger proportion of the population, the amount of sampling error is reduced) 
Independent Research Boards: groups that review research with attention to the manner of selecting subjects, protection of subjects' rights, steps to secure informed consent, protection of subject confidentiality, steps for protecting subjects from any study risks  
2. Sampling Guidelines in pilot studies, small samples may be used; in speech and hearing science, samples of eight to fifteen subjects may be used in experiments in which physiological reactions are involved; for studies attempting to validate new measurement instruments, researchers may be told to sample quite large, at least 200 events; to use some statistical tools, such as multiple correlation or factor analysis, researchers need a minimum number of events for every variable included in the study 

3.
Guidelines Based on Sampling Error 
confidence intervals: the probability that sample statistics "capture" population parameters, within certain margins for error 
C. Statistical Effects of Small Samples when small samples are used, only very big effects stand out statisticians have observed that samples of thirty or more events tend to produce identical distributions use of volunteers: samples of volunteers may differ greatly from other members of the population III. Forms of Sampling 

A. Random Sampling  random sampling: securing data such that each event in
the population has an equal chance of being selected systematic or periodic sampling: selecting respondents according to a predetermined schedule other than a random sequence 
1. Simple Random Sampling  simple random sampling: selection of data such that each event in the population has an equal chance of selection 
2.
Stratified Random Sampling 
stratified random sampling: samples are defined on the basis of known proportions within the population and random sampling is completed within each group 
3. Cluster Sampling  cluster sampling: sampling in which groups or areas (clusters) are randomly selected and from which an actual sample is drawn 
B. Nonrandom Sampling advantages of nonrandom sampling: 1. often allows the researcher to get samples that otherwise would be unavailable 2. nonrandom sampling often invited by field and quasiexperimental research limitations: 1. tends to show great biases; 2. no sampling error computation is possible; 3. severely limits conclusions that may be drawn by researchers 

1.
Accidental or Convenience Sampling 
accidental or convenience sampling: selection of events that are most readily available 
2. Quota Sampling  quota sampling: samples are defined on the basis of the known
proportions within the population and nonrandom sampling is completed within each group 
3.
Purposive or Known Group Sampling 
purposive or known group sampling: selection of events from groups that are known to possess a particular characteristic under investigation 
4.
Snowball Sampling 
snowball sampling: selection of events on the basis of referrals from initial informants 
IV. Dealing With Sampling Problems A. Subject Refusal to Participate ways respondents may refuse to participate: 1. they can decline initially to accept a questionnaire, answer any questions, or mail in a survey; 2. they can provide incomplete responses 

informed consent: the requirement that individuals be permitted to withdraw from an experiment or study  
B. Looking for Evidence of Randomization in Research Articles we expect that research articles either go into detail explaining how they handled randomization, or we expect that some reference will be made to a table of random numbers 