Introducing Communication Research 2e © 2014 SAGE Publications Chapter Eight Sampling: Who, What and How Many?
Key Concepts Population. Sample. Census. Parameters. Probability sampling - ▫random, stratified random, systematic, multi-stage cluster Nonprobability sampling - ▫convenience, purposive, quota, network Sampling frames Sampling error Introducing Communication Research 2e © 2014 SAGE Publications
Sampling Terms Population - every person in the group you are studying. Sample - part of a population selected for study. Census - study of entire population. Parameters - describe a population. Statistics - describe a sample. Sampling frame - list from which a sample is drawn. Introducing Communication Research 2e © 2014 SAGE Publications
The Two Major Types of Sampling Probability – ▫Random selection. Every unit has an equal chance of being selected. Nonprobability – ▫Sampling is based on judgment by the researcher. Introducing Communication Research 2e © 2014 SAGE Publications
Nonprobability Sampling Convenience Volunteer Purposive or judgment Quota Snowball or network Introducing Communication Research 2e © 2014 SAGE Publications
Probability Sampling Simple random - ▫Lottery, random numbers. ▫ ▫ Systematic - ▫Stratified random. ▫Multistage cluster. Introducing Communication Research 2e © 2014 SAGE Publications
Convenience Sample First ten items selected arbitrarily Introducing Communication Research 2e © 2014 SAGE Publications
Stratified Random Sample m 40% of the sample is 30% of the sample is 20% of the sample is 10% of the sample is Percentage of each group in sample is the same as the percentage of that group in population. Random sampling takes place within each group. Introducing Communication Research 2e © 2014 SAGE Publications
Systematic Sample m Random starting point selected. Then every n th unit. (n = 6 in this example) Introducing Communication Research 2e © 2014 SAGE Publications
Sampling Error The degree to which the sample statistics differ from the parameters of the universe or population from which it was selected. Margin of error, confidence level and size of sample are related. Introducing Communication Research 2e © 2014 SAGE Publications
Sample Size Depends on – ▫Purpose of the research. ▫Desired level of statistical confidence. ▫Homogeneity of the population. ▫In practice, the time and resources available for research. Introducing Communication Research 2e © 2014 SAGE Publications
Standard Error (%), Probability of Answers and Sample Size Probability of Answers Size of Sample 90:1050: %15.8% %5.0% %1.6% Introducing Communication Research 2e © 2014 SAGE Publications
Sampling Frame Issues Postal Sampling ▫may be dated ▫cannot access the transient & homeless Phone Sampling ▫problems accessing mobile numbers ▫consumer resistance to calls Internet Sampling ▫difficulties generalizing from Internet population to general population. Introducing Communication Research 2e © 2014 SAGE Publications
Chapter Summary Census - study of an entire population. Sample - part of population chosen for study. Probability sampling – random basis. Nonprobability sampling – judgment of researcher Larger sample sizes reduce sampling error. Statistical formulae give us sample size for a given margin of error, or vice versa. Introducing Communication Research 2e © 2014 SAGE Publications
Vocabulary Review Census Convenience sampling Homogeneity Multistage cluster sampling Network or snowball sampling Nonprobability sampling Pilot Probability sampling Purposive or judgmental sampling Quota sampling Random digit dialing Random numbers Random numbers generator Random sampling Sampling frames Sampling interval Sampling units Stratified random sampling Systematic sampling Volunteer sampling Introducing Communication Research 2e © 2014 SAGE Publications
Web Resources Council of American Survey Research Organizations - Research Randomizer - StatPac - Survey Sampling International - WebSM - World Association for Public Opinion Research Introducing Communication Research 2e © 2014 SAGE Publications