COMMERCE 3MA3 Chapter Notes - Chapter 11: Data Security, Nonprobability Sampling, Multistage Sampling
Chapter Eleven: Basic Sampling Issues 10/16/18
● Basic Concepts in Sampling
● The population is the entire group under study
● Probability sample = each person has an equal probably of being selected
The Concept of Sampling
● Population
● Sample
● Census - the best information comes from here
Why samples are acceptable in research:
● Sampling is considered statistically sound
● A census is expensive and time-consuming
● It may be impossible to find an accurate, up-to-date sample frame for a full
census
● Fewer opportunities for errors / biases to impact data
Developing a Sampling Plan
Developing a Sampling Plan
Step One: Define the Population of Interest
● Target audience
● Demographics, geographics, behavioural characteristics, awareness, etc.
Document Summary
The population is the entire group under study. Probability sample = each person has an equal probably of being selected. Census - the best information comes from here. It may be impossible to find an accurate, up-to-date sample frame for a full census. Fewer opportunities for errors / biases to impact data. Used for screening respondents for fit to complete survey. Survey method: mail, internet, telephone, mall-intercept, etc. People from the list will become sample units. Step four: select a sampling method: probability sampling: Every element of the population has a known likelihood of selection for the survey. Specific elements from the population have been selected in a non-random manner. Level of accuracy you want to achieve. Time and money you have to complete survey study. Plan to conduct probability or non-probability sampling. Determine the phases of the sample selection process. Sampling error: an error that occurs because the sample selected is not perfectly representative of the population.