Sampling Chapter 5 Introduction Sampling The process of drawing a number of individual cases from a larger population A way to learn about a larger population by obtaining information from a subset of a larger population Example Presidential polls are based upon samples of the population that might vote in an election Introduction Why Sample? To learn something about a large group without having to study every member of that group Time and cost Studying every single instance of a thing is impractical or too expensive Example Census Introduction Why Sample? Improve data quality Obtain in-depth information about each subject rather than superficial data on all Introduction Why Sample? We want to minimize the number of things we examine or maximize the quality of our examination of those things we do examine. Introduction Why Sample? When is sampling unnecessary? The number of things we want to sample is small Data is easily accessible Data quality is unaffected by the number of things we look at Example You are interested in the relationship between team batting average and winning percentage of major league baseball teams There are only 30 major league teams Data on team batting averages and winning percentages are readily available Introduction Why Sample? Elements A kind of thing the researcher wants to look at Quiz – Question 1 Suppose you are interested in describing the nationality of Nobel prize-winning scientists. What would an element in your study be? What would the population be? Introduction Why Sample? Population The group of elements from which a researcher samples and to which she or he might like to generalize Quiz – Question 2 In the case of presidential elections in the United States the population is ________ and the elements of this population are _________. Introduction Why Sample? Sample A number of individual cases drawn from a larger population Introduction Sampling Frames, Probability versus Nonprobability Samples Target population A population of theoretical interest Introduction Sampling Frames, Probability versus Nonprobability Samples Sampling frame or study population The group of elements from which a sample is actually selected Quiz – Question 3 The local television station conducted a study of TV viewers in the local viewing region. A list of all residential customers who subscribed to cable TV was obtained from the cable company. The list had 200,000 households as subscribers. The TV station samples every 40th household on the subscriber list. An interviewer visited each household and conducted the survey on viewing habits of household members. What is the sampling frame of the study? Introduction Sampling Frames, Probability versus Nonprobability Samples Nonprobability Samples A sample that has been drawn in a way that doesn’t give every member of the population a known chance of being selected Introduction Sampling Frames, Probability versus Nonprobability Samples Probability A sample drawn in a way to give every member of the population a known (nonzero) chance of inclusion Probability samples are usually more representative than nonprobability samples of the populations from which they are drawn Introduction Sampling Frames, Probability versus Nonprobability Samples Biased Samples A sample that is not representative from the population which it is drawn Probability samples are LESS likely to be biased samples Introduction Sampling Frames, Probability versus Nonprobability Samples Generalizability The ability to apply the results of a study to groups or situations beyond those actually studied A probability sample tends to be more generalizable because it increases the chances that samples are representative of the populations from which they are drawn. Introduction STOP AND THINK Can you think why researchers haven’t used cell phone numbers in polling until recently? What problem may result from only using landline numbers? Focal Research “Calling Cell Phones in ’08 Pre-Election Polls” Examines the hypothesis than Barack Obama fared better in probability samples including landline- and cell phone-users than in samples including landline users alone. Focal Research Thinking about ethics Because of the sampling technique employed, the Pew pollsters never knew the identity of their respondents, so respondent anonymity was never in danger. Moreover, participation in the survey was voluntary. Sources of Error Associated with Sampling Types of Survey Error – due to sampling Coverage Error Nonresponse Error Sampling Error Sources of Error Associated with Sampling Coverage Errors Errors that results from differences between the sampling frame and the target population Sources of Error Associated with Sampling Coverage Errors People are typically left out, if samples are drawn from phone books, car registrations, etc… Unlisted Phone Numbers – one of the greatest potentials for coverage error Pollsters use random digit dial to avoid unlisted numbers Random-digit dialing A method for selecting participants in a telephone survey that involves randomly generating telephone numbers What are potential future problems, with using telephone listings to draw a sample? Sources of Error Associated with Sampling Coverage Errors Parameter- A summary of a variable characteristic in a population Sources of Error Associated with Sampling Coverage Errors Statistic-A summary of a variable in a sample Sources of Error Associated with Sampling Nonresponse Error Errors that result from differences between nonreponders and responders to a survey Stop and Think What kinds of people might not be home to pick up the phone in the early evening when most survey organizations make their calls? What kinds of people might refuse to respond to telephone polls, even if they were contacted? Sources of Error Associated with Sampling Sampling Error Any difference between the characteristics of a sample and the characteristics of the population from which the sample is drawn Sources of Error Associated with Sampling Sampling Error Sampling Variability The variability in sample statistics that occurs when different samples are drawn from the same population Sources of Error Associated with Sampling Margin of error Suggestion of how far away the actual population parameter is likely to be from the statistic Types of Probability Sampling Simple Random Sampling Systematic Sampling Stratified Sampling Cluster Sampling Multistage Sampling Types of Probability Sampling Simple Random Sampling A probability sample in which every member of a study population has been given an equal chance of selection One way to draw a simple random sample, is to put all possibilities on paper, cut them up, and then draw a sample from a hat Research Randomizer (http://randomizer.org) Types of Probability Sampling Simple Random Sampling Sampling distribution The distribution of a sample statistic A visual display of the samples Types of Probability Sampling Types of Probability Sampling Systematic Sampling A probability sampling procedure that involves selecting every kth element from a list of population elements, after the first element has been randomly selected Example Divide the total number of elements by the number you want in your sample 24/6 = 4 Randomly select a number between 1 and 4 and then select every 4th element from that number Types of Probability Sampling Systematic Sampling Selection interval The distance between the elements selected in a sample Selection Interval (k) = population size sample size Types of Probability Sampling Stratified Sampling A probability sampling procedure that involves dividing the population in groups or strata defined by the presence of certain characteristics and then random sampling from each stratum Example If you had a population that was 10% women and you want a sample that is also 10% women Types of Probability Sampling Stratified Sampling Steps to draw a stratified random sample 1. 2. 3. Group the study population into strata or into groups that share a given characteristic Enumerate each group separately Randomly sample within each strata Types of Probability Sampling Cluster Sampling A probability sampling procedure that involves randomly selecting clusters of elements from a population and subsequently selecting every element in each selected cluster for inclusion in the sample Cluster sampling is an option if data collection involves visits to sites that are far apart Types of Probability Sampling Cluster Sampling Example You are conducting a study of Kentucky high school students You could obtain a list of all high school students in the state and complete random sampling A cluster sample would be more practical Obtain a list of all high schools in Kentucky Random sample the high schools from the list Obtain a list of students for each high school selected and then contact each of those students Types of Probability Sampling Multistage Sampling A probability sampling procedure that involves several stages, such as randomly selecting clusters from a population, then randomly selecting elements from each of the clusters Types of Probability Sampling Multistage Sampling Example Random Digit Dial Stage 1: Areas Codes randomly sampled Stage 2: Three digit local exchanges randomly sampled Stage 3: Last four digits randomly sampled Stage 4: Asking the person who answer the phone for the appropriate person you want to interview Quiz – Question 4 You want to draw a sample of the employees at a large university ensuring that in your sample you have people represented from all personnel categories including administrators, faculty, secretarial staff, cleaning staff, mail room staff, technicians, and students. What type of probability sample would be best? Types of Nonprobabilty Sampling Purposive Sampling Quota Sampling Snowball Sampling Convenience Sampling Types of Nonprobability Sampling Purposive Sampling A nonprobability sampling procedure that involves selecting elements based on a researcher's judgment about which elements will facilitate his or her investigation Types of Nonprobability Sampling Quota Sampling A nonprobability sampling procedure that involves describing the target population in terms of what are thought to be relevant criteria and then selecting sample elements to represent the “relevant” subgroups in proportion to their presence in the target population Types of Nonprobability Sampling Snowball Sampling A nonprobability sampling procedure that involves using members of the group of interest to identify other members of the group Types of Nonprobability Sampling Convenience Sampling A nonprobability sampling procedure that involves selecting elements that are readily accessible to the researcher Sometimes called an available-subjects sample Choosing a Sampling Technique Is it desirable to sample at all or can the whole population be used? Is it important to generalize to a larger population? Political preference polls Do you have the access and ability to perform probability sampling? Major considerations Methods Theory Practicality Ethics Summary Sampling is a means to an end. We sample because studying every element in our population is frequently beyond our means or would jeopardize the quality of our. On the other hand, we don’t need to sample when studying every member of our population is feasible.