2 Basic ConceptsPopulation: the entire group under study (or of interest)Exercise: Define population for a study seeking to assess SUU student attitudes towards a) program quality and delivery, b) program content, and c) social environment.Sample: subset of the populationUsed to represent the populationSample unit (elements): basic unit investigated (choose sampling units/elements when sampling)Individuals, households, etc.Census: data collected from EVERYONE in population
3 Basic Concepts (continued) AGAIN: total error = sampling error + nonsampling errorSampling error: error due to taking a sample (+/-zs)Nonsampling error: everything else (measurement, data analysis, etc.)Sample frame: list from which the sample is selectedSample frame error: Pop’n members not in frame, and members in frame not in pop’n of interest
4 Reasons for Sampling Cost Too much information to handle Sampling can be more accurateNonsampling errors can overwhelm reduction in sampling errorsSampling work behaviors exampleCensus BureauTime problem
5 Developing a sampling plana 1. Define the population of interest.2. Choose a data-collection method (mail, telephone, Internet, intercept, etc.).3. Identify a sampling frame.4. Select sampling method5. Determine sample size.6. Develop operational procedures for selecting sampling elements/units.7. Execute the operational sampling plan.
6 PROBABILITY SAMPLING METHODS Each member of population has a ‘known’ probability of being selectedSimple Random Sampling: Each member has an equal probability of being selectedBlind Draw MethodTable of Random NumbersUseful for small samples, when Random Digit Dialing (or +1) is appropriate, and computerized lists
7 PROBABILITY METHODS (Cont’d) Stratified Sampling: Population is segmented (stratified), and then samples are chosen from each strata using some other methodCan be more efficient (smaller sampling error)Homogeneous within, heterogeneous withoutUseful when interested in different strata (e.g., small numbers, etc)Disproportionate versus proportionate
8 PROBABILITY METHODS (Cont’d) Cluster Sampling: Population is divided into groups, or clusters, and then clusters are randomly chosen.Homogenous without, heterogeneous withinEvery unit in cluster examined, ORA Random (or systematic) sample is taken from chosen cluster (2-stage or 2-step approach)Careful with the probabilities!
9 PROBABILITY METHODS (Cont’d) Systematic Sampling: Randomly choosing a starting point and then choosing every nth member.Example: Need 52 data points (daily sales) for a yearSkip interval = 365/52=7.01Randomly choose 1 day out of first 7, then choose every 7th one after that.Variation: Choose every nth visitor
10 NONPROBABILITY SAMPLING METHODS Probability of selection not known, and hence representativeness cannot be assessedTechnically, confidence intervals, H0 tests, etc. not appropriateConvenience Samples:Shopping mall intercepts, classes asked to fill out questionnaires, etc.Judgment Samples: Someone puts together what is believed to be a relatively representative sampleEx.: Test markets
11 Nonprobability Sampling (Cont’d) Referral (or Snowball) SamplesQuota SamplesEXAMPLE: Choose sampling units so their representation equals their frequency in the pop’n (e.g., 52% females, 48% males)
12 The Sampling Process Identifying the Target Population Reconciling the Population, Sampling Frame DifferencesDetermining the Sampling FrameSelecting a Sampling FrameProbability SamplingNon-Probability SamplingThe Sampling ProcessDetermining the Relevant Sample SizeExecute SamplingData Collection From RespondentsHandling the Non-Response ProblemInformation for Decision-Making
13 Nonresponse Bias Reason for nonresponse: Handling nonresponse Refusal Lack of ability to respondNot at homeInaccessibleHandling nonresponseImprove research designCall-backsEstimate effectsSample nonrespondents; trends