SurveysGeneral IdeaSome new vocabularyPlan of StudyChallenges to writing good questionsSteps to running a surveyMisspecifying target populationSurveys that use census as sampling frameSelecting samplesObtaining survey dataUnreliable methods of selecting samplesAdditional potential pitfallsExample of voluntary response surveyExamples of problematic survey designsHite’s surveyHite’s surveySlide 17Another problematic survey designRandom sampling comment 1Random sampling comment 2Random sampling comment 3Random sampling comment 4FPP Chapter 19SurveysGeneral Idea Parameter StatisticInferenceSamplePopulationSome new vocabulary PopulationSampleParameterStatisticInferenceBiasNon-response biasResponse biasSimple random sampleConvenience samplingFrame coverage biasJudgment samplingVoluntary samplingProbably others that I’ve missedPlan of Study 1. Issues in questionnaire design2. Methods for selecting units to survey3. Administration of surveysChallenges to writing good questions1. Defining objectives and specifying the kind of answers needed to meet objectives of the question2. Ensuring all respondents have a shared, common understanding of the question3. Ensuring people are asked questions to which they know the answers4. Asking questions respondents are able to answer in the terms required by the question5. Asking questions respondents are willing to answer accurately6. Asking questions that don’t lead respondent to a certain answerSteps to running a survey1. Establish the target population2. Obtain a sampling frame (this can be very difficult)3. Select a sample4. Obtain data from the sampled unitsMisspecifying target population1994 Democratic gubernatorial primary in ArizonaAll polls predict Eddie Basha would trail front-runner by at least 9 pointsResult of election: Basha wonTarget population used in polls: registered voters who had voted in previous primariesSurveys that use census as sampling frameU.S. census often used as frame for many federal and social surveys target population here is folks living in U.S.U.S. census misses some people can you think of any examples?Samples take from frame are non-representative even before samplingSelecting samplesUnits sample should be representative of the target populationHow do we ensure this?Select a subset of units from the frame at randomMost common method is to obtain a “simple random sample”If random sample is large enough, it should have characteristics that mirror the characteristics of the population frame.Obtaining survey dataRemember the following when designing a surveyImperative that purpose of survey is stated clearlyConfidentiality should be promised and keptAt ISU there is a group that ok’s confidentiality of surveys is metMethod for asking questions should be the same for all sampled unitsUnreliable methods of selecting samplesWhat follows are examples of how NOT to select a sampleConvenience sampling:Picking units that are easy to measureJudgement sampling:Picking units you judge as representative of the populationVoluntary response sampling:Picking units who respond voluntarilyWhat are some examples of each?Additional potential pitfallsNonresponse bias:Units that do not respond differ from those that do. These folks will be under representated.Frame coverage bias:Frame doesn’t include all of target populationCan we think of some examples?Example of voluntary response surveyNightline call-in poll:Ted Koppel asked people to call his show to express their opinion on whether the United Nations should continue to have its headquarters in New York186,000 people called in with 67% saying no.Independent random sample: 72% said yes.Examples of problematic survey designsShere Hite’s book, Women and Love: A Cultural Revolution in Progress (1987), claims:84% of women “not satisfied emotionally with their relationships” (pg. 804)95% of women “report forms of emotional and psychological harassment from men with whom they are in love relationships” (pg. 810)70% of women “married five or more years are having sex outside of their marriages” (pg. 856)Hite’s surveyTo whom did she send a survey?100,00 questionnaires mailed to professional women’s groups, counseling centers, church societies, and senior citizens’ centers.Her target population was women. What was her actually population?Hite’s surveyWhat did the survey look like?127 essay questions on questionnaire4.5% of these questionnaires returnedWhat was not taken into account?Hite’s surveyHow did she ask the questions?Questions use vague words like “love”. People have different interpretations of such wordsQuestions were leading“Does your husband/lover treat you as an equal? Or are there times when he seems to you as an inferior? Leave you out of decisions? Act superior?” (pg. 795)Another problematic survey designThe article “Abortion Rights Groups Surveying Voters’ Views”, by Jack Coffman, appeared in the December 26, 1989 issue of the St. Paul Pioneer Press Dispatch.Problems with Minnesota surveyRandom sampling comment 1Say you collect data on units using a method other than a random sample, and you know these data are not representative of the population of interest. Then, you take a random sample from these collected data. This random sample is representative of the population.Wrongo !!Large random samples are representative of the population in the frame.Effectively, this methods uses the unrepresentative, collected data as a frame.By randomly sampling from a unrepresentative sample, you just get a smaller unrepresentative sample.Random sampling comment 2Say you obtain data that are representative of the target population. Should you take a random sample from these collected data?This question arises when researchers use data collected by others, for example in a Stat 101 project.No!If you have a representative sample, use it. This sub-sampling method just reduces the amount of data you work withRandom sampling comment 3A census is a measurement of outcomes for all units in the population. For example the U.S.. Government does a census of the population every 10 years to apportion seats in the House of Representatives. It also takes censuses of agriculture and business.Why do survey instead of