Chapter 08 Surveys and questionnaires

Surveys and questionnairesChapter 8 Most commonly used quantitative methodUsed for obtaining information about what people do, and respondents’ attitudes or characteristicsIn experimental, quasi-experimental, and descriptive research designsWhat is a survey?System for collecting comparable information across many peopleSelf-reportFace-to-face PhoneOnline MailComparing survey typesWhat is best method for reaching your population?What is the best way to collect data about your topic?Will type of survey influence your response rate?Designing a surveyDevelop the research designEvaluate existing questionnaires/surveysRecommended over creating your ownHas undergone extensive testing and revisionMinor changes are okaySubstantial changes will require that you pretest or pilot test the questionnaireWriting your own questionnaireStart with literature reviewDesigning survey itemsStraightforwardOne complete thought written in sentence or question formatRespondent should know how to answerAvoid abbreviations and slang expressionsShorter is better than longerWhat is a good survey item?Is the item straightforward?Does the item contain a complete thought?Does participant have choice in how to answer?Will the participant know how to answer?Will participants understand the terminology or abbreviations?Will participants understand the purpose of the item?Will participants have the information requested?Is the item written in such a way that participants will be willing to respond?Is the item too long?Closed questionsRespondents given a question or statement and given a set of responses to select fromAll responses must be known in advanceCreates easily comparable responsesUse a recall cue or stimulus statement to draw participants’ attention to issue, topic, or timeframeResponse sets for closed questionsNominal or categorical responsesExhaustiveMutually exclusiveEquivalentLikert-type scales5- or 7-point scale Includes middle or neutral responseSemantic differential scalesBipolar adjectives anchor 7-point scaleLikert-type Scale Response setsVery oftenFairly oftenOccasionallyRarelyNeverCompletely satisfiedMostly satisfiedUnsureMostly dissatisfiedCompletely dissatisfiedNoneVery MildMildModerateSevereAlways UsuallySometimesRarelyNeverStrongly disagreeDisagreeUndecidedAgreeStrongly agreeOpen questionsRespondents use their own words to respond Makes data less comparable and more difficult to interpretConsider what constitutes an adequate answerBuild that request into the questionUse a recall cue to draw participants’ attention to issue, topic, or timeframeRecord everything participant saysCode responses after all data are collectedDesigning flow of survyLogical, clear path for respondentEach section needs instructionsRequest only the demographic information included in your designInclude a “thanks”Must look professionalPretesting the surveyAlso called pilot testingShould be done if you developed a questionnaire or modified an existing oneFour approaches Cognitive approachConventional Behavior codingExpert panelSampling issues for surveysSample size and response rate are not the sameResponse rate or return rate = number of people who provide usable responsesAcceptable response rates vary by survey techniqueMay be differences in those who choose to respond and those that don’t respondUnusable responsesSurvey reliability and validityInternal reliability must be computed for multiple item questionnairesCronbach’s alphaVaries from 0 to 1.00Generally .70 considered acceptableThree types of validity should be consideredContent validityFace validityConstruct validityAnalyzing and reporting survey dataData of all participants combined to create a picture of the wholeOften limited to descriptive purposesData can test existing modelsData from closed questions Descriptive statisticsData from open questionsCategorized or content analyzed