Bài giảng Business Research Methods - Chapter 15: Data Preparation and Description

Chapter 15Data PreparationandDescriptionMcGraw-Hill/IrwinCopyright © 2011 by The McGraw-Hill Companies, Inc. All Rights Reserved. Learning ObjectivesUnderstand . . .The importance of editing the collected raw data to detect errors and omissions.How coding is used to assign number and other symbols to answers and to categorize responses.The use of content analysis to interpret and summarize open questions.2Learning ObjectivesUnderstand . . .Problems with and solutions for “don’t know” responses and handling missing data.The options for data entry and manipulation.3Goal of Data Decription“The goal is to transform data intoinformation, and information into insight.Carly Fiorina former president and chairwoman, Hewlett-Packard Co4PulsePoint: Research Revelation55The percent of white-collar workers who answer work-related calls or e-mail after work hours.5Data Preparation in the Research Process6Monitoring Online Survey DataOnline surveys need special editing attention. CfMC provides software and support to research suppliers to prevent interruptions from damaging data .7EditingCriteriaConsistentUniformly enteredArranged forsimplificationCompleteAccurate8Field EditingSpeed without accuracy won’t help the manager choose the right direction.Field editing reviewEntry gaps identifiedCallbacks madeValidate results9Central EditingBe familiar with instructions given to interviewers and codersDo not destroy the original entryMake all editing entries identifiable and in standardized formInitial all answers changed or suppliedPlace initials and date of editing on each instrument completed10Sample Codebook11Precoding12Coding Open-Ended Questions6. What prompted you to purchase your most recent life insurance policy? _______________________________ _______________________________ _______________________________ _______________________________ _______________________________ _______________________________ _______________________________ _______________________________13Coding RulesCategories should beAppropriate to the research problemExhaustiveMutually exclusiveDerived from one classification principle14Content AnalysisQSR’s XSight software for content analysis.15Content Analysis16Types of Content AnalysisSyntacticalPropositionalReferentialThematic17Open-Question Coding Locus of Responsibility Mentioned Not MentionedA. Company________________________________________________B. Customer________________________________________________C. Joint Company-Customer________________________________________________F. Other________________________________________________ Locus of ResponsibilityFrequency (n = 100)A. Management 1. Sales manager 2. Sales process 3. Other 4. No action area identifiedB. Management 1. Training C. Customer 1. Buying processes 2. Other 3. No action area identifiedD. Environmental conditionsE. TechnologyF. Other1020731512852018Handling “Don’t Know” ResponsesQuestion: Do you have a productive relationship with your present salesperson?Years of Purchasing YesNoDon’t KnowLess than 1 year10%40%38%1 – 3 years3030324 years or more603030Total100%n = 650100%n = 150100%n = 20019Data Entry20Missing DataListwise DeletionPairwise DeletionReplacement21Key TermsBar codeCodebookCodingContent analysisData entryData fieldData fileData preparationData recordDatabaseDon’t know response EditingMissing dataOptical character recognitionOptical mark recognitionPrecodingSpreadsheetVoice recognition22Appendix 15aDescribing Data StatisticallyMcGraw-Hill/IrwinCopyright © 2011 by The McGraw-Hill Companies, Inc. All Rights Reserved. Research Adjusts for Imperfect Data“In the future, we’ll stop moaning about the lack of perfect data and start using the good data with much more advanced analytics and data-matching techniques.”Kate Lynch research directorLeo Burnett’s Starcom Media Unit24FrequenciesUnit Sales Increase (%)FrequencyPercentageCumulative Percentage56789Total12321911.122.233.322.211.1100.011.133.366.788.9100Unit Sales Increase (%)FrequencyPercentageCumulative PercentageOrigin, foreign (1)67812211.122.222.211.133.355.5Origin, foreign (2)5679Total1111911.111.111.111.1100.066.677.788.8100.0AB25Distributions26Characteristics of Distributions27Measures of Central TendencyMeanModeMedian28Measures of VariabilityInterquartile rangeQuartile deviationRange Standard deviationVariance29Summarizing Distribution Shape30___Symbols31Key TermsCentral tendencyDescriptive statisticsDeviation scoresFrequency distributionInterquartile range (IQR)KurtosisMedianModeNormal distributionQuartile deviation (Q)SkewnessStandard deviationStandard normal distributionStandard score (Z score)VariabilityVariance32