Bài giảng Business Research Methods - Chapter 5: Clarifying the Research Question through Secondary Data and Exploration

Clarifying the Research Question through Secondary Data and ExplorationChapter 5Learning ObjectivesUnderstand . . . The process of using exploratory research to understand the management dilemma and work through the stages of analysis necessary to formulate the research question (and, ultimately, investigative questions and measurement questions). What is involved in internal data mining and how internal data-mining techniques differ from literature searches.Pull Quote“It is critical to use serious business judgment about what types of information could possibly be useful and actionable for an organization. We have seen enormous resources expended on “data projects” that have no realistic chance of payoff. Indiscriminately boiling a data ocean seldom produces a breakthrough nugget.” Blaise Heltai, general partner,NewVantage PartnersExploratory Phase Search StrategySearch StrategyDiscovery/ AnalysisSecondary SourcesIndividual Depth InterviewsExpert InterviewGroupDiscussionsIntegration of Secondary Data into the Research Process Objectives of Secondary SearchesExpand understanding of management dilemmaGather background informationIdentify information to gatherIdentify sources for and actual questionsIdentify sources for and actual sample framesConducting a Literature SearchDefine management dilemmaConsult books for relevant termsUse terms to searchLocate/review secondary sourcesEvaluate value of each source and content Whiteboard technology makes an easier discussion of symptoms relevant to the management-research question hierarchyLevels of InformationPrimary Sources:MemosLettersInterviewsSpeechesLawsInternal recordsSecondary Sources:EncyclopediasTextbooksHandbooksMagazinesNewspapersNewscasts Tertiary Sources:IndexesBibliographiesInternet search enginesIntegrating Secondary DataThe U.S. Government is the world’s largest source of dataTypes of Information SourcesEncyclopediasDirectoriesHandbooksTypesIndexes/BibliographiesDictionariesEvaluating Information SourcesAuthorityFormatAudienceEvaluationFactorsPurposeScopeThe Evolution of Data MiningEvolutionary StepInvestigative QuestionEnabling TechnologiesCharacteristicsData collection (1960s)“What was my average total revenue over the last five years?”Computers, tapes, disksRetrospective, static data deliveryData access (1980s)“What were unit sales in California last December?”Relational databases (RDBMS), structured query language (SQL), ODBCRetrospective, dynamic data delivery at record levelData navigation (1990s)“What were unit sales in California last December? Drill down to Sacramento.”Online analytic processing (OLAP), multidimensional databases, data warehousesRetrospective, dynamic data delivery at multiple levelsData mining (2000)“What’s likely to happen to Sacramento unit sales next month? Why?”Advanced algorithms, multiprocessor computers, massive databasesProspective, proactive information deliveryData-Mining ProcessBusiness Research ProcessStage 1: Clarifying the Research QuestionManagement-research question hierarchy begins by identifying the management dilemmaManagement-Research Question HierarchySalePro’s HierarchyFormulating the Research QuestionTypes of Management QuestionsThe Research QuestionExamine variablesBreak questions downFine-TuningEvaluate hypothesesDetermine necessary evidence Set scope of studyInvestigative QuestionsPerformance ConsiderationsAttitudinal IssuesBehavioral IssuesGantt ChartMindWriter Project PlanKey TermsBibliographyBibliographic DatabaseData MartData MiningData VisualizationData WarehouseDictionaryDirectoryEncyclopediaExpert interviewExploratory researchHandbookIndexIndividual depth interview Investigative questions Literature search Management question Measurement questionCustom-designedPredesigned Primary sources Research questions Secondary sources Source evaluationPurposeScopeAuthorityAudienceFormat Tertiary sourcesAdditional Discussion opportunitiesChapter 5Snapshot: BlogsFrequent chronological publication of personal thoughts & web links1 Billion blogs and growing847 = average followers61% are hobbyists59% are men79% have college degreesMost have Facebook access to their blogsSnapshot: Deception LineBusiness intelligence is fertile ground.Comprehensive literature searchExpert interviewsFormer employee interviewsMonitor competitive publicationsAttend presentations by executivesShare proprietary informationSnapshot: Surfing the Deep Web“ Although many popular search engines boast about their ability to index informationon the Web, some of the Web’s information is invisible to their searching spiders. The most basic reason is that there are no links pointing to a page that a search engine spider can follow. Or, a page may be made up of data types that search engines don’t index—graphics, CGI scripts, or Macromedia Flash, for example.”Snapshot: Cloud Affects ResearchA computing environment where data and services reside in scalable data centers accessible over the Internet.“[The organization] pays only for [server] capacity that [it] actually uses.”“There’s no hardware to purchase, scale,and maintain, no operating systems, database servers, or application servers to install, no consultants and staff to manage it all, and no need for upgrades.”Data no longer reside on organizations serversSnapshot: Mining FeelingsSentiment analysis and opinion mining: apply computational treatment to opinion, sentiment, and subjectivity in textual form. Difficult comment analysis problemsFalse NegativesRelative SentimentCompound SentimentScoring SentimentSentiment ModifiersConditional SentimentSnapshot: Odin Text“Most firms have a wealth of rich unstructured data within their organization that they need to understand.”Monitors customer commentsDraws attention to new, important trendsCalculates sentimentFilters ‘noise’User-determined analysisSnapshot: Online Professional CommunitySponsored content websiteShop-talk communityProfessional collaboration communityResearch Thought Leaders“Companies are certainly aware of datamining, but most companies are not makingeffective use of the data collected. Theyare not so good at analyzing it or applyingthese insights to the business.”Gregory Piatetsky-Shapiropresident KdnuggetsPulsePoint: Research Revelation33The percent of financial executives who have full confidence in their current risk strategies.Data Mining in BusinessPercent of ActivityMarketingFinancialAnalysisSalesCustomerServiceFraud DetectionDistributionInsuranceNetworkManagementClarifying the Research Question through Secondary Data and ExplorationChapter 5Photo AttributionsSlideSource7Wavebreakmedia Ltd/Getty images11Courtesy of U.S. Census Bureau23DreamPictures/Blend Images LLC27Thinkstock/Jupiterimages28Steve Cole/Getty Images29Comstock Images/Jupiterimages32Courtesy of Anderson Analytics33Courtesy of 1to1Media