The reflection of hierarchical cluster analysis of co-occurrence matrices in SPSS.pdfVIP

ResearchPaper Thereflectionofhierarchicalclusteranalysis ofCO．．occurrencematricesinSPSS QiujuZHOU．FuhaiLENG LoetLEYDESDORFF。 NationalScienceLibrary,ChineseAcademyofSciences，100190Beijing，China zUniversityofAmsterdam，AmsterdamSchoolofCommunicationResearch(ASCoR)， Received：Ju1．29，2015 PO Box15793，1001NGAmsterdam，theNetherlands Accepted：Ju1．31，2015 Abstract Purpose：Todiscusstheproblemsarisingfrom hierarchicalclusteranalysisofco—occurrence matricesinSPSS，andthecorrespondingsolutions． Design／methodology／approach：WedesigndifferentmethodsofusingtheSPSShierarchical clusteringmoduleforco．．occurrencematricesinordertocomparethesemethods．W eofferthe correctsyntax to deactivate the similarity algorithm forclusteringanalysiswithin the hierarchicalclusteringmoduleofSPSS． Findings：W henoneinputsco—occurrencematricesintothedataeditoroftheSPSShierarchical clusteringmodulewithoutdeactivatingthe embedded similarity algorithm，theprogram calculatessimilaritytwice，andthusdistortsandoverestimatesthedegreeofsimilarity． Practicalimplications：Weofferthecorrectsyntaxtoblockthesimilarity algorithm for clusteringanalysisintheSPSShierarchicalclusteringmoduleinthecaseofcooccurrence matrices．Thissyntaxenablesresearcherstoavoidobtainingincorrectresults． Originality／value：Thispaperpresentsamethodofeditingsyntaxtopreventthedefaultuse ofasimilarityalgorithm forSPSS’shierarchicalclusteringmodule．Thiswillhelpresearchers， especially thosefrom China，toproperly implementtheco—occurrencematrixwhenusing SPSSforhierarchicalclusteranalysis，inordertoprovidemorescientificandrationalresults． Keywords Co—occurrence matrices；Hierarchicalcluster analysis；SPSS；Si