資源描述：
《侯杰泰結(jié)構(gòu)方程模型.ppt》由會(huì)員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在教育資源-天天文庫。
1、結(jié)構(gòu)方程模型及其應(yīng)用StructuralEquationModelandItsApplications侯杰泰Kit-TaiHau香港中文大學(xué)教育心理系EducationalPsychologyDept,TheChineseUniversityofHongKong1結(jié)構(gòu)方程模型及其應(yīng)用StructuralEquationModelandItsApplicationsI簡(jiǎn)介IIntroduction2100個(gè)推理測(cè)驗(yàn)分?jǐn)?shù)21,31,32,05,06,09,10,22,29,18,11,01,39,92,23,27,93,97,30,02,96,40,53,78,04,98,36,07
2、,08,24,54,55,77,99,34,03,86,87,59,60,15,62,63,43,52,28,79,58,65,95,81,85,57,14,17,33,16,19,20,37,25,69,84,61,64,68,70,42,45,72,83,89,44,38,47,71,00,73,12,35,82,56,75,41,46,49,50,94,66,67,76,51,88,90,74,13,26,80,48,91均值Mean=53，標(biāo)準(zhǔn)差SD(StdDev)=153100名學(xué)生在9個(gè)不同學(xué)科間的相關(guān)系數(shù)(correlationcoefficientmatrix)4
3、5再生/隱含矩陣(reproduced/impliedmatrix)678檢查模型的準(zhǔn)確性(accuracy)和簡(jiǎn)潔性(parsimony)擬合優(yōu)度指數(shù)（goodnessoffitindex），簡(jiǎn)稱為擬合指數(shù)(fitindex):、NNFI、CFIdf=[不重復(fù)元素non-duplicatingelements,p(p+1)/2]–[估計(jì)參數(shù)estimatedparameters]在前面例子df=9x10/2–21=249依據(jù)及指定模型找出與相距最小的樣本相關(guān)（或協(xié)方差）矩陣correlation/covariancematrix一個(gè)或多個(gè)有理據(jù)的可能模型(alternative
4、models)輸出Output輸入InputSEM程式program(e.g.,LISREL)、各路徑參數(shù)（因子負(fù)荷loading、因子相關(guān)系數(shù)factorcorrelations等）各種擬合指數(shù)10結(jié)構(gòu)方程模型及其應(yīng)用StructuralEquationModelandItsApplicationsI簡(jiǎn)介IIntroduction11100名學(xué)生在9個(gè)不同學(xué)科間的相關(guān)系數(shù)(correlationcoefficientmatrix)1213141516171819202122_____________________________________________________
5、____________________________________________(no.ofestimatedparameters)模型dfNNFICFI需要估計(jì)的參數(shù)個(gè)數(shù)______________________________________________________________________________________________M12440.973.98221=9Load+9Uniq+3CorrM227503.294.47118=9Load+9UniqM326255.647.74519=9Load+9Uniq+1CorrM426249.656
6、.75219=9Load+9Uniq+1CorrM527263.649.72718=9Load+9UniqM624422.337.55821=9Load+9Uniq+3CorrM721113.826.89824=9Load+9Uniq+6Corr______________________________________________________________________________________________23模型比較(ModelComparison)自由度(df),擬合程度(fit),不能保證最好，可能存在更簡(jiǎn)潔(parsimonious)又?jǐn)M合(fit
7、)得很好的模型輸入Input:相關(guān)（或協(xié)方差）矩陣correlation/covariancematrix一個(gè)或多個(gè)有理據(jù)的可能模型(alternativemodels)輸出Output:既符合某指定模型，又與差異最小的矩陣估計(jì)各路徑參數(shù)parameter（因子負(fù)荷loading、因子相關(guān)系數(shù)factorcorrelations等）。計(jì)算出各種擬合指數(shù)(goodnessoffitindexes)24依據(jù)及指定模型找出與相距最小的樣本相關(guān)（或協(xié)方差）矩陣correlation/co
