近年台灣高齡化現象越發嚴重，家庭人數持續下降，根據內政部統計，平均每戶家庭的家庭成員約為2.7人，又以嘉義縣、雲林縣及南投縣更為嚴重。對於家中有年長者或需長時間照護之個案或長者，就需要外來照護，故此家庭的照顧功能逐漸下降。此外，目前台灣約有79.4萬人需要被照護，但其中只有約30萬人由家庭成員自行照顧。日常照顧的壓力逐漸增加，使得需相關部門需要投入更多的社會成本或資金去協助照顧有需要之個案。個案所填寫或接受訪查之問卷(以下稱為原始問卷)內容含量太多，需要花費較長時間填寫。由於所投入之社會成本甚大且有限，但需要援助之個案數量也龐大，開始出現不能負擔現象，故此需要制定一有效率的方法評估個案是否立即需要援助。因此，透過台北市社會局提供原始問卷資料，利用結構方程式，同時配合克隆巴赫係數與因子分析反映信度與效度，找出具有代表性(或較重要性)的題目，化簡原始問卷，並為每一位個案計算分數，協助衡量是否需要即刻援助。

In recent years,the phenomena of aging society is getting worse and the numbers of family members are decreasing.According to the reports of Ministry of the Interior,the average numbers of family members for each family are about 2.7 persons,especially as Chiayi,Yunli and Nantou are worst.For those elders and patients who need long-term-care,the function of home-caring is declining.In other words,there are about 79.4 thousands of people are waiting for caring but there are only about 30 thousands of them are cared by their own family members.It means that for the revalant departments are forced to face the problem of caring.Therefore,the numbers of applying for caring support increase rapidly.Thus,the revalant departments have to review for each applicant through a meaningful and significant questionnaire.By using Structural Equation Modeling(SEM) ,Cronbach’s Alpha as reference of reliability,Factor Analysis as reference of Validity and pretest data to modify the original questionnaire which provided by Taipei City Government, Department of Social Welfare to reduce the numbers of questions and calculate a representative score for each applicant for evaluating the level of urgency.

第一章 緒論
第一節 研究背景與動機 1
第二節 問卷介紹 1
第三節 敘述統計 3
第二章 研究方法
第一節 信度 15
第二節 效度 16
第三節 資料結構方程式模型分析 18
第四節 模型建立流程 22
第三章 分析結果
第一節 各項模型信效度 26
第二節 各項模型架構 31
第三節 最終模型信效度 34
第四章 更多模型探討
第一節 增加一子題項到各部分(E、I及G部分) 39-49
第二節 增加新潛在因子到各部分(E、I及G部分) 39-49
第三節 兩部分同時增加新潛在因子 50
第五章結論與建議
第一節 最總問卷分數計算 53
第二節 結論 56
參考文獻 57
附件
I.Cronbach's Alpha計算與推導 58
II.最終模型報表 59
III.建議之因素分析結果 61
IV.資料來源

[1]Cronbach,L.J.(1951).Coefficient Alpha and the Internal
Structure of Tests.Psychometrika,16,297-334.
[2]Kerlinger,F.N(1973).Foundations of Behavioral Research.
(2nd ed.).NY:Holt,Rinehart & Winston.
[3]Joeskog,K.G.(1973).A general method for estimating a linear
structural equation systems.In A.S.Goldberger &
O.D.Duncan(Eds.),Structural equation models in the
social science(pp.85-112).New York:Academic.
[4]Hu,L.,Bentler,P.M.(1999).Cutoff criteria for fit indexes in
covariance structural Equation Modeling,6(1),1-55.
[5]Bentler,P.M.(1995).EQS structural equations program
manual.Encino,CA:Multivariate Software.
[6] Bentler,P.M.(1992).On the fit of models to covariances and
methodology to the Bulletin.Psychologcal Bulletin,112,400-
404.
[7]Browne,M.W.,& Cudeck,R.(1993).Alternative ways of assessing
model fit.In Bollen &J.S.Long(Eds.),Testing structural
equation models(pp.136-162).Newbury Park,CA:Sage.
[8]Bentler,P.M., & Bonett,D.G.(1980).Significance tests and
goodness of fit in the analysis of covariance
structures.Psychological Bulletin,88,588-606.
[9]Ahmann, J.S. and Glock, M.D. (1981) Evaluating Student
Progress: Principles of Tests and Measurements. London:
Allyn and Bacon, Inc.
[10]Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill.