信度效度分析 - web.thu.edu.tw

1 1.. (consistency).. 2.. ( ) . (constant) . (random error) .. 1. (respondent) . 2.. 3.. 4.. 3.. 1. (test-retest method): .. (test-retest reliability) . 4. 2. (split-half method): .. 5. 3. (equivalent-forms method): .. 6. 4. (Kuder-Richardson reliability).. rKR 20 = (. k )(1 . pq ). k 1 S 2. K . pq .. S2 . 7. L. J. Cronbach . k . = .. S i 2.. i + 2 S ij 2. k 1 S .. K . 2. S i . Sij . 8.. Sc6 Sc7 Sc8 Sc9. Sc6 Sc7 Sc8 Sc9 . =. 4 . 1 . ( + + + ) .. 4 1 ( + ..+ ) + 2 ( + ..+ ) . = 9.. 10.. 1. (content validity) .. 2. (criterion-related validity) .. 11. 3. (construct validity) .. 4. (nomological validity): . (lawlike validity) .. 12.. 8 5 . Duane Davis (2004), Business Research for Decision Making , sixth edition, .. 13.. 1.. 2.. 3. (+). 4. (+). 5. (+). 6.. (-). (+) (+) .. 1.. 2.. 3. (-). 4. (+). 5.. (-). 1.. 2.. :+ - . 8 6 . Paul Peter & Gilbert A. Churchill (1986), Journal of Marketing Research , 23 (February), 14.. z . z . 15. 1/2. z ( ).

2 ( ) .. : .. 16. 2/2. z Cronbach .. Cronbach .. Cronbach . z . 17.. z . 30 . : .sav .. 1. Analyze/scale/reliability analysis 2.. 3.. 18.. 2.. 19. 1/2. z = .. Corrected Item-Total correlation Cronbach Alpha if item Deleted . Corrected Item- Total correlation .. 20. 2/2. 21. 1/2. z .. 22. 2/2. z . ( ) .. (factor loading) ( ) .. ( ) .. 23.. z .. 97 .. ( ..sav). 1. Analyze/Data Reduction/Factor 2.. 3.. 24.. 2.. 25. -1. z KMO= Bartlett's test Chi-sq = (Sig= < ) . KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy..806. Bartlett's Test of Sphericity Approx. Chi-Square df 105. Sig..000. z Principal components . eigenvalue>1 . Total Variance Explained Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative %. 1 2 3 4 5 6 .926 7 .730 8 .674 9 .603 10 .540 11 .462 12 .390 13 .334 14 .316 15.

3 269 26 Extraction Method: Principal Component Analysis. -2. z 6 . Rotated Component Matrix a Component . 1..814. 2..169. 3. 4. 5..767..742..143..245..115..102..372..13 5..110..673 .192 .154..493..460..457..398..120..409..168. .368..760..658..217..270..493 .608 .124 .235 ..207..824..665..121..202..128 .188..774..711..309..252 .273 .136 .710..113 .461 Extraction Method: Principal Component Analysis..646. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 15 iterations. 27. -3. z .. 28.. z .. 1. Analyze/scale/reliability analysis 2.. 3.. 29.. 2.. 30. -1. z .. = . z .. = . z . = . z . = . 31. -2. z . z .. 32. (pretest). 40~50 (more is better).. ( ).. (SPSS). 1.. 2. 27 73 . 3.. 4. t . 5.. SEM . 1. (Bollen, 1989). 2.. ~ (Noar, 2003). 3.. (Bollen, 1989). 4.. 5.. 6. 5 7 . 7. (Bollen, 1989). (composite reliability; CR). CR .. ( Hair,1997), Fornell and Larcker (1981) .. ( )2/. (( )2+( . )) (J reskog and S rbom , 1996).

4 (Average variance extracted AVE). AVE . VE .. Fornell and Larcker(1981) . ( . ) .. AVE= ( 2)/(( )2+. ( )). (J reskog and S rbom , 1996).