LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (August 2011)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Wed, 17 Aug 2011 05:43:07 -0700
Reply-To:   autumnwu <>
Sender:   "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:   autumnwu <>
Subject:   Help!SEM fitting index all failed,though reliability and validity test passed.Reason?!
Content-Type:   text/plain; charset=UTF-8

Dear all brilliant members,

I am a new analyzing thesis data. Im conducting a survey about people's behavior intention, but i came into a serious problem and i couldnt figure out where goes wrong though i tried other way test it again. I need your valuable opinion and kind help walking through this most difficult part for me.

SPSS briefly reslut:

Proposed model: Factor 1-8, 8 independent variables, Factor 9-11, 3 dependent variables. (number of measurement items in brace) Factor 1 (3) Factor 6 (3) Factor 9 (2) Factor 2 (4) Factor 3 (5) Factor 7 (5) Factor 10 (3) Factor 4 (4) Factor 5 (4) Factor 8 (4) Factor 11 (2)

1.Sample Collecting: Convenience sampling is adopted as a method collecting questionnaires.82 valid questionnaires are received. 2. Reliability test: Factor 2,3,4,5,6,7,8,10,11 which Cronbach' α > 0.8 are maintained, factor 1,9 which Cronbach'α around 0.6 are deleted. 3. Validity test: Measurement items which factor loading below 0.5 are not invovled. Besides, measurment item which has two valid value which above 0.5 is deleted.So 1 measurement items in Factor 3 is deleted and two measurement items in Factor 7 are deleted. After rotated SPSS again, all factor loading is above 0.5, but Factor 2 stays one factor, Factor 3,4,5 reducted a new one, Factor 6,7 reducted a new one, Factor 8 stays and Factor 9,10,11 reducted a new one. KMO and bartlett value all above 0.8. 4. Correlation test: All factor relations at 0.01 significance level under Bivariate Correlation test.

AMOS test: Modified model: Factor 1-4, 4 independent variables, Factor 5, 1 dependent variables. (number of measurement items in brace) New Factor 1 (4) Factor 3 (6) Factor 5 (5) New Factor 2 (12) Factor 4 (4) Notes for model: Minimum was achieved, Chi-square=1022.214 Degrees of freedom=431 Probability level=.000 RMR=0.112,GFI=0.572,AGFI=0.508,NFI=0.617,TLI=0.710,CFI=0.732

!!! The result in SPSS test is kind of ok, but why the test in AMOS all the fitting index are failed to meet the standardized level?! It has no relation with the size of my questionnaire. It gives almost the same result or even bit worse result when i doubled the number of my questionnaire. Also the proposed model is supported by previous study. I couldnt get where goes the problem what makes the modified model doesnt fit the test in AMOS. Any one could give a reasonable answer or suggestion that i could looking for the possible problem!! Thanks a lot for your time and patient!

-- View this message in context: Sent from the SPSSX Discussion mailing list archive at

===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

Back to: Top of message | Previous page | Main SPSSX-L page