Hello, everyone
Seeking some clarification regarding model fit through SPSS AMOS.
While performing an analysis for a variable with 5 items (each positively coded). It has been obtained that the values of factor loading as well as model fit except RMSEA are found to be good. The details are as follows:
N=361
Factor Loading Index
I1
Morning!
Actually you have a problem also with X2/df - it is too high. Both these indicators (also RMSEA) are sensitive to the sample size. Your sample size seems to be OK. So:
You can also think about reducing the number of relations tested, about introducing modification indices where appropriate. But before you do it, ensure that the quality of your data is OK (normal distribution, type - ratio and not ordinal).
I hope it helps.
Best,
Eugene
Hey Sayoni,
can you please provide information concerning the multivariate skewness and kurtosis? Chi-sq statistic, which is used to compute RMSEA, may be inflated by severe violations of multivariate normality. Normalized Chi-sq statistic (CMIN/DF ratio) seems also rather high (a rule of thumb is < 2.0 or less conservatively < 5.0).
Providing information about the items' metrics (minimum, maximum, type of scale) would be also helpful for choosing a proper estimation method for your data.
Best wishes,
Martin
RMSEA becmes trouble if it exceeded .05 , and in your case it is .133 which means there is error in the model.
i completely agree with Yevgen! you have a problem with CMIN/DF=5.64!! CMIN/DF: is a chi-square value correcting for sample size. This value should be less than 3 in a good fitting model. i suggest also to check the quality of your data (normal distribution, type - ratio and not ordinal or outliers). then retest your model! If the problem persists, model modification is required to obtain a better-fitting model. modification indices in AMOS allows to generate the expected reduction in the overall model fit chi-square for each possible path that can be added to the model.
regards,
Thank you, everyone (especiallyYevgen, Martin & Salah ) for clarifying my doubt....The data quality seems to be OK..I have made some changes with respect to the MI values accordingly and thereby value of CMIN/DF and RAMSEA comes to be within acceptable range..
Once again..thanks a lot..!!
the sample size is relatively good. however, check the number of factor extracted also. seems to be too many. therefore think to reduce them where applicable
my recent analysis with 410 children showed these results;
RMSEA 0.111
Chi2 30.160
CFI 0.94
To my knowlegde, these data has bad model fit. My question is what should i do next?
CFA or EFA? Or some other ways to solve this?
Hi Sayoni Santra ,
How were you able to reduce the RMSEA value. As i am facing the same problem though, the CMIN/DF is in acceptable range.
Dear Sayoni Santra ,
May i know how to Check the modification indices.
Steps for it and values checkings
Thanks ahead
Dear Mustafa,
In AMOS ,go to 'Analysis Properties', in that go to 'Output ' tab , there you will find modification indicies. Check it . Values will be checked as usual.
I was able to run an analysis on AMOS.
However, I have a low fit of my data to the model, and given time contrasts I doubt I would be able to double the number of data I got (I only have 217 responses..).
What could I do? At the moment I have a
- CMIN/DF: 18,5
-CFI: 0,46
-RMSEA: 0,285
I tried to go through the modification indices, but the only available covariance modification i could do between two 'errors' doesn't make sense and would only improve my model by 4,7.
Any suggestions?
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS