1. SEM Analysis SPSS/AMOS

2. Ski Satisfaction Download, from BlackBoard, these files Boot up AMOS
SkiSat-VarCov.txt
SkiSat.amw
SEM-Ski-Amos-TextOutput.docx
Boot up AMOS
File, Open, SkiSat.amw
See my document for how to draw the path diagram.

3. Identify Data File
File, Data Files, File Name. Select SkiSat-VarCov.txt. Open.

4. View Data File
View Data.

5. Love-Ski Properties Right-Click on Love-Ski Select Object Properties
Notice that I have fixed the variance to 1.

6. Path Properties
Right-click on the arrow leading from SkiSat to snowsat. Select Properties.
Notice that I have fixed the coefficient to 1.

7. Set Analysis Properties
Minimization History
Standardized Estimates
Squared Multiple Correlations
Residual Moments
Modification Indices
Indirect, Direct, and Total Effects

8. Calculate Estimates
Proceed With The Analysis

10. View Text (Output) Result (Default model) Minimum was achieved
Chi-square = 8.814
Degrees of freedom = 4
Probability level = .066 No significant, but uncomfortably close
Null is that the model fits the data perfectly

11. Standardized Weights Estimate SkiSat <--- senseek .399 LoveSki .411
foodsat
.601
numyrs
.975
dayski
.275
snowsat
.760

12. R2 The last four are estimated reliabilities. Estimate SkiSat .328
Estimate
SkiSat
.328
dayski
.076
foodsat
.362
snowsat
.578
numyrs
.950

13. Standardized Residual Covariances
Looks like we need to allow senseek to covary with dayski and numyrs.
senseek
dayski
foodsat
snowsat
numyrs
.000
2.252
.606
.754
.193
.660
.567
.313
.308
2.337
.488
.707

14. Standardized Total Effects
LoveSki
senseek
SkiSat
.411
.399
.000
dayski
.275
foodsat
.247
.240
.601
snowsat
.312
.303
.760
numyrs
.975

15. Standardized Direct Effects
LoveSki
senseek
SkiSat
.411
.399
.000
dayski
.275
foodsat
.601
snowsat
.760
numyrs
.975

16. Standardized Indirect Effects
LoveSki
senseek
SkiSat
.000
dayski
foodsat
.247
.240
snowsat
.312
.303
numyrs

17. Modification Indices: Covariances
This is the Lagrange Modifier Test. It is a significant Chi-Square on one degree of freedom. The fit of the model would be improved by allowing senseek and LoveSki to covary.
M.I.
Par Change
senseek
<-->
LoveSki
5.574
1.258

18.  Fit Comparative Fit Index = .919.
CFI is said to be good with small samples. Fit is good if > .95.
Root Mean Square Error of Approximation = .110
< .06 indicates good fit, > .10 indicates poor fit 

19. Modified Model Added a path from SenSeek to LoveSki
LoveSki is now a latent dependent variable
Fixed the regression coefficient from LoveSki to NumYrs at 1, giving LoveSki the same variance as NumYrs.
I had noticed earlier that LoveSki and NumYrs were very well correlated.
Added a disturbance for LoveSki, as it is now a latent dependent variable

21. 2 has dropped 6.761 points on one degree of freedom.
Minimum was achieved
2(3) = 2.053
Previously 2(4) = 8.814
2 has dropped points on one degree of freedom.
Probability level = .562
Null is that the model fits the data perfectly

22. Standardized Residual Covariances
No large standardized residuals. 
senseek
dayski
foodsat
snowsat
numyrs
.000
.891
.024
-.075
-.013
-.440
-.255
-.005
.138

23. Fit
CFI = 1.000
RMSEA = 0.000 