(annotation appears in red)

The novel output generated by the IML program is printed after the results for the factor analysis (e.g., the scree plot, eigenvalues, pattern coefficients, etc.).

`**** BEGIN OUTPUT FROM PROC IML ****`

`Simplified Weights for Items / Factors`

`ITEM FSCOEF`
`MEQ1
0 1`
`MEQ2
0 1`
`MEQ3
1 0`
`MEQ4
1 0`
`MEQ5
1 0`
`MEQ6
0 0`
`MEQ7
1 0`
`MEQ8
0 0`
`MEQ9
1 0`
`MEQ10 0
0`
`MEQ11 0
0`
`MEQ12 0
0`
`MEQ13 0
0`
`MEQ14 0
0`
`MEQ15 0
0`
`MEQ16 0
0`
`MEQ17 0
0`
`MEQ18 0
0`
`MEQ19 1
1`

`The weights chosen by the researcher are printed
first. I used the IML program for computing coarse factor scores (see main
page) based on the regression approach to determine the weights. As
can be seen above, the coarse factor score estimates for the first factor
are computed by summing items 3, 4, 5, 7, 9, and 19. The coarse factor
score estimates for the second factor are computed by summing items 1,
2, and 19.`

`Indeterminacy / Determinacy Indices`
`(Multiple R, R-Squared, and Minimum Correlation)`

` FACTOR MULTR
RSQR MINCOR`
` 1
0.914 0.835 0.669`
` 2
0.902 0.814 0.628`

`The indeterminacy/determinacy indices for
the factors are reported above. As shown by the high MULTR and RSQR values,
the factors are fairly determinate. The MINCOR values are also fairly high.
The MULTR values will be specifically used below.`

`Validity Coefficients`

` FACTOR VALID
MULTR`
` 1
0.890 compare to MULTR -->
0.914`
` 2
0.821
0.902`

`The validity (VALID) coefficients for the
coarse factor score estimates compare fairly well to the MULTR values.
The second factor showed a greater loss of information than the first.
If the researcher decides that the loss of information is too great, the
coarse factor score estimates will have to be improved by either adding
or deleting items. Again, the IML programs for computing coarse factor
score estimates will have to be used in this process. If the scores cannot
be sufficiently improved, a refined set of factor score estimates may be
preferred or a new factor analysis may be attempted (e.g., using a different
extraction method, trying a different rotation, or selecting fewer factors).`

`Univocality Matrix`
`(Rows = Factor Scores / Columns = Factors)`

` UNIV
FACTCOR`
` . 0.552
compare to FACTCOR --> .
0.563`
` 0.588 .
0.563 .`

`The Univocality information indicates the
extent to which the factor score estimates are insufficiently or too highly
correlated with other factors in the analysis. The rows of
the above matrices represent the factor score estimates and the columns
represent the factors. A desirable outcome is obtained when the two matrices
above match perfectly. In this analysis the values in the two matrices
are fairly similar. For example, the correlation between the first and
second factor (see the FACTCOR matrix) is .563, which compares favorably
to the correlation between the factor score estimates for the first factor
and the second factor (r = .552 in the UNIV matrix).`

`Correlational Accuracy`

`SCORECOR
FACTCOR`
` 1.000 .
compare to FACTCOR --> 1.000
.`
` 0.600 1.000
0.563 1.000`

`The Correlational Accuracy information is
easier to comprehend than the Univocality criteria above. The FACTCOR matrix
again represents the correlations among the factors, and the SCORECOR matrix
represents the correlations among the factor score estimates. Clearly,
if the factor score estimates are adequate representations of the factors,
the values in the two matrices should match. As shown above, the values
do match fairly well, although the coarse factor score estimates for the
two factors are slightly more correlated than the factors themselves.`