the substantive requirements for uncorrelated error terms and
unidemensionality. Thus, the inability to achieve a nonsignificant
X2 can be discounted. Several alternative measures indicate good
fit. The four-dimension model is compared with a null model in
exhibit 5. The null model, which assumes that all observed variables
are independent, is a baseline with which other models can
be compared. The four-dimension model is a significantly better
fit than the null model, with a normed-fit index, A, (Bollen 1989,
271), of .84. Two alternative indicators of overall fit reported in
exhibit 5, goodness of fit (GFI) and adjusted goodness of fit
(AGFI), are .88 and .86 respectively, indicating good fit for the
four-dimension model.