Structural Equation Modeling: Interval Estimation - Quick Tab

Noncentrality Interval Estimation and the Evaluation of Statistical Models

Select the Quick tab of the Structural Equation Modeling: Interval Estimation dialog box to access options to implement the methods for confidence interval estimation discussed in Browne and Cudeck (1992), Steiger and Fouladi (1997).

Chi-Square. In the Chi-Square box, enter the observed chi-square statistic, computed by the structural equation modeling program (or perhaps reported in an article or book).

Df. Enter the degrees of freedom for the chi-square test in the Df box.

Total N. Enter the total N for all groups in the analysis in the Total N box. If there were two or more groups, this value is the sum of the sample sizes.

Conf. Level. In the Conf. Level box, enter the confidence level for the confidence interval. Note that confidence limits are presented for three quantities (See Noncentrality-Based Indices of Fit):

RMSEA. These upper and lower limits are for a confidence interval on the Steiger-Lind (1980) RMSEA.

Gamma1. These upper and lower limits are for the population Gamma1 coefficient developed by Steiger (1989). This coefficient (see Equation 107) is the population equivalent of the Jöreskog-Sörbom GFI index of fit.

Gamma2. These upper and lower limits are for the population Gamma2 coefficient developed by Steiger (1989). This coefficient (see Equation 110) is the population equivalent of the Jöreskog-Sörbom (1986) adjusted GFI index of fit.

Manifest Vars. In the Manifest Vars., field enter the number of manifest variables actually included in the model being tested in each group in the design. Currently, SEPATH performs multiple group modeling only on covariance matrices with equal numbers of variables.   For example, if you perform a three group experiment involving five manifest variables in each group, this number is 5, not 15.

No. of Groups. In the No. of Groups field, enter the number of independent groups on which the analysis is based.