Fit Indices Accepted Value Absolute Fit Measures

b.
Model Evaluation
Model evaluation is a process to verify each of latent constructs having factor loading
more than 0.7. Generally, the model evaluation also evaluates the model using some indices
of goodness-of-fit model such as χ2/df, GFI, and RMSEA (measures of absolute fit), AGFI,
CFI, NFI, IFI, RFI and TLI (assess incremental fit), PCFI and PNFI (measure the parsimony
fit) as depicted in the Table 4.7 below.
Table 4.7: Goodness-of-fit Indices for Measurement Model

Fit Indices Accepted Value
Absolute Fit Measures
Chi-square/df(X2/df) <3
GFI (Goodness of Fit Index) >0.9
SRMR (Standardized Root Mean Square Residual) <0.05
RMSEA (Root Mean Square Error of Approximation) <0.10
Incremental Fit Measures
AGFI (Adjusted Goodness of Fit Index) >0.80
NFI (Normed Fit Index) >0.90
CFI (Comparative Fit Index) >0.90
IFI (Incremental Fit Index) >0.90
RFI (Relative Fit Index) >0.90
TLI (Tucker-Lewis index) >0.90
Parsimony Fit Measures
PCFI (Parsimony Comparative Fit Index) >0.50
PNFI (Parsimony Normed Fit Index) >0.50
Source: Hair et al., (2010); Byrne (2010) and Hooper, et al., (2008).
1) Fit indices indicate the degree to which a pattern of fixed and free parameters
specified in the model are consistent with the pattern of variances and covariances from
a set of observed data (Suhr, 2008).The goodness-of-fit indices are split into three
categories: Absolute Fit Measures, Incremental Fit Measures and Parsimony Fit
Measures.
2) An absolute measure of fit presumes that the best fitting model has a fit of zero.
The measure of fit then determines how far the model is from perfect fit.
3) Incremental Fit Measures is analogous to R2 and so a value of zero indicates
having the worst possible model and a value of one indicates having the best
possible. On the other hand, a comparative measure of fit is only interpretable when
comparing two different models and it can be computed for a saturated model, and so
the saturated model can be compared to non-saturated models (Hair et al., 2010).
4) The chi-square for the model is also called the discrepancy function, or chi-square
goodness of fit. In AMOS, the chi-square value is called CMIN. CMIN is the minimum
value of the discrepancy. CMIN is a Chi-square statistic comparing the tested model
and the independence model to the saturated model (Byrne, 2010). Computation of
degrees of freedom is the estimation of the difference between the number of distinct
sample moments and the number of distinct parameters to be estimated. The number of
distinct sample moments includes variances and co-variances, while the number
of distinct parameters to be estimated includes several parameters that are
constrained to be equal to each other count as a single parameter. Parameters that
are fixed at a constant value do not count at all. This is why the number of distinct
parameters to be estimated could be less than the total number of regression weights,
variances, co-variances, means and intercepts in the model. Byrne (2010) revealed that
CMIN/DF is the minimum discrepancy divided by its degrees of freedom. This ratio
which is used as a measure of fit with the ratio should be close to one for correct
models. The output page in AMOS when referred to, shows what some previous
researchers explained the rule of thumb for this ratio such as Wheaton et al., (1977)
who suggested that the researcher also should compute a ratio of approximately 5 or
less as beginning to be reasonable. Carmines and McIver (1981) suggested ratios in
the range of 2 to 1 or 3 to 1 are indicative of an acceptable fit between the
hypothetical model and the sample data, Marsh and Hocevar (1985) suggested this
ratio to be as low as 2 or as high as 5 in order for it to indicate a reasonable fit;
Byrne (2010) suggested that the ratio > 2.00 represents an inadequate fit. However
this present study refers to the ratio of CMIN/DF as less than 3.
5) In AMOS, the GFI (goodness of fit index) concerns the minimum value of the
discrepancy function with where the minimum value of the discrepancy function minus
from value of 1. GFI is less than or equal to 1. A value of 1 indicates a perfect fit.
Therefore this present study estimates the accepted value for GFI to be more than
0.9.
6) SRMR (Standardized Root Mean Square Residual) or RMR (root mean square residual)
as the square root of the average squared amount by which the sample variances and
co-variances differ from their estimates obtained under the assumption that proposed
model is correct. The smaller the value of RMR is, the better. An RMR of zero indicates
a perfect fit. The accepted value for RMR in this present study is less than 0.05.
7) The Root Mean Square Error of Approximation (RMSEA) estimates lack of fit compared
to the saturated model. RMSEA of 0.05 or less indicates good fit, and 0.08 or less
adequate fit. The columns labelled LO 90 and HI 90 contain the lower limit and
upper limit of a 90% confidence interval for the population value of RMSEA.
PCLOSE is the p value testing the null that RMSEA is no greater than 0.05. Therefore,
this present study employs a model with a RMSEA greater than 0.1.
8) The AGFI (Adjusted Goodness of Fit Index), NFI (Normed Fit Index), CFI (Comparative
Fit Index), IFI (Incremental Fit Index), RFI (Relative Fit Index), TLI (Tucker-Lewis
index) under the Incremental Fit Measures. The accepted value would be >0.80 for
AGFI and the others to be > 0.90.Hair et al., (2010) explained AGFI (adjusted GFI) as
an alternate GFI index in which the value of the index is adjusted for the number of
parameters in the model. Furthermore AMOS output states that the AGFI takes into
account the degrees of freedom available for testing the model. The AGFI is bounded
above by one, which indicates a perfect fit. It is not, however, bounded below by zero,
as the GFI is. The Normed Fit Index (NFI) is simply the difference between the two
models’ chi-square divided by the chi-square for the independence model. The
Comparative Fit Index (CFI) uses a similar approach (with a non-central chi-square)
and is said to be a good index for use even with small samples. The accepted value for
CFI is along the values of IFI and RLI. The Tucker-Lewis index (also called the non-
normed fit index or NNFI), another incremental fit index, does has such a penalty. While
TLI is concerned with the ratio of chi-square to its degrees of freedom in Independence
model and the proposed model in the AMOS.
9) Parsimony Fit Measures is explained by a table that contains PRATIO, Parsimony

Normed Fit Index (PNFI) and Parsimony Comparative Fit Index (PCFI). The accepted
values for PCFI and PNFI in this study are > 0.50. The parsimony ratio expresses the
number of constraints in the model being evaluated as a fraction of the number of
constraints in the independence model. The parsimony ratio is used in the calculation of
PNFI and PCFI. The Parsimony Normed Fit Index (PNFI), is the product of NFI and
PRATIO, whereas PCFI is the product of the CFI and PRATIO. The PNFI and PCFI are
intended to reward those whose models are parsimonious (contain few paths).

Fit Indices Accepted Value Absolute Fit Measures

Uploaded by

Copyright:

Available Formats

Fit Indices Accepted Value Absolute Fit Measures

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fit Indices Accepted Value Absolute Fit Measures

Uploaded by

Copyright:

Available Formats

b.

Table 4.7: Goodness-of-fit Indices for Measurement Model

9) Parsimony Fit Measures is explained by a table that contains PRATIO, Parsimony

You might also like