Hi:

For those who uses SEM models, you may be aware that the chi-square test of fit is not properly distributed at small samples or in the case of complex models and requires rescaling (else the Type 1 error is inflated). We have updated the swain and swaini modules, which provide 3 corrections to the test.

Here is the description of what swain does:

swain and swaini provide a suite of corrections to the
chi-square overidentification test (i.e., likelihood ratio test
of fit) for structural equation models whether with or without
latent variables. The chi-square statistic is asymptotically
correct; however, it does not behave as expected in small samples
and/or when the model is complex (cf. Herzog, Boomsma, &
Reinecke, 2007). In situations where the ratio of the sample
size to the number of parameters estimated is relatively small,
the chi-square test will tend to overreject correctly specified
models. Applied researchers should thus use a statistic having a
more appropriate Type 1 reject error rate to judge model fit in
finite samples (Bastardoz & Antonakis, 2014, 2016).

The module provides three corrections: (a) the Swain (1975)
scaling factor (see Herzog & Boomsma, 2009; Herzog, et al., 2007);
(b) an empirical correction T(MLEC1) by Yuan, Tian, and Yanagihara
(2015), and (c) the F-test, which can be thought of as the small
sample correction to the chi-square test (McNeish, 2020).

It is available from ssc. If you have not installed the first version, just type "ssc install swain"

For those updating swain, type "ssc install swain, replace"

Thanks.
John Antonakis
__________________________________________ John Antonakis Professor of Organizational Behavior Director, Ph.D. Program in Management Faculty of Business and Economics (HEC) University of Lausanne Internef #618 CH-1015 Lausanne Switzerland Tel ++41 (0)21 692-3438 Fax ++41 (0)21 692-3305 http://www.hec.unil.ch/people/jantonakis Editor in Chief: The Leadership Quarterly __________________________________________