AIC

AIC

ØAIC 

The AIC  table can be found under: View → Text Output → Model fit → AIC


AIC

Model

AIC

BCC

BIC

CAIC

Default model

129,234

111,234

234,567

345,678

Saturated model

134

133,445

345,678

456,789

Independence model

3432,101

3444,565

3344,22

3355,23

 

v AIC

AIC stands for Akaike Information Criterion (Akaike, 1987) and is used to measure the quality of the statistical model for the data sample used. AIC score useful only in comparison with other AIC scores of the same data set.

The lower the AIC value, the better.

AIC = 129,234

 

v BCC

BCC = Browne-Cudeck criterion used specifically to analyze moment structures and impose a larger penalty on less parsimonious models.

 

v BIS

BIC = Bayesian Information Criterion applies a larger penalty for complex models compared to AIC, BCC, CAIC and therefore has a greater tendency to select parsimonious models.

BIC = 234,567 for our tested (default) model.

 

v CAIC

CAIC = Consistent Akaike Information Criterion (Atilgan & Bozdogan, 1987) occurs only when means and intercepts are not explicit in the case of one group. CAIC applies a penalty to complex models that is higher than AIC and BCC, but less severe than BIC.

 

The reference values for these criteria depend on the sample size, number of parameters, and model complexity. As such, there are no fixed reference values. However, some guidelines suggest that models with lower AIC, BIC, BCC, and CAIC values are preferred over models with higher values.

Lower values of these criteria indicate better model fit.




Cite this article in your research paper:
APA

Statistische Beratung Leonardo Miljko (datum) How to interpret SEM model fit results in AMOS. Retrieved from https://www.StatistischeDatenAnalyse.de/images/services/How_to_interpret_SEM_model_fit_results_in_AMOS.pdf .

Harvard

Statistische Beratung Leonardo Miljko  January 10, 2020 How to interpret SEM model fit results in AMOS. viewed datum < https://www.StatistischeDatenAnalyse.de/images/services/How_to_interpret_SEM_model_fit_results_in_AMOS.pdf >


Wichtiger Hinweis: Der Originalinhalt ist auf Kroatisch. Die Übersetzung ins Deutsche und Englische erfolgte über einen Web-Übersetzer. Wir entschuldigen uns für die Fehler.

Kontakt

Statistische Beratung und Datenanalyse Leonardo Miljko gew.
Zum Alten Hofgarten 21,
DE-94405 Landau an der Isar
Tel.: +49-178-8955-751
Viber/WhatsApp: +387-63-997-996
Statistische Beratung und Datenanalyse Leonardo Miljko gew.
Erfahrungen & Bewertungen zu Statistische Beratung und Datenanalyse

 

Erfahrungen & Bewertungen zu Statistische Beratung und DatenAnalyse