|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Effect size vs. baseline model |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
i have a simple model: MODEL: fishery ON year; in which fishery is categorial. i am using ESTIMATOR = WLS the effect size of the variable is significant (-3.957) but the baseline model is rejected X^2 = 15.659 , df=1 p=0.0001 what is the correct statistical inference from the output. how can the effect size be significant and the model not? thanks in advance victor. |
|
|
Model fit and the significance of parameter estimates are not the same. If you have a model that does not fit the data, the results of this model should not be interpreted. |
|
Back to top |
|
|