# GLM output

#### andyw

##### New Member
I am conducting a series of GLMs in order to find the minimum adequate model, the largest model in which all factors are significant. In some of my models all factors are significant, except for the intercept... what does this mean??

thanks

andy

#### Mean Joe

##### TS Contributor
The intercept is often ignored.
Non-significant intercept means its 95% confidence interval covers 0.

#### Dason

That doesn't mean you should remove it from the model though.

#### andyw

##### New Member
cool thanks... so if i were to plot each of the significant terms from the GLM output then the intercept would have to be included for each one?

ie... the intercept is 0.25;
- factor x has coefficient of 0.5, thus a line of y=0.5x + 0.25
- interacting factor x*z has a coefficient of 0.4, thus a curve of y=0.4(x^2) + 0.25
- a further interacting factor x*z*a has a coeff. of 0.1; thus a curve of y=0.1(x^3) + 0.25

does that seem to make sense?

thanks

andy