I have an experiment where I have 4 treatments and 2 outcomes (1 or 0) possible for each treatment. This is the result I calculated in R but I don't understand why only 3 of my treatments come up in the summary. How do I interpret this? The 4 treatments relate to nest sites that may either be a control, no egg, quail egg or artificial egg.

> fm <- glm(Disturbed~Trial+Treatment, data= data, family = binomial)
> summary(fm)
glm(formula = Disturbed ~ Treatment, family = binomial, data = data)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-1.4823  -0.6681  -0.3715   0.9005   2.3272  

                 Estimate Std. Error z value Pr(>|z|)   
(Intercept)        0.6931     0.5477   1.266  0.20569   
TreatmentControl  -3.3322     1.1711  -2.845  0.00444 **

TreatmentNo egg   -2.0794     0.8466  -2.456  0.01404 * 
TreatmentQuail    -0.8267     0.7536  -1.097  0.27263   
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 77.694  on 59  degrees of freedom
Residual deviance: 62.183  on 56  degrees of freedom
AIC: 70.183

Number of Fisher Scoring iterations: 5

Why is there no Artificial egg treatment?


If you have a factor with four levels you only have three degrees of freedom and so one level is chosen as a reference level against which all the others are compared. In this case the one first in alphabetical order has been chosen. You could relevel the factor so that control was the reference. In any event the intercept has the role of the reference level for a simple model like this.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.