1
$\begingroup$

I am new to statistics and have read that this might be attributed to multicollinear varibles from "2 not defined because of singularities" and may be fixed with a dummy variable trap. Could it be because there are too few observations in those 2 categories?

enter image description here

Would this also work for the two way anova,is the problem the same?If so can some provide a step by step procedure for me; or a video link on how to fix it?

Here is plot of means if it helps as EDA

enter image description here

Improve this question
$\endgroup$
2
  • $\begingroup$ What does table(gravidity, method.delivery) look like? $\endgroup$
    – mdewey
    Commented Jun 3, 2018 at 12:48
  • $\begingroup$ @mdewey I added a graph of the plot of means if it helps $\endgroup$
    – glockm15
    Commented Jun 3, 2018 at 13:55

1 Answer 1

Reset to default
1
$\begingroup$

The graph now added clearly show that there are no observations in two of the cells (two of the breech cells) so the interaction cannot include them. This is probably nothing to worry about although much depends on the scientific question. One might want to know why only second and subsequent babies are breech but I know nothing of obstetrics.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Good point! There was only 1 observation for breech, guess I will just remove it $\endgroup$
    – glockm15
    Commented Jun 3, 2018 at 14:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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