Regression modelling error message 'response must be numeric'

Machine Learning and Modeling
1

I'm trying to analyse if Class (Working_class, Middle_class) has an effect on whether subjects use standard or non-standard pronouns (specifically standard 'you' and non-standard 'yous'). I was getting the error message 'response must be numeric', so I converted the pronouns in my data to you=0 and yous=1.
I've tried both linear and logistic regression:

data.model<-glmer(IV_yous~Class+(1|Informant), data=my.dfyou)

This gets the error message:
Error in mkRespMod(fr, REML = REMLpass) : response must be numeric
In addition: Warning message:
In glmer(IV_yous ~ Class + (1 | Informant), data = my.dfyou) :
calling glmer() with family=gaussian (identity link) as a shortcut to lmer() is deprecated; please call lmer() directly

data.model<-lmer(IV_yous~Class+(1|Informant), data=my.dfyou)

This gets the error message:
Error in mkRespMod(fr, REML = REMLpass) : response must be numeric

Do I need to convert my Class data to 0 and 1 too? Thanks.

2

I think you need a family = binomial argument to ask the function to do logistic regression.

2 Likes
3

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.