[R-sig-ME] Logistic modelling with guessing parameter

[Ken Knoblauch]

Peter Harrison <pharr011 at ...> writes:
<< snip >>
> I've been using a script kindly posted by Ken Knoblauch 
back in 2010
> (https://stat.ethz.ch/pipermail/r-sig-mixed-models
/2010q4/004531.html) to achieve this using
> (1|p_ID). But when I do this, I get the following 
error message: (maxstephalfit) PIRLS step-halvings
> failed to reduce deviance in pwrssUpdate.


You ought to be able to use the link directly from 
the psyphy package.  A lot has changed with lme4
and a little with psyphy so that the modification from
then isn't necessary.

When I try your problem, I get
library(lme4)
library(psyphy)
data <- read.csv("https://dl.dropboxusercontent.com/
s/szq2e2sxwfsuxo6/data.csv", header = TRUE)
mod1 <- glmer(responses ~ accuracy + (1|p_ID), 
family = binomial(mafc.logit(2)), data=data)
summary(mod1)
Generalized linear mixed model fit by maximum likelihood (Laplace
  Approximation) [glmerMod]
 Family: binomial  ( mafc.logit(2) )
Formula: responses ~ accuracy + (1 | p_ID)
   Data: data

     AIC      BIC   logLik deviance df.resid 
  6742.6   6762.8  -3368.3   6736.6     6180 

Scaled residuals: 
    Min      1Q  Median      3Q     Max 
-3.9571 -1.1363  0.4520  0.6233  0.9299 

Random effects:
 Groups Name        Variance Std.Dev.
 p_ID   (Intercept) 0.4454   0.6674  
Number of obs: 6183, groups:  p_ID, 229

Fixed effects:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept)   4.1656     0.2771   15.03   <2e-16 ***
accuracy     -5.6355     0.3772  -14.94   <2e-16 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Correlation of Fixed Effects:
         (Intr)
accuracy -0.971

with no errors

> Thanks so much!
> Peter
>