Winbugs and Jags free Item Response Theory from the dot matrix plots of proprietary software and open up a multicoloured world of posterior predictive model checking. Fitting IRT models using brute force is not for the impatient, however. That’s why, just as early psychometricians shipped off their calculations to teams of monks. I’ve shipped off my model fitting to...

Read more »

Ten reasons to love the cloud:

1. The names on Heroku. empty-moon-9726. Suddenly the cloud seems prosaic when you can have the moon.
2. Being reunited with putty. It’s like catching up with an old friend. At first the conversation is a little stilted, but then you soon remember just how well you got on and what brought you together in the first...

   Read more »

Psychometrics, Qu’est-ce que c’est?

Say psychometrics to people and they think IQ tests. Fair enough.

I think eRm:

1
2
3
4
5
6
7

# Rasch model with beta.1 restricted to 0
data(raschdat1)
res <- RM(raschdat1, sum0 = FALSE)
print(res)
summary(res)
res$W

The joy of fitting your first Rasch model in R is unparallelled. Go on try, it. Hmmm, a list of numbers. No idea what they mean?

ok. so you take an IQ...

Read more »

Ever since R was born (evoked?) geeks have been trying to get it to talk HTML. A list of web interfaces for R is updated on CRAN here. Aims are various. Some seek to replace R with a traditional GUI. Others are more ambitious and open up a glimpse of an architecture that provides live analysis of ever...

Read more »

So you want to run R in the cloud so you can set your Gibbs sampling off, forget about it, and not be paranoid about power cuts and reboots. Andrew Gelman hosted a good debate on the pros and cons of R in the cloud on his blog.

The consensus seems to be RStudio and EC2. P.S. If...

Read more »