Rasch Analysis

Hello all,

someone asked me to look into Rasch Analysis. I have found a few articles that I am going to read but I was just curious if anybody is familiar enough with this technique to provide a very brief, high-level overview about this technique?

It sort of sounds like some combination of logistic regression and Factor Analysis ....but maybe not?
Last edited:


I'd say just logistic regression. It's a 1PL IRT model (though a Rasch purist would hate me for saying this). Basically it answers the question "Given your ability level what is the probability you'll answer this item correctly?" I learned about IRT models but not a lot about Rasch. Some say the IRT 1PL is the same as the Rasch but as I already stated a Rasch purist would argue differently. That's really all I know about Rasch models.

Spunky is a psychometrics person so hopefully he sees this and provides more insight.
After reading some articles and things online and based on Trinker's response above, I don't quite see how Rasch Analysis is terribly different from logistic regression. It seems like the main difference is that typically a logistic regression would measure some type of observable thing while Rasch Analysis is trying to get at unobservable or latent variables.

Can anybody elaborate any further on whether or not this thought is actually correct? Is Rasch Analysis simply just a logistic regression geared more towards latent variables?

I've been searching online for somewhere that does some type of walk-through or example of Rasch Analysis but have been unsuccessful at finding anything that better explains the steps taken for this statistical method. I'm curious if anybody has come across anything that provides a type of "walk-through"?


IRT (I take the stance that Rasch falls under IRT; but again some Rasch people would say I'm full of stuff) uses logistic regression but has three different formulas (depending on what model you choose (1PL, 2PL or 3PL)). I wouldn't say it is logistic regression, more it's a model that uses logistic regression. I'd recommend you explore IRT models more (1, 2 and 3 PL). But it's not too complicated, IMO.

old topic but I need some advice.
My prof wants me to do a Rasch Analysis for my paper. My aim is to check if there are some factors (like gender, age, etc...) that can explain if someone is willing to answer political questions or to refuse.
I got some nice data, but I'm not sure about the use of Rasch. I took about lets say 30 variables with political questions and recoded them (0 = no answer given, 1 = answer given). I read the STATA article about Rasch but I think this is rather confusing (http://www.stata.com/support/faqs/statistics/rasch-model/) Here, they want to figure out if some math questions are harder than others. Why do I need a complex model here? I'd just look at the descriptive statistics and check the percentage answered correctly. The lower the rate, the harder the math question?. Duh, isn't that kinda obvious? I stil dont get the use of the Rasch model here :/ any help?