Talk:Rasch model
This is the talk page for discussing improvements to the Rasch model article. This is not a forum for general discussion of the article's subject. |
Article policies
|
Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL |
![]() | This article is rated C-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||||||||||||||||||||||
|
Untitled
editWhat about the contributions done by Erling B. Andersen?
psychometrika Volume 71, Number 1 / March, 2006
dichotomous data
edit"In general, the probability of a person responding correctly to a question with difficulty lower than that person's location is greater than 0.5, while the probability of responding correctly to a question with difficulty greater than the person's location is less than 0.5"
dichotomous is two choices, right? how can the probability fall below .5? surely no test is designed so that an ignorant choice scores less than random choice? am I missing something?
- a dichotomous variable can take on two values; e.g. 0,1. when people mark something correct or incorrect the score is a dichotomous variable. an example is a mathematics problem - there are many ways to get it wrong but (often) only one way to get it right, and the score of 0 or 1 is used to denote an incorrect or correct response. the dichotomy is right vs wrong. dichotomous does not imply two choices are available to a respondent in a multiple choice format. Holon 11:31, 2 May 2007 (UTC)
This probability is not the probability of a person marking randomly and happening to answer correctly, rather it is the probability that the person answered correctly when trying, or the probability that they know the answer and mark the correct answer for that reason. In Rasch analysis, these probabilities are computed against the probabilities of other persons answering the same and other questions correctly, and of the same person answering other questions correctly. If the person is determined to have a low probability of answering questions correctly across the entire set of items, then the probability that they will answer a hard question correctly can drop below .5. This can only happen when compared to scores on other items. If guessing is occurring, it will show up as a misfitting item or, if the person is guessing a lot, a misfitting person. Hope this helps!
The Rasch model is a special IRT model
editFrom the present text: "The Rasch model for dichotomous data is often regarded as an item response theory (IRT) model with one item parameter. However, rather than being a particular IRT model, proponents [see Wikipedia:WEASEL] of the model regard it as a model that possesses a property which distinguishes it from IRT models."
If we accepted this statement, then we would need to rewrite the article item response theory, redefining IRT models in quite a twisted way, excluding just the Rasch model from the otherwise open set of possible models. And all this for no other reason than an unexplained view of unnamed "proponents". Seems a very bad idea to me. -- Frau Holle (talk) 14:20, 12 August 2008 (UTC)
Response: There are crucial philosophical and practical differences between Rasch models (following the lead of Georg Rasch) and IRT models (following the lead of Frederic Lord). Some of these differences are tabulated at http://www.rasch.org/rmt/rmt193h.htm Martha Stocking summarized the philosophy behind IRT models as follows: "Building statistical models is just like this. You take a real situation with real data, messy as this is, and build a model that works to explain the behavior of real data." (New York Times, 2-10-2000). This is not the philosophy behind Rasch models. Rasch models embody separable parameters (each with a sufficient statistic), formulated for ideal probabilistic data, in order to extract from real data a set of linear measures as close to the ideal measures as is statistically possible. Winsteps (talk) 01:17, 22 October 2008 (UTC)
Interesting link, thank you. Nothing against an exposition of "philosophical" motivation & interpretation of mathematical models. At present, however, the articles Rasch model and item response theory have a verbosity that makes them hard to digest for mathematically-minded readers. -- Frau Holle (talk) 13:02, 23 October 2008 (UTC)
Scales image
editI was going to revert this edit but I wonder if the editor has a point. Should the image simply be removed?--A bit iffy (talk) 21:56, 13 November 2008 (UTC)
- Further to above, I've now simply removed the image as it seems pointless. It doesn't add meaning to the article.--A bit iffy (talk) 09:34, 15 November 2008 (UTC)
Intelligence citations bibliography for updating this and other articles
editYou may find it helpful while reading or editing articles to look at a bibliography of Intelligence Citations, posted for the use of all Wikipedians who have occasion to edit articles on human intelligence and related issues. I happen to have circulating access to a huge academic research library at a university with an active research program in these issues (and to another library that is one of the ten largest public library systems in the United States) and have been researching these issues since 1989. You are welcome to use these citations for your own research. You can help other Wikipedians by suggesting new sources through comments on that page. It will be extremely helpful for articles on human intelligence to edit them according to the Wikipedia standards for reliable sources for medicine-related articles, as it is important to get these issues as well verified as possible. -- WeijiBaikeBianji (talk, how I edit) 02:10, 2 September 2013 (UTC)