Item response theory made easy with user-friendly jMetrik software | 1PL, 2PL, 3PL & 4PL

Подписаться 18 тыс.

Просмотров 10 тыс.

50% 1

This videos demonstrates how to fit 4 item response theory models using jMetrik. I will show how to fit a 1-parameter logistic model (1PL), 2PL, 3PL, and 4PL. In addition, the correlations between ability measures estimated by the models as well as some useful graphs will be presented.
Useful links:
Download jMetrik: itemanalysis.com/jmetrik-down...
What can gaze behaviors, neuroimaging data, and test scores tell us about test method effects and cognitive load in listening assessments?: journals.sagepub.com/doi/10.1...
Book: www.routledge.com/Applied-Mea...
Paper: www.tandfonline.com/doi/abs/1...
A comprehensive review of Rasch measurement in language assessment: Recommendations and guidelines for research: journals.sagepub.com/doi/full...

Опубликовано:

7 авг 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 26

@segsb7085 3 года назад

This was great! Thank you so much..very detailed and easy to follow!

@13sonal Год назад

Thank you very much for a detailed and informative video. I was trying to install extensions in SPSS for ITR but was not able to. This gave me a much appreciated alternative.

@rishiekumarloganathan907 2 года назад

Thanks sir...I would like to ask regards d chi square in determining model fit and how to calculate -2LL for each model

@zohregh2213 2 года назад

Thank you for this informative tutorial. I was wondering what do you recommend for missing data? Thanks

@VahidAryadoust 2 года назад

IRT is sensitive to missing data, but Rasch measurement is quite tolerant of it.

@zohregh2213 2 года назад

@@VahidAryadoust Thank you so much, one more question is how do you interpret the result if the questionnaire is a health-related one, let's say for example if it's to measure the severity of disease?

@VahidAryadoust 2 года назад

@@zohregh2213 I am not sure about that. I suggest you ask a medical expert.

@sharmaparsoon007 3 года назад

Thanks it was a very informative tutorial, how can do analysis if student has skipped the question?

@VahidAryadoust 3 года назад

You can replace them with '0' or just leave the missing cell in your data. I will be ignored.

@mikekalkbrenner4518 Год назад

Thank you so much for creating this helpful video. When I try to run the analysis, it's "estimating parameters" for 10+ hours and never produces the results. I tried using two different computers. Does anyone have any advice or suggestions?

@VahidAryadoust Год назад

possible causes: miskeyed-in data (check your data); too much missing data; very very large sample size; misspecification of the parameter.

@carinasong9237 3 года назад

Thank you so much, Prof. Vahid. But I have a question. Is there any possibility to conduct 2-PL or 3-PL analysis for polytomous scores or/and rater severity in Jmetrik?

@VahidAryadoust 3 года назад

You can do 2PL/3PL for polytomous data, but I am not sure if rater severity can be handled using jMetrik. Please see my videos on Facets. best wishes

@carinasong9237 3 года назад

Sorry, I didn't make it clear. I wonder is that Rasch Model (also 1PL) test many facets like examinee ability, rater severity, task difficulty, criterion difficult .... 2PL/3PL only test two facets--- item difficulty and examinee ability with item discrimination parameter and guessing parameter added? Is there any way testing both many facets and item discriminition and guessing parameter?

@carinasong9237 3 года назад

@@VahidAryadoust BTW. I've conducted a Rasch analysis before. Your Facet video deepened my command of Facet a lot!

@VahidAryadoust 3 года назад

@@carinasong9237 great! best wishes.

@VahidAryadoust 3 года назад

@@carinasong9237 Facets will not be the suitable software for that type of analysis. I am not aware of IRT 2PL or 3PL with a facets format. sorry.

Год назад

Best wishes from Brazil, master. Is there any possibility to set a score system from 0 to 1000 on IRT model? I need to give a score for a classroom of 45 students, they did a test with 45 items and I wanna personalize their scores equal our ENEM (SAT). If so, can you give me a brief configuration on jmetrix? tks in advance.

@VahidAryadoust Год назад

Hi, Theoretically, it is possible, but as the sample size is small in your case, I imagine it will yield high standard errors.

Год назад

@@VahidAryadoust Tks professor any video showing the steps?

@doctorresearcher7550 2 года назад

Hi there! I am running the data, and I did all the steps which you mentioned but I am not getting the results, it is running parameters for over an hour. I had closed the tab and ran other things like frequency etc and those things are perfectly running. Kindly help

@VahidAryadoust 2 года назад

It seems that the parameter estimation does not converge. I am not sure about the reason though. Try using a different sample and re-run the analysis.