A Conceptual Introduction to Item Response Theory: Part 6. Applications of IRT

This is the last of the 6 part module and it describes applications of IRT including: 1) scale selection and development, 2) calibrating different scales to a common metric, and 3) computer adaptive testing

[@logingeorge6194]

Just sat through all six of the videos. VERY HELPFUL. Graphics and examples are great and help with the learning. Humor is much appreciated too! :-) THANK YOU for making these available!

[@tombuis4438]

Thanks Karon that was a real pleasure! This is an excellent example of how to convey this kind of material in an engaging and easily digestible way that I also found to have a very kind and pleasant tone, like you want us to learn way more than you want to show us how smart you are 🤙🏼 low-ego teaching! Thanks again 🙋🏻‍♂️

[@itskindachanny]

Love the presentation (your unexpected remarks are hilarious), it's incredibly helpful to learn the overview of IRT!

[@BigPinkBootyDaddy]

a rare example of accessible psychometric terminology, indeed it is and so are your videos. thank you.

[@mattwood1339]

Thanks very much for this series! This conceptual walk-through has really helped me get my head around IRT.

[@nnette69]

all 6 Very help-full en very nice to listen to! and also great with some humor :)

[@janetong4467]

Excellent presentation and animation on a challenging topic! Thank you for sharing this module.

[@0501victoria]

Thanks for these videos. Really helped to hear the conceptual approach!

[@chun1702]

Thank you for doing this! Very interesting, nice presentation.

[@Zonno5]

I know this has been 6 years ago, but a series on differential item functioning would be useful as well.

[@trevorpope1913]

Thank you for putting together this excellent set of presentations.

[@pratyasheeojah6311]

Thank yo so much, for explaining IRT so easily.

[@wdpoison]

I just finished all the 6 parts. It helped a lot. Thank You.

[@mercedesmcbride8939]

Excellent series, Karon. Thank you very much!

[@cmoten3]

Thank you for these videos. I got a better understanding of the principles of IRT!

[@Zonno5]

Super useful and really intuitive. Thanks for sharing, really appreciate it. I would love some more maths and examples if you have them.

[@user-ds4lx9vi6h]

Thank you so much for this playlist.
On CAT, when someone gets a wrong answer, is the next question is "1 degree" lower in difficulty, or it "jumps" a few degrees down in order to save time/questions to determine the test-taker score?

[@rexyanhx]

Very helpful for beginners like me! Thank you so much Karon.

[@jassulychalendonayre8538]

This series have been really helpful. You said in one of the videos you were planning on expand the content, I'd be very happy if you do :) Would it be also possible to make videos about the mathematics and equations involved? I think in your style of explaining it would be so much easier to understand it.

[@anouschkaporte2478]

Thank you do much for your clear and funny explanation. You made IRT kinda fun!

[@rahulsachan5010]

very helpful. god bless you.

[@amandaharris887]

Very helpful. Thank you. One thing that causes confusion for me is the statement, oft repeated, that IRT is not sample dependent. Surely it must be dependent upon the sample used to define the initial parameters for both the levels of theta and the item parameters? I think there's something I'm missing here that I can't get my head around.

[@mpraveeng1972]

Really helpful to get the basic idea.

[@oddmed]

Thanks so much. You saved my life.

[@karoncook]

:)

[@linyuliao3417]

Thank you very much for the wonderful videos introducing IRT. It would be nicer if there are real examples showing how IRT works with relevant data and analytical tools.

[@Aj-cg6be]

linyu liao I am working on one, will keep you posted on the updates.

[@xgesaa]

Thanks a lot. Very easy explained

[@anuraagbhatt]

could you suggest me some books that might be available online for further study of the topic ?

[@sanand1987]

thanks Karon for the efforts. really informative. would like to ask you though if you can elaborate on how would the results of a test administered based on the IRT model be if the test was administered to a group of students on different days and the level of the difficulty varied clearly enough (based on the post test response from a number of students) on different days. can we expect a biasing in the results due to the day the students chose to appear for the exam?

[@francisschrag869]

Fran Schrag Very helpful module; the introduction of the math formula could have been enhanced, I think, if before telling us what each symbol meant, you'd spent a little more time explaining what the point of the formula was, maybe by drawing parallels with other formulas we might be familiar with. I also think at the end, you might have a summary of the advantages of IRT over the earlier test theory.

[@vajinder785]

thanks for video .it is nicely explained .but i still have one doubt .
what will happen if a person appears in a multiple choice questions based exam with good knowledge, marks some easy questions wrong and difficult one right .then what will be the interpretation .please reply.

[@karoncook]

In this case, the standard error of measurement would be large and the testing would be extended (assuming CAT was using a standard error stopping rule). Assuming the mistakes were in the beginning, the CAT algorithm should adjust with a pretty good estimate, though really poor guesses in the beginning can bias the final estimate somewhat.