Would be good to hear the possible causes of the randomness implied by the model. Surely, for any given person, there is no uncertainty on whether they can touch their earlobe, they either can or cannot. Thus, if our model was a complete description of the person's ability, it would deterministically predict the outcome. Thus, I assume that here we concede to the idea that Rasch model is not a complete description of the person's ability. It would be interesting to learn how much of real world information is sacrificed in order to adhere to the rules of this model, and whether it is possible to do better with the data at hand
Something that is not explained is where does the latent variable come from. From my understanding it is based on the rescaling of the sum of the items.
![[#2024MAMA] ROSÉ (로제), Bruno Mars - APT. | Mnet 241122 방송](http://i.ytimg.com/vi/dgGqD28J6aQ/mqdefault.jpg)
This video was great. I'm watching this now as a first-year doctoral student and the examples, images, and explanations have been very helpful
I love it... Nice job
Thanks for posting these. Great music!
Very good series on IRT....Very helpful to give a starting point
Would be good to hear the possible causes of the randomness implied by the model. Surely, for any given person, there is no uncertainty on whether they can touch their earlobe, they either can or cannot. Thus, if our model was a complete description of the person's ability, it would deterministically predict the outcome. Thus, I assume that here we concede to the idea that Rasch model is not a complete description of the person's ability. It would be interesting to learn how much of real world information is sacrificed in order to adhere to the rules of this model, and whether it is possible to do better with the data at hand
Ms. Cook mentioned that the range of logits are usually similar to Z-scores, so -4 to +4.
Something that is not explained is where does the latent variable come from. From my understanding it is based on the rescaling of the sum of the items.
Thanks a ton man...u saved my ass from getting kicked in my exam
It is not very clear that how you transfer responses (very low-very high) to logit.
For plotting VH you should take another item ...
Touching earlobe is not difficult item which can decide the high function of shoulder
why -4 to +4? why not -6 to +6 or for that matter why not -x to +y where x and y are two positive real numbers
2:13 😂😂😂
noooo! a pox