Item Response Theory
- 1. Item Response Theory Rahul Bhojwani rahulbhojwani2003@gmail.com bhojw005@umn.edu
- 2. Different Test Theories: ● Classical Test Theory ● Item Response Theory
- 4. Let’s start with IRT Terminologies: ● Item ● Ability and Difficulty ● Scale
- 5. Type of Items: ● Dichotomous (True or False) ● Polytomous (Multiple Choice)
- 15. IRT entails 3 assumptions: ● A unidimensional trait denoted by 𝛉 ● Local independence of items; ● The response of a person to an item can be modeled by a mathematical item response function (IRF).
- 27. Adding more parameters ● 1 PL model ( Difficulty) ● 2 PL model ( Difficulty and Discrimination) ● 3 PL model ( Difficulty, Discrimination and Guessing)
- 43. Item Calibration: ● How to generate item and test taker difficulty from the response matrix: ○ Joint Maximum Likelihood (UCON) ○ PROX ■ http://www.winsteps.com/winman/estimation.htm ■ https://conservancy.umn.edu/bitstream/handle/11299/98553/v01n2p 281.pdf?sequence=1
- 44. Item Calibration: ● How to add new items to the scale: ○ Same students → New items ○ New students → New items ○ New students → Old items ● How to match two different scales:
- 45. Get scale of new test taker: ● Maximum Likelihood ● Bayesian model ● Link: https://www.rasch.org/memo23c.pdf ● Test Characteristic curve
- 52. Questions??
- 54. Extras: