Table of Contents | Search Technical Documentation | References
A two-parameter logistic (2PL) model is used for the short constructed-response items that were scored as correct or incorrect. The form of the 2PL model is the same as that of the 3PL model, with the cj parameter fixed at zero. The fundamental equation of the 2PL model defines the probability that a person whose score on scale k is characterized by the unobservable variable θk will respond correctly to item j as
where
xj | is the response score to item j, 1 if correct and 0 if not; |
|
aj | where aj > 0, is the slope parameter of item j, characterizing its sensitivity to scale score; and |
|
bj | is the threshold parameter of item j, characterizing its difficulty. |
The probability of an incorrect response to the item is further defined as