Premium
Performance of a Proposed Method for the Linking of Mixed Format Tests With Constructed Response and Multiple Choice Items
Author(s) -
Tate Richard
Publication year - 2000
Publication title -
journal of educational measurement
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.917
H-Index - 47
eISSN - 1745-3984
pISSN - 0022-0655
DOI - 10.1111/j.1745-3984.2000.tb01090.x
Subject(s) - item response theory , computer science , psychology , multiple choice , statistics , econometrics , mathematics education , psychometrics , mathematics , significant difference
The error associated with a proposed linking method for tests consisting of both constructed response and multiple choice items was investigated in a simulation study. Study factors that were varied included the relative proportion of constructed response items in the test, the size of the year‐to‐year change in the ability metric, the number of anchor items, the number of linking papers to be reassessed, and the presence of guessing. The results supported the use of the proposed linking method, In addition, simulations were used to illustrate possible linking bias resulting from (a) the use of the traditional linking method and (b) the use of only multiple choice anchor items in the presence of test multidimensionality.