Linking and Scaling
This chapter presents a number of methods for linking the calibration results from two or more tests. It covers two general cases. Linking when the test forms are designed to be equivalent and to be used with the same examinee population, and linking when the test forms do not have the same content and/or statistical specifications and the examinees are not from the same population. The first case corresponds to the case of test equating. The goal is to report information about the locations of examinees in the multidimensional space that can be used interchangeably independent of test form. The second case includes the multidimensional extension of vertical scaling and reporting results in the same coordinate system when test specifications change. The development of large item pools and item banks also may require the linking of calibrations from nonparallel test forms.
KeywordsTest Item Test Form Direction Cosine Item Parameter Item Bank
- Haebara T (1980) Equating logistic ability scales by a weighted least squares method. Japanese Psychological Research 22:144–149Google Scholar
- Holland PW (2007) A framework and history for score linking. In Dorans NJ, Pommerich M, Holland PW (eds) Linking and aligning scores and scales. Springer, New YorkGoogle Scholar