This Digest describes matrix sampling of test items as an approach to achieving broad coverage while minimizing testing time per student. Matrix sampling involves developing a complete set of items judged to cover the curriculum, then dividing the items into subsets and administering one subset to each student. Matrix sampling, by limiting the…

Modified Parallel Analysis (MPA) is a heuristic method for assessing "approximate unidimensionality" of item pools. It compares the second eigenvalue of the observed correlation matrix with the corresponding eigenvalue extracted from a "parallel" matrix generated by a unidimensional and locally independent model. Revised…

The dominant issues in considering the use of tests developed outside the classroom to measure the impact of curriculum reform on higher order thinking are reviewed by a panel interviewed for this discussion. Panel members are: (1) Stuart Kahl, (2) Robert Linn, (3) Senta A. Raizen, (4) Lauren Resnick, and (5) Thomas A. Romberg. It is conceded…

A scaling procedure is proposed, based on item response theory (IRT), to fit non-hierarchical test structure as well. The binary scores of a test of English were used for calculating the probabilities of answering each item correctly. The probability matrix was factor analyzed, and the difficulty intervals or estimates corresponding to the factors…