A STUDY OF A NETWORK-FLOW ALGORITHM AND A NONCORRECTING ALGORITHM FORTEST ASSEMBLY

The network-flow algorithm (NFA) of Armstrong, Jones, & Wu (1992) and the average growth approximation algorithm (AGAA) of Luecht & Hirsch ( 1992) were evaluated as methods for automated test assembly. The algor ithms were used on ACT and ASVAB item banks, with and without error in the item parameters. Both algorithms matched a target test informatio n function on the ACT item bank, both before and after error was intro duced. The NFA matched the target on the ASVAB item bank; however, the AGAA did not, even without error in this item bank. The AGAA is a non correcting algorithm, and it made poor item selections early in the se arch process when using the ASVAB item bank. The NFA corrects for nono ptimal choices with a simplex search. The results indicate that reason able error in item parameters is not harmful for test assembly using t he NFA or AGAA on certain types of item banks.