Answers are grounded in this paper and cite their sources.
From one very useful point of view, granted certain reasonable restrictions, tests of the same length actually do have the same standard error(s) of measurement, which can thus be computed even before the test items have been written. Such standard errors are now in use with certain nationally distributed tests. Several puzzling questions regarding these standard errors are discussed and answers are given.
No takes yet. Share an insight, caveat, or question.
Frederic M. Lord (1956) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: