The standard error of measurement usefully provides confidence limits for scores in a given test, but is it possible to quantify the reliability of a test with just a single number that allows comparison of tests of different format? Reliability coefficients do not do this, being dependent on the spread of examinee attainment. Better in this…

In most reliability studies, the precision of a reliability estimate varies inversely with the number of examinees (sample size). Thus, to achieve a given level of accuracy, some minimum sample size is required. An approximation for this minimum size may be made if some reasonable assumptions regarding the mean and standard deviation of the test…

In some testing programs an early item analysis is performed before final scoring in order to validate the intended keys. As a result, some items which are flawed and do not discriminate well may be keyed so as to give credit to examinees no matter which answer was chosen. This is referred to as allkeying. This research examined how varying the…

A monte carlo simulation was conducted to assess the effects in an adaptive testing strategy for test batteries of varying subtest order, subtest termination criterion, and variable versus fixed entry on the psychometric properties of an existent achievement test battery. Comparisons were made among conventionally administered tests and adaptive…

The optimal test length, using predictive validity as a criterion, depends on two major conditions: the appropriate item-difficulty rather than the total number of items, and the method used in scoring the test. These conclusions were reached when responses to a 100-item multi-level test of reading comprehension from 136 non-native speakers of…