The Army Alpha (AA) is an intelligence test developed during WW1 by Robert Yerkes and first introduced in 1917. The AA represents a landmark in the history of intelligence testing, being among the first instruments ever designed for large-scale administration. The test was also validated on exceptionally large sample sizes; about 1.7 million men sat the AA during the war. Even after its retirement, it continued to influence the development of later standardized intelligence tests, including the SAT and AGCT.

This revised form of the Army Alpha is mostly the same as the originals, with some changes to accommodate the online adaptation:

• Subtest 1 (Following Directions) has been completely overhauled. Given the nature of the original subtest (which required physical writing, therefore making it difficult to automate), and serious questions about its psychometric reliability, the subtest has been replaced with one that is similar in nature but more friendly to an online administration.
• Subtests 2 through 8 are essentially unchanged. The total test itself consists of eight subtests, and there were originally 5 distinct Army Alpha forms. Each subtest was selected from the five original Army Alpha forms by identifying the form containing the most difficult overall set of items for each subtest. The resulting test is therefore a hybrid form of the most difficult subtests across Forms 5 through 9, rather than a total reproduction of any single original form.
• To help with discrimination in the higher ranges, each subtest was extended with more difficult ceiling items. Time limits were appropriately revised to account for the ceiling extensions.

The test is currently in a norming phase, but I'll try to have norms out when the sample size is adequate. As of now, you'll receive your total raw score and raw scores per each subtest upon completion. Your participation is greatly appreciated.