CONSTRUCTION OF COMPREHENSIVE ACHIEVEMENT EXAMINATIONS FOR NAVY OFFICER CANDIDATE PROGRAMS
Abstract
Naval science achievement examinations were constructed with the intention of measuring integrated knowledge rather than mere memory for factual information. These examinations included a substantial proportion of items measuring problem-solving and decision-making ability. The OCS examinations were: (1) the time-limits were appropriate for measuring power rather than speed; (2) the tests met satisfactory standards of reliability and score dispersion; (3) the subject matter was heterogeneous; (4) a single general factor of achievement in OCS was indicated; (5) the test items showed a reasonably satisfactory distribution of difficulty; (6) a reasonably large number of items showed satisfactory validity against a criterion which consisted of the total weighted scores on all examinations; (7) high validities were found equally often for items measuring sheer factual information and those requiring problem solving and decision-making. The Wherry-Gaylord iterative factor-analysis procedure was applied to the OCS examinations. The following dimensions of achievement were identified: verbal reasoning, and judgement, spatio-temporal reasoning, and ability to remember factual information. The technique of individual operating characteristics was applied to the interpretations of factor keys. The graphical presentations of operating characteristic curves confirmed that the content of the examinations is extremely heterogeneous. OCS students were compared with NROTC students with respect to performance on naval science achievement test.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1953
- Accession Number
- AD0023118
Entities
People
- Benjamin Schohan
- John B. Caroll
Organizations
- American Institutes for Research