The aim of this study was to show that the superiority of corrected-for-guessing scores over number right scores as true score estimates depends on the ability of examinees to recognize situations where they can eliminate one or more alternatives as incorrect and to omit items where they would only be guessing randomly. Previous investigations show examinees lack this ability. An instructional strategy for teaching examinees to use partial information in taking multiple choice tests under formula scoring conditions was tested on 280 fourth, fifth and sixth grade, private and public school students from the United States Virgin Islands. This method is outlined and followed by sections dealing with the administration of pre-tests and post-tests and the provision of training to selected groups (in the form of an algorithm for attacking test items, and practice sessions with immediate feedback). It is indicated in the final discussion of research results that the technique was successful in training examinees to omit only items for which they had no more than a chance probability of correctly responding. References, figures of the algorithm and pre-test practice items, and tables of descriptive statistics are appended. (Author/AEF)