An experimental examination was conducted using a multiple-choice test for the purpose of constructing an examination in mathematics. Some 1,125 secondary school students from various school districts were chosen as candidates. The test consisted of twenty mathematical situations requiring five true or false answers for each test item. No time limit was specified for the completion of the test. Markers were instructed to mark each correct answer +1, each incorrect answer -1, and 'no response' scored zero. This system of marking received considerable criticism, therefore, the test papers were marked both on the 'right minus wrong' basis and by taking into account only 'right' answers. Results show consistent sex difference in favor of boys in terms of scores on the test. The multifacet test used in this experiment was shown to have a good level of reliability and an acceptable level of validity. However, there appeared to be a distinct possibility that an average score could be obtained by guessing alone. To avoid this possibility, the alternative answer of 'don't know' was suggested for future testing. The four appendices present some observations from the mathematics panel of the Ministry of Education, the preliminary test, the final test used, and tables of statistics. (LS)