Several studies have shown that, on average, women perform slightly better than men on free-response tests, while men perform slightly better on multiple-choice tests. Two advanced placement examinations, United States History (USH) and European History (EH), were chosen for study because previous studies have shown that sex differences on the free-response parts of these tests were non-significant, while sex differences on the multiple-choice parts were large. Several ratings and analyses were made of random samples of 400 free-response booklets from the 1986 administrations of both examinations. Samples of 199 males and 199 females for USH and 198 males and 199 females for EH were used. The following significant predictors were identified: (1) the advanced placement multiple-choice score; (2) historical content; (3) English composition quality; and (4) the number of words written. Regression equations were developed to estimate overprediction or underprediction by sex of free-response scores from the multiple-choice score alone and from multiple predictors. Sampled cases were matched with Scholastic Aptitude Test and English Composition scores. Results suggest that format effects are real and cannot be attributed to bias in scoring or to totally irrelevant variables. When scoring was conducted analytically with a focus on historical content, no sex differences were observed in the free-response portions. Males and females differ in how they respond to two different but legitimate ways of assessing history skills, raising the question of how much influence basic skills and aptitude should have on outcomes of assessment. Twenty tables present study data, and 7 appendices contain supplemental information, including 14 tables and 15 test rating forms. An 18-item list of references is included. (SLD)