There is a general concern in Trinidad and Tobago and the wider English-speaking Caribbean about the gender differential in performance on regional and national examinations. Previous studies had shown boys not performing as well as girls. Utilizing scores from the math component of the Continuous Assessment Programme (CAP), this study investigated whether the trend was evident among the younger primary school students. The final sample comprised 1,682 students in Standard 1, Standard 2, and Standard 3 (age group 7-9 years). Total scores, Zscores, scores for students attempting all items, and composite scores for the low-level and high-level items were computed. The proportion of boys and girls in the upperÂ (z greater than or equal to 1.00) and lower (z less than or equal to -1.00) tails of the distrubution were examined, as was the gender differential in the number of students omitting test items. Overall, girls scored higher than boys, more boys than girls omitted items, and a significantly greater proportion of boys were in the lower tail of the distribution. The Hindu schools were the exception to this general finding. The non-response to items could be addressed by schools teaching test-taking skills. However, more important may be the underlying reasons for the phenomenon, and the implications for boys' future academic achievement and employment opportunities.