In early-year engineering mathematics programmes in Dublin Institute of Technology (DIT), students were commonly required to attempt five out of eight questions in their end-of-year examination. As these questions were based on well-defined areas, it allowed students to omit certain topics and still perform impressively. This observation was re-iterated by the fact that the most common problem for which engineers sought help in the Maths Learning Centre last year was basic integration, with 56% coming from second or third year, at which stage they should be highly familiar with integration. One way to address the problem is by reducing (or even eliminating) choice questions on mathematics papers in earlier years: if the material covered is necessary groundwork for later years, it should not be possible for students to omit it entirely. In this study, we build upon the results of an anonymous survey to determine students’ opinions of reduced choice in early years. Within DIT, last year’s changeover from year-long to semester-long modules afforded a natural move from a three hour end-of-year exam to two two-hour end-of-semester exams. Two new approaches to choice were introduced by the authors. The first, with mechanical engineers, completely eliminated choice: students were required to answer all questions on the end-of-semester papers. The second approach, with building services engineers, replaced the final exam (a choice of five out of eight questions with question one compulsory) by end-of-semester exams featuring a choice of three out of four questions with question one compulsory. This greatly decreased choice without eliminating it. We compare and contrast both approaches, and address the question of whether reduced choice encourages students to learn difficult topics or adversely affects their marks.