Internationally, mathematics education reform has been directed toward characterizing educational goals that go beyond topic/content/skill descriptions and develop students' problem solving. The Revised Bloom's Taxonomy and MATH (Mathematical Assessment Task Hierarchy) Taxonomy characterize such goals. University entrance examinations have been seen as one way of accomplishing these goals and influence learning, teaching, and assessment in mathematics. The present study analyzed mathematics items (N = 1077) in Turkish university entrance examinations in 1998-2013 and objectives (N = 621) in mathematics curricula in 2005, 2011, and 2013 to determine the extent to which they represent the dimensions/categories of these taxonomies and the degree to which items are aligned with objectives in terms of reflecting the dimensions/categories of these taxonomies. The findings reveal that the items demand, to a large extent, automated computational skills; this is also evident in the relevant mathematics curricula. Implications for practice are discussed and could play a role in reforming assessment.