The purpose of this study is to interpret the gains in Turkish eighth graders’ TIMSS mathematics performance between 2007 and 2011 in terms of the two national assessments conducted to place eighth graders into high schools in Turkey, namely OKS 2007 and SBS 2011. Content analysis technique was employed by three experts to analyze mathematics items in these national assessments. Items in the national assessments were categorized into the content domains (numbers, algebra, data & chance, and geometry) and cognitive domains (knowing, applying, and reasoning), which were described in the TIMSS 2011 assessment framework. As a result, it was found that the percentage of number content items decreased from 24% in OKS 2007 to 15% in SBS 2011 while the geometry content items increased from 36% in OKS 2007 to 50% in SBS 2011. On the other hand, there was an increase in the percentage of items in the knowing domain (32% in OKS 2007 to 50% in SBS 2011) and a decrease in the percentage of items in the other two cognitive domains. In terms of students’ mathematics performance, no significant differences were observed among the given years in numbers content scores of Turkish eighth graders while their scores in the other content domains improved, especially in geometry, as well as the improvement in their knowing cognitive scores between TIMSS 2007 and 2011. As a result, in terms of the content domains, this study determined that changes in the focus of the Turkish assessments in relation to TIMSS is likely a reason for the increase in geometry in comparison to number performance, with all respect to the other factors affecting student achievement. It also raises concerns regarding the distribution of the cognitive domains in the national assessments, considering the inconsistency of the distributions with students’ performances in the specific domains.