The purpose of this study is to analyze the statistical reasoning levels of preservice elementary school teachers. With this purpose, pre-service teachers consisting of 29 groups worked on a model eliciting activity (MEA) in scope of an elective course they were taking. At the end of the class, they were asked to present their solutions while working on the MEA in form of a detailed report. The data of the study consisted of these reports and solution sheets. Content analysis method was used in the analysis of the data. As a result, it was found that when the participants were asked to interpret a data set in a table, they could not establish a relationship between measurements of central tendency and variation, and their reasoning was limited or mistaken. The general tendency when pre-service teachers encounter a data set is that they think the only value representing the data set is the arithmetic mean. Additionally, it was found that, although the pre-service teachers were able to correctly compute the measures of variation such as standard deviation and interquartile range, they did not have sufficient knowledge about what these measures tell us about the variation of the data set.