Statistics courses have become an essential part of many programs in higher education. Thus,rnstudents from a broad spectrum of disciplines take statistics courses in higher education. Structuralrnequation modeling can be defined as a class of methodologies that seeks to represent hypothesesrnabout the means, variances, and covariances of observed data in terms of a smaller number ofrnâ€˜structuralâ€™ parameters defined by a hypothesized underlying conceptual or theoretical model. Thernmain objective of this study is to investigate the structural relationships among mathematicsrnachievement, attitudes toward statistics, and statistics outcomes by testing a structural model, calledrnâ€œStatistics Attitudes-Outcomes Modelâ€. In the current study, a survey design was adopted and arnquestionnaire was used to gather information from 205 undergraduate statistics non-major AddisrnAbaba University students who registered for introductory statistics courses offered by thernDepartment of Statistics in the first semester of the 2013/14 Academic Year. The hypothesizedrnmodel was tested by Structural Equation Modeling (SEM) technique using Smart PLS 2.0 statisticalrnsoftware. Participants of this study generally reported positive attitudes toward statistics except thatrnthey perceived the difficulty of statistics as neutral and they were indifferent in terms of theirrnindividual affection to statistics. Mathematics achievement variable had medium total standardizedrneffects on statistics outcomes variable; however, effort and interest variables had small total effectsrnon explaining statistics outcomes, and difficulty had non-significant total effect on explainingrnstatistics outcomes. The results indicated that, studentsâ€™ past and overall mathematics achievement,rninterest in statistics and the effort they expand to learn statistics were found as important factors forrnexplaining their statistics outcomes