In this research the researcher uses graphical, tabular and maximum likelihood estimations techniques to figure out the most important variables that affect students academic status. Both the graphical illustrations and the maximum likelihood results of ordered logit and probit estimates indicate that gender , age , previous academic background (i.e., students grade twelve university entrance result as a proxy), department choice have a significant effect on the academic status of student at a higher institution. The findings show that female and male students are not equal in their academic status. Female students have a higher probability to be academically poor than male given all other factors the same for both sex. One of the most interesting finding of the present study is the effect of department choice on students academic status. Forcing students to join a department without their choice will negatively affect students academic status. The other variable which directly affect the probability of academic status in our model is previous academic background using grade twelve university entrance exam result as a proxy variable. Based on the above findings of the current result gender support via tutorial should be given great emphasis for female students. In addition to providing tutorial, a special guidance and counseling program should be strongly organized and involved in the institution specially for fresh students. It would be better for female students to have a specific room for reading With in a library with Internet facility. Regarding department choice, even though it is totally impossible to assign students in their department choice in a developing country like Ethiopia due to limited resources, great emphasis should be given to give priority for students interest. This is because students choice makes students active participants in their educations, thereby increasing levels of engagement and increases their academic performance. Moreover, students academic background should be seriously emphasized in the beginning when students are assigned for each faculty and university.

Keywords: Keywords Academic Performance, Ordered Logit, Ordered Probit, Brant Test, Marginal Effects