In the eastern part of Ethiopia, land scarcity is one of the most important problems constraining food production. This problem has compelled farmers to practice monocropping and exposed the land to soil erosion and infertility. Therefore, it is important to practice a cropping pattern that enhances food production per unit area and also alleviates the problem of land degradation. Cassava (Manihot esculenta) could be a good candidate for this purpose since it is a drought tolerant crop that may produce higher yields per unit area. However, farmers in eastern Harerghe zone have no acquaintance with how cassava is processed into different food items. Therefore, the objectives of study were to evaluate performance of cassava and to determine the mixing ratio of sorghum-cassava flour for suitable food preparation in the study area. The treatments consisted of three cassava varieties (Kello, Qulle and local check). The experiment was laid out in randomized complete block design and replicated three times. Data were analyzed using the Generalized Linear Model of SAS Statistical Software package. Results revealed that there was significant (P<0.05) differences among the varieties for root diameter, root yields and number of roots per plant at Fedis and Erer. However, all others parameters did not significantly differ among the varieties. Kello and Qulle had increased root diameters by about 18.3% over the local check at Fedis while the local variety exceeded Kello and Qulle varieties in the number of roots per plant by about 14.1 and 36.7%, respectively. Kello and Qulle variety had superior root yields over the local variety by about 11 and 5 tons, respectively. There was also significant (P < 0.05) differences across the two locations for all parameters, except root length. The result showed that Qulle was more attractive in root color and ranked first among the evaluated varieties. Thus, it was found to be the most preferred variety in terms of sensory traits such as taste. Sole cassava flour was more preferred for cake making than the mixing ones for the three varieties. For porridge preparation, the ratio of 25 and 75% of cassava and sorghum flour, respectively, was the most preferred formulation for Qulle and the local variety while 50 and 50% of cassava and sorghum flour ratio was the most preferred formulation by farmers. According to attendants, Injera was not physically attractive when the ratio of cassava flour increased as compared to sorghum flour for the three varieties. Thus, sole sorghum flour was preferred by farmers for Injera making. In conclusion, Kello and Qulle variety performed well at Fedis and similar agro-ecology and they were most preferred by farmers for cake and porridge making.