Chickpea is one of the grain food legumes contributing an enormous amount of protein to the human diet in Southern Ethiopia. Though a lot of improved varieties were released by research centers farmers depend on low yield and local varieties. Hence, participatory variety selection is one of the methods used to evaluate varieties through involvement of users. Participatory Variety Selection (PVS) were conducted during 2015/2016 in Mirab Badwacho and Damot Fullasa districts of South region, Ethiopia to assess the performance of chickpea (Cicer arietinum L.) varieties and to evaluate farmers' selection criteria for chickpea. Six improved varieties with local check were laid out in a randomized complete block design with four replications. Significant variation among chickpea varieties were observed for most the agronomic traits collected except for number of pod per plant which was not significant. Concerning location, the majority of the traits were showed significant difference indicating dissimilarity in agro ecologies of the two districts. The study also revealed that in some cases the researchers' selection match with farmers' preferences. However, in general farmers have shown their own way of selecting a variety for their localities. These parameters include earliness, diseases and pest resistance, seed colour, branch number and length and seed size. Hence, including farmers' preferences in a variety selection process is a paramount important. Therefore, based on attentively measured parameters, farmers' favourites and the agro ecologies of the site the varieties Natoli, Dalota and Arerti are selected for the area. The varieties Habru and Ejere should also be given due consideration by farmers for its earlier maturity in the study area.