This experiment was conducted in the spring 2007 at Ardakan in order to assess the genetic diversity (or variation) in morphological and phonological traits in 204 naked barley lines and to determine the affecting factors on yield by multivariate statistical analysis. Due to the high number of genotypes, measurements were based on planting rows without replication. Selection was according to the morphological and phonological traits. Genotypes showed significant differences in all measured traits especially for seed yield. The highest seed yield was obtained from genotypes by numbers 28, 31, 34, 36 and 37. Yield had the highest correlation coefficient with green cover percentage (r=%43), days to flowering (r=%39) and height (r=%38). According to the path coefficient analysis, the yield components including of tillers of seeds and 1000 seed weight showed positive direct effect on grain yield and 1000 seed weight showed the highest amont for it. Principal Component Analysis (PCA) showed 4 main components including HI, spike position relate to height, lodging percent and potential seed yield According to the factor analysis based on the first factor stem elongation was distinguishly separated from the other traits. So may low temperature had a major effect on it. Dendogram derived from cluster analysis based on squared euclidian distance divided the trait into two groups. Grain yield showed the most similarity with date of flowering. According to results of stepwise regression, emergence percentage, height and days to flowering caused the most variation in yield respectively, and they entered to the regression model. Therefore, in yield improvement approaches, selection can be done according to the vegetative traits including green cover percent, height and days to flowering.