The aim of the experiment was to determine the correlations between tuber yield and yield components and to measure the direct and indirect effects of yield components on tuber yield of potato genotypes. The experiment was conducted using eleven potato genotypes with one standard check variety (Belete) and a local check at Kulumsa Agricultural Research Center, Bekoji sub-station during 2020-2021. Data were collected for tuber yield, tuber dry matter content (%), plant height (cm), stem number, specific gravity, and phenological parameters. Since the two seasons had heterogeneous error variance, analysis was carried out separately. In 2020, correlation analysis indicated that the tuber yield was positively and significantly associated with stem height (0.608), marketable tuber (0.997) and positively associated with tuber dry matter content, stem number per plant, days to maturity, while it was negatively correlated with specific gravity, days to emergence and days to flowering at the genotypic level. Total tuber yield was positively and significantly correlated with days to maturity (0.640), stem height (0.791), marketable tuber (0.977) and stem number at the genotypic level. While specific gravity was positively and significantly correlated at the phenotypic level; and negatively at the genotypic level in the second season. Results of path analyses indicated that marketable tubers had the highest positive direct effect (+1.059) on total tuber yield followed by days to flowering (+0.057) and specific gravity (+0.026) in 2020 while marketable tuber (0.7107), stem height (0.2842), tuber dry matter content (0.0078), days to maturity (0.1250) had also positive direct effect in 2021. In addition, most of the indirect effects of yield components on tuber yield were significant and positive. Significant effects of marketable tuber, tuber dry matter content, days to maturity and stem height on total tuber yield, indicated that they could be used as the criteria for potato improvement and breeding programs.