Two field experiments were carried out in Zarzoora Experimental Station at ElBeheira Governorate in 1998/99 and 1999/2000 seasons to studythe effect oftwo farming systems of land preparation (the con-ventional method of plowing and no tillage systems), four phosphorus fertilizer levels (zero, 15, 30 and 45 kg P205/ fed) and five weed con‌trol treatments (unweeded, Fusilade super 12.5 % EC, hand hoeing once, Fusilade + hand hoeing once and hand hoeing twice) under four statisti-cal procedures(analysis of variance, correlation matrix, multiple linear re-gression, path coefficient and principal component analysis ) on yield and its components of faba bean. Results indicated that fresh weight of grassy weeds and dry weight of total weeds in both seasons, fresh weight of total weeds and dry weight of grassyweeds in the second season were significantly re‌duced with tillage system. On the opposite, seed and biological yields/ fed of faba bean in both seasons, plant height, number of branches/ plant, number of pods/ plant and weight of 100 seeds in one season sig‌nificantly increased by tillage. Number of seeds/ plant and weight of pods and seeds/ plant were not significantly affected by tillage in both seasons. Fresh and dry weights of weeds were significantly reduced by in‌creasing P-fertilizer level in both seasons, except dry weight of total weedswhich was significantly reduced in the first season Growth, yield and yield components of faba bean were significantly increased by in‌creasing P- fertilization in both seasons, except the number of seeds/ plant which was increased in the second season only. Whereas, number of branches/plant and 100 seeds weight were not significantly affected in both seasons. Weed control treatments decreased significantly the biomass of weeds compared with the unweeded treatment in both seasons. Weed control treatments caused a satisfactory effect on all faba bean charac‌ters under study in both seasons. Hand hoeing twice aswell as hand hoe‌ing once + Fusilade herbicide recorded the highest values for seed and biological yields/fed in both seasons. The highest value of correlation coetficientwasfound between seed yield/ plant and number of seeds/ plant ( 0.991) in both seasons. Stepwise multiple linear regression and path coefficient analysis also agreed upon number of seeds/ plant and weight of pods/ plant as the major contributors to seed yield/ plant. The principal component analysis grouped the studied variables into two major components which alto‌gether accounted for 92.8% of the total variation. Thus, we recommend farmers to grow faba bean under tillage condition with phosphorus fertili‌zation and controlling associated weeds manually or chemically.