This study set out to assess the following: (1) Relationship between Social psychologicalrnfactors (id entification and de-individuation) and fan aggression, (2) Difference in aggressionrnamong fans' different levels of identification (3) Differences in aggressive behavior of footballrnfans across s ocia-demographic facto rs. (4) Situational factors (venue, event, and staffrnvariables) predicting fan aggression, and (5) Association between de-individuation andrnsituational factors. A random sample of spectators (N = 220) at football league game in AddisrnAbaba Stadium were selected and completed a survey designed to identify or measure thernaforementioned variables.) The data was analyzed using s tatistics including, t-test,rndescriptive mean comparison, one-way analysis of variance, pearson correlation, linear andrnmultiple regression. Linear regression showed that fan identification predicted fanrnaggression. Consistent with previous studies, highly identified fans were more likely tornbehave aggressively at games than moderately identified fans and lowly identified fans .rnMultiple regression indicated that de-individuation components: anonymity, arousal, grouprnsize, and loss of self awareness emerged as significant predictors of fan aggressionrnrespectively. Venue-situational factors: stadium location, temperature, noise level, andrnseating arrangement were rated important variables respectively. The two club fans did notrnshow significant differences in the above venue variables. Event-situational factors: crowdrndemography, alcohol availability, performer behavior, event significance, and event durationrnwere found to be important in contributing to aggression. The two clubs fans significantlyrndiffered in event s ignificance, performance quality, crowding, and event duration among thernevent variables. Staff-situational factors: experience, communication, and training were ratedrnimportant contributors to aggression for both club fans. The t- test also revealed that the twornclubs fans significantly differed in communication among, presence, and experience of thernpolice officers in their contribution to fan aggression. From socio-demographic factors, agernand longevity of fandom were found to have significant effect on aggression. The researchrnresults indicated that those fans aged between 21-25 tended to show higher levels ofrnaggression than any other age category. Fans with 6-10 years of fandom tended tornexperience lower levels of physical aggression than fans with more than 15 years ofrnfandom. Analysis us ing a Pearson correlation matrix showed a pos itive correlation betweenrnde-individuation and situational factors: staff, event, and venue characteristics.