Purpose: The study sought to check the efficiency of Nairobi Securities Exchange with regard to dividend announcements on semi-strong form of efficiency. Methodology: The study employed event study methodology which is descriptive in nature. Census was carried out to determine which dividend announcements qualified for analysis. The period of study extended for five years from 2012 to 2016. Window period covered 30 days before and 30 days after the announcement. Average abnormal returns were evaluated for significance at 95% confidence level. Findings: The results indicated that the market was efficient for 4 years except one year where the market was found to be inefficient in semi strong form perhaps due to the prevailing economic conditions during the year. Unique contribution to theory, practice and policy: This study recommended that the NSE be checked for efficiency from time to time. This is informed by the fact that an efficient market allocates the resources optimally from areas they are less required to sectors they are highly required hence contributing to economic development. The study recommends regulatory bodies to come up with strategies to enhance and sustain efficiency at NSE.