TBILISI (Reuters) - U.S. data and polling firm Edison Research said on Friday that the differences between an exit poll it conducted of Georgia's parliamentary election on Saturday and the official results of the vote "cannot be explained by normal variation alone and suggests local-level manipulation of the vote."

Georgia's electoral commission has said the ruling Georgian Dream won with 54% of the vote, a result that opposition parties and the country's president have said is fraudulent.

Edison's exit poll, commissioned by the pro-opposition television channel Formula, showed Georgia's four main opposition parties on course to achieve a majority in the election.



(Reporting by Felix Light; Editing by Tom Hogue)