ANOTHER LOOK AT CLOSENESS AND TURNOUT - THE CASE OF THE 1979 AND 1980 CANADIAN NATIONAL ELECTIONS

Political scientists have long considered the relationship between the perceived closeness of an election and voter turnout. Downs (1957) and Riker and Ordeshook (1968) predicted that increased closeness would lead to increased turnout, largely because of voter reaction- Cox and Munger (1989) suggest, however, that any such relationship is likely to be the result of variation in elite activity (Key 1949) or in roll-off rates. This paper uses the 1979 and 1980 Canadian national elections to analyse these claims. The proximity of the two elections allows for a better estimate of the perceived closeness of the 1980 election. Analysis of the number of votes cast in the 282 Canadian ridings shows a strong relationship between closeness and turnout, which persists even when controls are imposed for variation in campaign spending, an important form of elite activity.