Many Canadians identify with one party at the federal level and another in provincial politics. Split-level partisanship is far less frequent in the United States. Nevertheless, it is an open question whether such dual partisanship is traceable to the same determinants in both countries. Following Niemi et al. (1987), I examine whether split-level identifiers are less likely to participate in politics, as well as to have lower levels of efficacy than people who identify with one party (either being fully consistent or partially consistent). No support for either hypothesis is found. Split-level identifiers participate just as much as fully consistent and partially consistent partisans. Their efficacy is generally equal to that of other groups as well. The only exceptions suggest that split-level partisanship reflects citizens' political environments. If people face two very different party systems at the federal and provincial tiers, they are likely to have different patterns of identification regardless ...