Comparing how Clinton, Edwards and Obama all perform against Republicans in the general election in a single public opinion survey is certainly the best, quantifiable means of determining “electability” around. However, one of the difficulties in determining which of the leading Democratic candidates is the most “electable” is that few national polling organizations ask general election trial heats for Clinton, Obama and Edwards during the same survey. So, in an attempt to filter out the noise and come closer to finding an answer to the “electability” question that will almost certainly haunt the nomination campaign during the fall, I have produced charts going back to the start of April that show the results from all of the polls that ask general election questions for how all three of Clinton, Edwards and Obama perform against Rudy Giuliani. All polls are taken from Real Clear Politics:
Democrats vs. Giuliani, Table One
Rasmussen also conducted trial heats for the three Democratic frontrunners during this time period, but they were not all conducted at the same time. So, instead, here is the two-poll Rasmussen mean, according to polls taken from March 28-June 10:
Giuliani 46.5--46.0 Clinton
Giuliani 47.5--41.0 Obama
Edwards 48.0--43.0 Giuliani
Clinton’s deficit in general election matchups against Rudy Giuliani, relative to Obama and Edwards, has entirely disappeared in non-Rasmussen polls. While she once clearly performed worse than Edwards and Obama in non-Rasmussen general election trial heats against Rudy Giuliani, that is no longer the case. In fact, now she performs slightly better than Edwards and Obama in non-Rasmussen polls.
Different polls are stronger for some candidates than they are for others. For example, the LA Times / Bloomberg poll is far worse for Clinton than it is for Obama and Edwards. The reasons for this are not entirely clear, but are probably connected to an oversampling of certain demographics that are either very favorable or very unfavorable to a given candidate. This could mean an oversampling of independents (which would hurt Clinton relative to Obama and Edwards), an oversampling of women (which would help Clinton relative to Obama and Edwards), or an oversampling of young voters (which would help Obama relative to Clinton and Edwards). There does not appear to be a broad pattern consistent across more than a handful of polls.
Rasmussen is all over the map. Even as most other polls are showing Clinton improving against Giuliani relative to Obama, Rasmussen shows the exact opposite. Since mid-June, Rasmussen also shows Edwards as the most electable, while the other four polls taken during the same time period show the opposite. Once again, this creates serious problems when attempting to discern broad trends across all polls.
In short, across the board I don’t think that there is any clear evidence pointing to Hillary Clinton as less electable than Barack Obama and John Edwards at this time. Since June 11th, she only performs worse in the overall mean when all of Rasmussen’s four polls taken during that time period are included. This will further complicate attempts to take away her perceived advantage within the rank and file on “electability,” and thus further strengthen her position in the nomination campaign. If another candidate is going to pass her in the campaign, this does not strike me as a promising route for any candidate to pursue.