The split-screen in the 2004 presidential debates helped Bush.

Yet it was the Bush side that wanted a single screen focused on the speaker, precluding reaction shots:

"Republicans thought they knew what they were doing by asking for single-screen, and the Democrats and all the pundits argued that it had hurt Bush because of the split screen. But the data shows that's not true," says Dietram Scheufele, a UW-Madison journalism professor. "It hurt Kerry quite a bit and didn't hurt Bush at all. The pundits didn't live up to reality."...

The study, published in the February issue of the journal Communication Research, put that assertion to the test, as 700 university students were asked to evaluate a five-minute-long debate clip in single screen and split-screen formats. The study was conducted in the two weeks prior to the 2004 election....

"The split-screen debates hurt Kerry and not really Bush," [Scheufele] says. "It was largely a function of what people thought about the two candidates in the first place. Split-screen coverage made Bush supporters more extreme in their support for the president and their opposition to Kerry. Kerry voters, on the other hand, didn't like Bush in the first place, but the split-screen coverage also didn't change much about their support for Kerry."

For Bush, the split-screen format shored up his base and helped him with GOP-leaning undecided viewers.

"When they saw Kerry on split screen and saw his smirks or writing something down in reaction to what Bush said, that produced a much more negative view towards Kerry," he adds. "People who leaned toward Bush in the first place felt even worse about Kerry."

The less Kerry the better!