Disclaimer. Don't rely on these old notes in lieu of reading the literature, but they can jog your memory. As a grad student long ago, my peers and I collaborated to write and exchange summaries of political science research. I posted them to a wiki-style website. "Wikisum" is now dead but archived here. I cannot vouch for these notes' accuracy, nor can I say who wrote them.
Gerber and Green. 2005. Do phone calls increase voter turnout? An update. Annals of the American Academy of Political Science.
Nonpartisan telephone appeals do nothing to boost voter turnout.
By including the placebo group, the authors can control for "reachability." If we just compare those who received a "get out the vote" (GOTV) appeal to those who didn't, GOTV appeals appaear to have boosted turnout from 48 to 64.5%, but this result is fallacious. If we compare turnout among those who received a GOTV appeal to those who received a placebo (blood drive) appeal, then the turnout rates are 64.5 and 67.2 percent (suggesting that GOTV appeals actually hurt turnout, but insignificantly). Thus, those who are "reachable" turn out more than those who are not. Being "reachable" makes you more likely to vote; being "reached" does not.
Similar design, but without the placebo. Methods were adjusted slightly as a result. Controls for whether the voter's local Congressional race was competitive. Finds that competitiveness doesn't matter; nonpartisan phone GOTV appeals don't matter.
Previous studies have found significant effects. But these studies all have a small N (which makes false positives more likely). In an earlier study (Gerber, Green, and Nickerson 2001), the authors noted a strong negative correlation between strength of effect and sample size, suggesting that the journals' bias against publishing insignificant results has led to only false positives being published. []
Research by the same authors
Research on similar subjects