Unit of observation:
Experiment A: individuals (turnout) Experiment B: precinct (persuasion)
Experiment A: 815 treatment precincts contained a total of 147,660 voters and 782 control precincts, containing 144,942 voters; Experiment B: The control group comprised 65,657 voters.
Experiment A: Precincts containing fewer than 100 or more than 700 Republican households; Experiment B: Precincts containing fewer than 100 or more than 700 Republican households. Republican households were identified based on whether a vote had been cast in a prior Republican primary election, and Perry supporters were identified by a micro-targeting firm.
Experiment A: Random assignment of precincts; Experiment B: Random assignmemt of households. Robotic calls were directed only to phone numbers belonging to individuals in the treatment group.
Governor Perry recorded an automated phone call on Willett’s behalf. This call was directed at more than 200,000 households that, according to micro-targeting data purchased by the campaign, were both likely primary voters and strong Perry supporters. The text of the automated message is as follows: "This is Governor Rick Perry. Tuesday is primary day and I want to encourage you to vote for my friend, Don Willett, for Supreme Court. I appointed Don Willett to the Supreme Court because he’s a proven conservative who understands that the role of a judge is to interpret the law, not create it from the bench. Keep Don Willett working for the people of Texas on the Supreme Court by voting in the Republican primary on Tuesday."
Automated phone calls (robo calls)
Voter turnout (mobilization) and persuasion
Shaw, Daron R., Donald P. Green, James G. Gimpel, Alan S. Gerber