1. Bousquet 2004, http://www.sptimes.com/2004/07/29/news_pf/State/GOP_ flier_questions_n.shtml.
3. Unfortunately, there is very little published research on aspects of election administration, like absentee ballot return rates, so it is impossible to produce a reliable national percentage of absentee ballots counted of those returned. The rates at which absentee ballots are counted appears to vary considerably across election jurisdictions, even within the same state: for example, according to data from California counties compiled by the U.S. Election Assistance Commission in its 2004 “Election Day Survey,” the rate of uncounted absentee ballots ranges from a few percent in many counties, to double-digit non-counted rates in places like Riverside County (11.1%), Sonoma County (11.8%), San Joaquin County (13.4%), Tulare County (17.4%), and Yuba County (27.4%); see http://www.eac.gov/election_survey_2004/statedata/Califo rniajurisdictions.html for these data.
4. See http://www.collinscenter.org/initiatives/initiatives_show.htm?doc_id= 105009.
5. For analysis of Georgia's transition to e-voting in 2002 and the immediate effect on residual vote rates, see Stewart (2004). Georgia's secretary of state, Cathy Cox, released data on 18 November 2004 depicting a drop in the presidential residual vote rate (or the percentage of ballots that showed no choice in the presidential race) from 3.5 percent in 2000 to 0.39 percent in 2004. The press release issued by Cox's office can be found at http://www.sos.state.ga.us/ pressrel/111804.htm; there interested readers can download data from the 2000 and 2004 elections (2004_pres_undervote_analysis.xls).
6. Tampa Tribune, 2004, 18. A parallel debate that we suspect will become much more involved in the near future concerns electronic statewide voter registration systems. The Help America Vote Act of 2002 requires that states move to centralized and computerized voter registration lists, but that same piece of legislation is silent as to the specifics of how these systems are to be developed, what standards they should meet, and how they might be tested and certified for use, raising concerns about potential problems with these new electronic voter registries (Alvarez 2005b).
7. There are also voting technologies that blur the distinction between punch card and optical scan ballots; an example is the “InkaVote” system used in Los Angeles County, California. There the voter has a ballot card that looks virtually identical to a punch card ballot, and the voter uses the same type of vote recorder device in the polling location to help identify the correct spots on the ballot to mark her vote preference, but instead of punching a hole in the ballot