Third-Party “Hatchet” Ads: An Exploratory Content Study Comparing Third-Party and Candidate Spots From the 2004 Presidential Election

Abstract

The Supreme Court's 2010 Citizens United ruling lifted several key rules limiting electioneering communication. These changes are predicted to have significant effects on political campaigns. Namely, the ruling allows third-party sponsored electioneering up until Election Day. Because of the widespread presence of third-party issue advertising in 2004 presidential race, that election offers researchers one of the first opportunities to compare the content of third-party spots with candidate sponsored spots. This study examined the differences between third-party and candidate-sponsored spots, to look at differences in areas of “magic word” inclusion (e.g., “vote for …,” “vote against …”), negativity, and overall message consistency. Our findings show that few candidates use magic words, third-party spots were significantly more negative, addressed more issues than candidate spots, and made fewer explicit references to issues. Based on our results, we recommend future research on the effects of ad negativity, sponsor salience, and third-party and candidate message consistency.

Notes

¹“Third-party” is used to identify the producers of “independent” PAC, 527 and other non-candidate affiliated entities that advertise for the ostensible purposes of influencing the outcome of an election. We adopt this language from Kaid and Dimitrova's (2005) study The Television Advertising Battleground in the 2004 Presidential Election.

²According to the FEC, “electioneering communication” is any broadcast, satellite, or cable communication that refers to a clearly identified candidate, is publicly distributed shortly before an election, and is targeted at the relevant electorate; http://www.fec.gov/pages/brochures/electioneering.shtml

³Throughout the results section we refer to third-party spots as either pro-Kerry or pro-Bush. While evidence doesn't allow us to determine if the spot producers actually desired the success or failure of a candidate, the spots were deemed helpful to a candidate if they appeared to assist a candidate by promoting an issue with which a candidate was aligned or by attacking the issue positions of the opponent.

⁴Updated videostyle categories obtained from L. L. Kaid through personal correspondence.

^a n = 107.

^b n = 197.

^c n = 30.

^d n = 74.

^e n = 79.

^f n = 99.

*Chi-square indicates difference is significant at p ≤ .05.

**Chi-square indicates difference is significant at p ≤ .10.

^aThird-party spots, n = 109; Bush and Kerry spots, n = 173.

^bThird-party spots, n = 108; Bush and Kerry spots, n = 173.

^cThird-party spots, n = 108; Bush and Kerry spots, n = 172.

^dThird-party spots, n = 102; Bush and Kerry spots, n = 95.

^eThird-party spots, n = 109; Bush and Kerry spots, n = 173.

^fThird-party spots, n = 108; Bush and Kerry spots, n = 172.

^gThird-party spots, n = 109; Bush and Kerry spots, n = 173.

^hThird-party spots, n = 109; Bush and Kerry spots, n = 170.

*Chi-square indicates difference is significant at p ≤ .05.

**Chi-square indicates difference is significant at p ≤ .10.

^aPro-Bush third-party spots, n = 30; Bush spots, n = 74.

^bPro-Bush third-party spots, n = 30; Bush spots, n = 73.

^cPro-Bush third-party spots, n = 30; Bush spots, n = 73.

^dPro-Bush third-party spots, n = 26; Bush spots, n = 41.

^ePro-Bush third-party spots, n = 30; Bush spots, n = 74.

^fPro-Bush third-party spots, n = 30; Bush spots, n = 73.

^gPro-Bush third-party spots, n = 30; Bush spots, n = 74.

^hPro-Bush third-party spots, n = 30; Bush spots, n = 74.

*Chi-square indicates difference is significant at p ≤ .05.

**Chi-square indicates difference is significant at p ≤ .10.

^aPro-Kerry third-party spots, n = 79; Kerry spots, n = 99.

^bPro-Kerry third-party spots, n = 79; Kerry spots, n = 99.

^cPro-Kerry third-party spots, n = 68; Kerry spots, n = 51.

^dPro-Kerry third-party spots, n = 76; Kerry spots, n = 54.

^ePro-Kerry third-party spots, n = 79; Kerry spots, n = 99.

^fPro-Kerry third-party spots, n = 78; Kerry spots, n = 99.

^gPro-Kerry third-party spots, n = 79; Kerry spots, n = 99.

^hPro-Kerry third-party spots, n = 79; Kerry spots, n = 96.

*Chi-square indicates difference is significant at p ≤ .05.

**Chi-square indicates difference is significant at p ≤ .10.

⁵Negativity t tests: 527 spots (M = .32, SD = .16) compared with candidate spots (M =.22, SD = .17), p < .000; Bush 527s (M = .40, SD = .18) compared with Bush spots (M = .27, SD = .16), p < .000; Kerry 527s (M = .28, SD = .13) compared with Kerry spots (M = .19, SD =.17), p < .000; Bush 527s (M = .33, SD = .19) compared with Kerry 527s (M = .33, SD = .19), p < .001.

^aPro-Bush third-party spots, n = 30; Bush spots, n = 74.

^bPro-Kerry third-party spots, n = 79; Kerry spots, n = 99.

*Chi-square indicates difference is significant at p ≤ .05.

**Chi-square indicates difference is significant at p ≤ .10.

^aThird-party spots, n = 108; Bush and Kerry spots, n = 169.

^bThird-party spots, n = 109; Bush and Kerry spots, n = 173.

^cPro-Bush third-party spots, n = 30; Bush spots, n = 71.

^dPro-Bush third-party spots, n = 30; Bush spots, n = 74.

^ePro-Kerry third-party spots, n = 78; Kerry spots, n = 98.

^fPro-Kerry third-party spots, n = 79; Kerry spots, n = 99.

*Chi-square indicates difference is significant at p ≤ .05.

**Chi-square indicates difference is significant at p ≤ .10.

^aThird-party spots, n = 109; Bush and Kerry spots, n = 173.

^bPro-Bush third-party spots, n = 30; Bush spots, n = 74.

^cPro-Kerry third-party spots, n = 79; Kerry spots, n = 99.

*Chi-square indicates difference is significant at p ≤ .05.

**Chi-square indicates difference is significant at p ≤ .10.