Measuring Level of Deviance: Considering the Distinct Influence of Goals and Tactics on News Treatment of Abortion Protests

Abstract

Research shows a pattern wherein the more deviant a protest group is the more critically it is treated. Level of deviance is often measured using a combined assessment of group goals and tactics. However, the separate influence of goals and tactics on news coverage of protest groups has not been explored. This study content analyzed newspaper articles (N = 168) of abortion protests in four newspapers (New York Times, Washington Post, Wall Street Journal, and Christian Science Monitor) from 1960 to 2006. Analyses revealed that group tactics—and not goals—emerged as the strongest predictor of how the group will be treated. Specifically, groups using more extreme methods were treated more critically with little influence of either legal position or public support. Implications of these findings for study of protest coverage and related measurement issues are discussed.

Notes

¹The 11th month was chosen to avoid periodicity for each prospective year from which articles were selected. Further, by using this as the sampling interval, we were able to collect data from all 12 months reducing the likelihood of any seasonal influence such as legislative cycles.

²We want to be clear in emphasizing that the shifts in those who oppose abortion under any circumstances have been slight and that the variation in opinion tends to be movement from those who favor abortion under any circumstances to those who favor it under limited circumstances (see Gallup, 2009).

^aPre-Roe n = 23 and post-Roe n = 77.

^b t = 3.03.

^cPre-Roe n = 40 and post-Roe n = 128.

^d t = 1.4.

^e t = 1.87.

³Some scholars have raised questions about the validity using polls to assess public opinion for volatile issues such as abortion that can be influenced by many external factors (Adamek, 1994). Further, Adamek critiqued the wording of such poll questions particularly focusing on the Harris question employed in these analyses. As a check on the validity of the Harris question we conducted additional analyses (not shown) using a poll question from Gallup (2009): “Do you think abortions should be legal under any circumstances, legal under certain circumstances, or illegal in all circumstances?” In each case using this question instead of the Harris question for our analyses (both for the bivariate and multivariate tests) replicates the same pattern demonstrated by the results using the Harris (2005) Poll question. In short, as support for abortion grew so did the level of criticism leveled at the pro-life movement. However, once included in the multivariate model, this relationship was no longer statistically significant. Further, like the Harris Poll question, the Gallup question is limited to the time period after the court decision as Gallup started including the question regularly in 1975.

^a n = 100.

^b n = 89.

^c n = 144.

^d n = 168.

*p < .05.

**p < .01.

*p < .05.

**p < .01.

*p < .05.

**p < .01.