Victim impact statements and crime heinousness: a test of the saturation hypothesis

Abstract

In Payne v. Tennessee (1991), it was argued that the harmful effects of a victim impact statement (VIS) during sentencing are diminished as the heinousness of the crime increases. We tested this hypothesis by giving 180 mock jurors a summary of a murder trial that varied the presence/absence of VIS along with the crime heinousness (more/less). Participants gave significantly harsher sentencing ratings when VIS was present than when it was absent, and crime heinousness also significantly affected sentencing ratings. However, a significant heinousness by VIS interaction failed to emerge for sentencing, perceptions of victim suffering, or anger toward defendant.

Keywords:

• mock jurors
• victim impact statements
• crime heinousness
• death penalty

Notes

1. In actuality, all capital crimes can be considered to be heinous acts, although Rosen (1986) points out that 24 states include ‘especially heinous’ crimes as an aggravating factor. We chose to use the terms ‘less’ and ‘more’ heinous in keeping with the differences in our manipulation and at the suggestion of an anonymous reviewer, but the ‘more’ heinous condition in our study should not be interpreted as an additional aggravating factor not present in the less heinous condition.

2. The coroner testified that it was unclear whether the victim was still alive when this act was committed. We included this fact because we felt it added to the dehumanizing nature of the crime, which would be judged as more heinous, and because this same act was committed in South Carolina v. Gathers (1989) in which the defendants received the death penalty.

3. States typically (e.g. see Thompson, 1989) follow death qualification standards articulated in Wainwright v. Witt ( 1985 ). This criteria differs from Witherspoon criteria in that when using Witt criteria, jurors are asked whether their attitudes concerning the death penalty would ‘substantially interfere’ with their ability to find the defendant guilty, knowing that he could receive the death penalty if convicted. We chose to use the older Witherspoon criteria for a number of reasons. We wanted to compare our results to previous studies concerning death penalty judgments and a number of these qualified using the Witherspoon standard (e.g. Fitzgerald & Ellsworth, 1984; Myers et al., 2004). The Witherspoon standard is more clearly operationalized, and thus may be preferred for research purposes (Robinson, 1993). Guilt judgments were a moot issue in this study because guilt had already been decided, and so to avoid a large number of participants leaving the question blank because they felt the question was not relevant to them, we employed the Witherspoon standard. Finally, in previous studies where both criteria were used, small differences emerged in who was screened out of the sample (both in the number, and the actual individuals screened out) when we compared Witherspoon versus Witt standards (Myers et al., 2001).

Death qualification took place after completing sentencing rather than during juror selection, which is the practice during actual capital trials. This departure from actual trial procedures detracts from the generalizability of our findings. It is difficult to determine the extent to which this limitation may have impacted our results, but the percentage of our total sample who were removed (22%) after death qualification is similar to that found in studies (e.g. 25% for Butler, 2008) where a community sample was used and the qualification took place prior to presenting participants with trial stimuli.

4. As we will discuss in more detail in the discussion section, this small effect size led to low power levels and null findings. However, our sample size would have been adequate to detect an interaction effect of moderate magnitude.

5. We do not report gender differences because gender differences regarding VIS and capital sentencing: (a) has little practical relevance from a policy standpoint as gender differences in jurors cannot be addressed procedurally, and (b) past studies concerning VIS and capital sentencing have failed to find gender differences in response to VIS (e.g. Myers, Lynn, & Arbuthnot, 2002; Myers et al., 2004). Thus, we did not plan for gender differences here. Therefore, although our sample sizes were not sufficiently large to test for a three-way interaction involving VIS, heinousness, and gender on sentencing judgments, it is noteworthy that neither the main effect for gender, nor the two-way and three-way interactions produced effect sizes of any sizeable magnitude (i.e. all η ² values < 0.01).