Letting them speak: Analysis of victim impact statements in capital sentencing proceedings

Letting them speak: Analysis of victim impact statements in capital sentencing proceedings

Victim impact statements are much less instigative than commonly thought, with prosecutors having a great influence over the witness statements. This is the bottom line of a recently published article in Psychology, Public Policy, and Law. Below is a summary of the research and findings as well as a translation of this research into practice.

Featured Article | Psychology, Public Policy, and Law | 2018, Vol. 24, No. 4, 474–488

The heterogeneity of victim impact statements: A content analysis of capital trial sentencing penalty phase transcripts


Bryan Myers, University of North Carolina Wilmington
Narina Nuñez, University of Wyoming
Benjamin Wilkowski, University of Wyoming
Andre Kehn, University of North Dakota
Katherine Dunn, American University


Victim impact statements (VISs) are controversial in capital sentencing proceedings largely due to their questionable relevance to sentencing, the concern that characterizations of the victim may lead to arbitrary sentencing judgments, and the belief that the emotional nature of this evidence may be inflammatory. A sample of 192 capital trial VIS transcripts was analyzed for content as well as a linguistic analysis of emotionality. The findings reveal that these statements are highly varied, including their format, length, and relation between the witness and the victim. Despite a legislative mandate that they address the emotional, financial, and physical suffering experienced by victim survivors, testimony of this nature occurs in a minority of the cases. Most commonly, these statements tend to characterize the victim and their qualities, relay the witness’ shock at first learning of the victim’s death (i.e., trauma narratives), and address the significance of the deceased to the family unit. In approximately one third of the transcripts reviewed, the witness made mention of the defendant, but this rarely included any mention of a desire for vengeance or recommended punishment. Linguistic analysis revealed that emotional content was prevalent throughout the testimony, with sadness emerging as more pervasive than anger. However, the degree of emotional language contained in these statements was not particularly high—and was comparable to that typically encountered in everyday life (e.g., newspapers, novels). Implications, particularly with regard to the potential for VISs to be considered inflammatory, are discussed.


victim impact statements, capital sentencing, content analysis, juror decisions, emotions and judgment

Summary of the Research

“Victim impact statements (VISs) refer to statements given either in writing or orally that detail the impact of the defendant’s crime on victims. In capital trials, this testimony is delivered by victim survivors who have included members of the decedent’s family, but they have also been presented by friends, coworkers and even emergency first responders. VISs are often described as educating the jury concerning both emotional as well as financial hardships that have arisen as a direct consequence of the loss of the victim. […] VISs in capital sentencing proceedings are controversial for a number of reasons, but chief among these issues are their relevance to the sentencing decision and the potential that their emotional nature may interfere with jurors’ capacity to decide in a reasoned and impartial manner.” (p. 474)

“Some early research has indicated that factors surrounding the victim such as their likability has been related to punishment or judgments relevant to sentencing. This pattern has been uncovered in actual trials where less reputable victims promote greater leniency toward the defendant. In studies explicitly examining VISs and victim qualities, we again see evidence that qualities of the victim matter to jurors.” (p. 475)

“Researchers have therefore sought to determine the effects of VISs on sentencing judgments, with a particular focus on whether VISs elicit strong emotions in jurors and whether these emotions are related to sentencing judgments. Early studies in this area failed to uncover a clear pattern suggesting that VISs elicit strong emotions and that these emotions interfere with reasoned decision making. The failure to uncover these effects, however, may have been due in part to a failure to distinguish between emotions of anger and sadness. […] The importance of measuring discrete emotional states finds support in social cognition research on emotions and judgment where repeated findings have emerged that anger and sadness produce differential effects on judgments. […] The research on mood and judgment fail to confirm the broad characterizations voiced by legal commentators who describe all emotions as antithetical to reason.” (p. 476)

“Jury simulation studies that have assessed discrete emotions have tended to produce findings consistent with those in the emotions and judgment literature whereby anger is associated with more punitive responses in jurors. […] The findings thus far provide empirical support to the contention that (a) VISs do elicit emotions in jurors, and (b) anger and not sadness has been linked to sentencing decisions.” (p. 475)

“This empirical evidence addressing the effects of VISs that elicit emotions in jurors is critical to determining whether VISs can be considered inflammatory. However, it is equally important to determine whether emotions such as anger and sadness are typically present in VISs. How typical is emotional content in VISs? In addition, while simulation studies show the importance of victim character on sentencing, it is not clear that character information frequently accompanies VISs. The present content analysis of transcripts of VISs is designed to address these gaps in our information on VISs. Moreover, a content analysis of VISs also affords the opportunity to judge whether simulation research conducted thus far, in general (and not just the research addressing emotionality and character), adequately reflect the VISs jurors typically encounter in actual trials.” (p. 475)

“We examined admissibility status of VISs for all 31 states that currently allow the death penalty, relying on Sanderford (2012) and www.deathpenalty.org as sources. […] Given the wide latitude with regard to the format of VISs, the number of witnesses who may offer testimony, and the general lack of guidance or restrictions surrounding who may testify, the length of their testimony, or the content of the testimony, it is clear that VISs have the potential to vary greatly from case-to-case. It is therefore critical to carefully examine these differences in the administration of VISs and the content of VISs to better establish the degree to which the jury simulation research adequately reflects current practices.” (pp. 476–477)

“In the present study, we obtained trial transcripts from a large number of capital trials (n = 131) which yielded a large number of VIS transcripts (75 cases with 192 VIS statements). One goal of the present investigation was to get a clearer picture of common elements found in VISs in capital sentencing cases. […] A second goal of the present study was to address the degree of emotionality contained within VISs.” (p. 477)

“In addition to some demographic and procedural questions (e.g., gender of witness, relation of witness to victim, VIS length, manner in which the VIS was introduced), the present investigation further identified a number of content areas that are of greatest relevance to the VIS debate: (a) financial harm, (b) physical harm, (c) psychological harm, (d) family significance of victim, (e) victim character, (f) expressed emotions, (g) referencing the defendant, (h) dehumanizing the defendant, (i) describing the crime, and (j) recommended punishment.” (p. 478)

“Analysis of 192 capital trial VIS transcripts revealed substantial variability in the statement content, highlighting the fact that no single study on VISs could adequately represent the information jurors experience in capital trials.” (p. 483)

“With regard to witness characteristics and procedural issues surrounding VIS presentation, the findings revealed that witnesses were nearly three times as likely to be female than male, they were most frequently a parent, and the question-and-answer format occurs approximately 2.5 times more often than the free-narrative format. Surprisingly, the free-narrative format is not more common than a postsentencing VIS delivered to a judge with the jury out of the room.” (p. 483)

“In terms of some commonalities in content, looking across all formats, VISs regularly contain characteristics of the victim, and expressions of the emotional suffering the victim survivors have experienced arises frequently. […] the analysis of the 192 VIS transcripts revealed that any mention of the defendant, including the witness using dehumanizing terms regarding the defendant as well as any recommendations for punishment, are very rare events—happening in less than one in every 10 cases. As a general rule, VISs tended to focus on the qualities of the victim that will be missed, and how the survivors were dealing with their grief. […] none of the content categories differed in frequencies across the sentencing outcomes. In fact, in almost all cases, life and death cases were remarkably similar in the VIS content. The lone exception was the likelihood that the testimony included how the survivor learned of the victim’s death. These narratives were more than twice as likely to arise in death cases than in cases leading to a life sentence, although this difference failed to reach statistical significance.” (pp. 483–484)

“In the present content analysis, mention of the victim and their personal qualities was common. In more than half the transcripts, the description of the victim focused chiefly on how their loss was likely experienced by the surviving family. […] Descriptions of the victim’s reputation in the community, or their social standing, were infrequently present in the VIS testimonies; occurring in less than one in every six transcripts. […] Looking across VIS formats we see that descriptions of the victim characteristics occur significantly less frequently in cases where the jury is not present. Perhaps letting the jury see the victim as more than a “faceless stranger” may be regarded by victim survivors as more pertinent than providing this information before a judge.” (p. 484)

“The data reveals that in approximately two thirds of the cases we reviewed, witnesses sought to convey the emotional suffering they experienced as a result of the crime. The emotional devastation associated with the death of a loved one is frequently conveyed through a description of the experience in which the witness first learned that the victim had been killed. These trauma narratives are relatively descriptive and delivered in a manner that promotes the likelihood that the jury will take their perspective and imagine what it would be like to learn this tragic news. In many cases, this testimony was prompted by the prosecutor, who specifically asked the witness to relay this information. Thus, these trauma narratives were more likely to be present in the question-and-answer VIS format.” (p. 484)

“A linguistic frequency analysis revealed that emotional content regularly occurred throughout VIS testimony, and sadness terms were 44% more prevalent than angry terms in these testimonies. However, the overall level of emotionality in these transcripts indicated that less than 0.5% of all words were angry, and only slightly more than 0.5% were sad. Consequently, concerns that the overall level of emotionality present in VISs is excessive could benefit from some perspective. […] While emotional language does find its way into VIS testimony, the overall level of anger and sadness does not distinguish it from levels present in other contexts encountered in everyday life.” (p. 484)

“With the exception of postsentencing testimony where the witness addresses the court and the defendant, but after the jury has already rendered a sentencing recommendation, a VIS was significantly more likely to contain sad content than angry content. […] The tendency for the prevalence of anger and sadness to vary as a function of whether the testimony is given pre- or postsentencing may be partly a function of the degree to which courts have limited the content of the testimony when presented in front of jurors, and partly a function of who the testimony is directed toward.” (p. 484)

“The relation between sentencing outcome and the degree to which the VIS terms reflect anger and sadness was examined, and there was little evidence in the present study to indicate higher levels of anger are associated with greater frequency of death penalty judgments. Indeed, the zero order correlations suggested that sadness might be a predictor of sentence, and it was positively related to life-sentences. This is inconsistent with the jury simulation research findings, and the small sample of independent cases and our inability to control for a number of important predictors of sentencing would suggest that future research is warranted.” (p. 485)

“In conclusion, the present investigation suggests that jury simulation research on VISs deviate from the “typical” VIS testimony in a variety of ways including the relation between the victim and the witness, the number of witnesses, and the format in which the VIS is presented. The question-and-answer format is most common, and suggests the prosecutors who direct the questioning have great influence on what information the jury hears during a VIS. In most instances, they direct witnesses to convey the emotional toll experienced by the victim survivors in dealing with their loss, and convey to the jury the personal qualities of the deceased. Very rarely do they address the defendant or allow witnesses to vent their anger toward the defendant. Anger is far less prevalent in VIS testimony than is sadness, and when anger is more common, it typically occurs when the jury is not present and therefore not in danger experiencing the prejudicial effects of this testimony.” (p. 486)

Translating Research into Practice

“The findings here suggest a number of changes that would be necessary for simulation research to better represent VISs that occur in actual trials (e.g., witness characteristics, examples of content). However, it is also important to identify areas in which existing policies concerning VISs may be discordant with the research findings, as well as highlighting findings that support existing practices. The chief issue of contention regarding VISs in capital sentencing concerns its relevance to the sentencing process.” (pp. 485–486)

“Whether victim character information should be heard and whether it is appropriate for testimony that is emotional to be present when jurors decide punishment are legal questions that cannot be answered by jury decision making researchers. Instead, the appropriate role here is for psychology to inform the debate by answering relevant empirical questions such as whether victim character information influence how jurors decide punishment, and which specific emotions lead to harsher sanctions. The findings in the present investigation are meant to shed some light on whether the kinds of studies researchers are conducting in the laboratories are reflective of the kinds of VIS evidence typically present in the courtroom.” (p. 486)

“The present findings indicate that while emotional language does populate VIS testimony, sadness is present much more commonly than is anger, which was encountered in less than one half of one percent of all words in the VIS.” (p. 486)

“If courts are concerned about the emotionality of VISs, one practical way to limit the degree to which anger is introduced into the courtroom is through the format by which the VIS is presented. As we noted, free-narrative format is much less common than a question-and-answer format whereby the prosecutor directs the testimony of the witness. Linguistically, anger was most common in the postsentencing format when the jury had already made their sentencing recommendation to the court. Anger was much lower during the penalty phase, with the lowest anger scores occurring for the question-and-answer format. Limiting VISs to this format may be one approach to limit the anger expressed in the testimony, and this format also affords the judge an opportunity to halt proceedings before a witness answers if the question posed by the prosecutor is judged to likely promote an angry response from the witness (e.g., if a prosecutor asked witness if they had something to say to the defendant). Consequently, this format allows victims to speak before the jury, but it does not give them free rein. Limiting this opinion testimony may allow the court to limit the degree of anger present in the VIS.” (p. 486)

“Our present findings suggest that any mention of the defendant is extremely rare, and calls for punishment are rarer still. Less can be said about the potential impact of this information on jurors, as researchers have still not systematically varied the opinions offered for punishment to determine if they do indeed influence sentencing judgments. However, it is perhaps worth noting that a VIS itself typically fails to produce strong effect sizes in simulation studies, and the most reliable effects occur when VISs provoke an anger response in participants. Going forward, systematically studying how statements directed at the defendant that specifically request harsh punishments would better inform decisions regarding the prejudicial effects of this testimony. Importantly, in the present investigation, any comments directed specifically at the defendant occurred vary rarely, and when it did, it was most likely when the jury was not present to hear the testimony.” (p. 486)

Other Interesting Tidbits for Researchers and Clinicians

“The manner in which we structured this content analysis merits attention. A top–down approach to the content analysis (sometimes referred to as directed or deductive) was used in the present investigation. […] This can be contrasted with a bottom–up approach (sometimes referred to as inductive, conventional, or grounded theory) where the materials would first be examined for important themes that arise, without any preconceived notions about the content areas examined. […] it could be argued that our analysis missed some important themes that characterize the nature of VISs in capital sentencing proceedings.” (p. 485)

“For the present investigation, this directed approach focused on the analysis of specific content that would best inform researchers who investigate the effects of VISs on jurors, and that would address issues of contention among legal commenters with regard to the constitutionality of VISs. […] An approach driven by the content of the transcripts rather than the specific a priori factors identified by investigators would likely yield a different picture of what constitutes typical VIS testimony.” (p. 485)

“A second limitation with the present investigation was the sampling method used to obtain VIS transcripts. The sample obtained was not random, but rather was a convenience sample drawn from all death penalty cases tried that year. […] An additional limitation is that while the outcome of the case (life or death) was not discussed as a hypothesis for our study with coders, they were nevertheless not kept blind to the outcome of the case.” (p. 485)

“Our study focused on VISs in death penalty cases that have been tried in the United States. Thus, we cannot speak to VISs that might be presented in noncapital cases or VISs presented in other countries that allow VISs (e.g., Australia, or the United Kingdom). However, with regard to VISs that are presented in death penalty cases, our work may be particularly relevant, given that the United States is the only known country that allows both the death penalty and VISs.” (p. 485)

“Lastly, our analysis of the level of emotionality in VISs fails to address subtle factors that likely play a significant role in how moving testimony may be for a given juror. […] This is an initial attempt at trying to measure the content of VISs, and so we encourage further research that takes alternative approaches to measuring the emotional impact of VISs.” (p. 485)

Join the Discussion

As always, please join the discussion below if you have thoughts or comments to add!

Authored by Kseniya Katsman

Kseniya Katsman is a Master’s student in Forensic Psychology program at John Jay College of Criminal Justice. Her interests include forensic application of dialectical behavior therapy, cultural competence in forensic assessment, and risk assessment, specifically suicide risk. She plans to continue her education and pursue a doctoral degree in clinical psychology.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.