Suggested do's and don'ts for future jury research: A swan song.

Abstract

I became involved in my first jury study (davis, Kerr, atkin, holt, & Meek, 1975) over 40 years ago, in 1972, working as a greenhorn research assistant in the lab of the late James h. davis (whose many significant contributions to our understanding of juries and other interesting groups are worthy of special note and praise; e.g., see Kerr & tindale, 2012). Like many young scholars, I got turned on to the study of juror and jury behavior and to the possibility that such work might contribute to an ideal of justice. over the past 40 years, I have never lost that enthusiasm or hope and have spent a fair fraction of my professional life delving into the jury's mysteries. I am now close to full retirement and so do not have quite the same vested professional interest in the future of the field that many do. Still, I care about that future and have some opinions (mostly contrarian ones, I am afraid) about how that future should be pursued. thus, I am honored and gratified to have my thoughts included in this volume and for the opportunity to offer a "swan song."

A meta-analysis of the emotional victim effect for female adult rape complainants: Does complainant distress influence credibility?

“…For example, if jurors are required to assess the honesty and reliability of the witness to judge credibility, credibility measures should include only these items. This would allow researchers to assess whether perceivers are using legal guidance in their credibility judgments, which is a research priority for criminal justice professionals (Kerr, 2017).…”

“…We recommend that researchers should use ecologically valid experimental protocols to replicate the effect of complainant emotional demeanor on credibility judgments (Bornstein, 2017; Koehler & Meixner, 2017; cf., Kerr, 2017; Kerr & Bray, 2005). Simulated decision-making study protocols are often criticized for lacking ecological validity, including inadequate sampling, inappropriate measurement of dependent variables, or missing investigative interview or trial procedure (Bornstein, 1999; Diamond, 1997; Koehler & Meixner, 2017).…”

“…We focused on studies which used an experimental methodology in this review. As the effect of complainant emotional demeanor on credibility judgments may inform policy changes, it is particularly important that the methodology used to investigate the effect is rigorous (Kerr, 2017). We make five suggestions for improving simulated decision-making studies for future research below including increasing internal validity, refining complainant emotion manipulations and credibility measures, performing direct replication studies, implementing ecologically valid protocols, and sharing data and study materials.…”

“…In rape cases, police officers, lawyers, judges, and jurors experience the complainant’s emotional demeanor in person during the investigation and trial proceedings (Maddox et al, 2012; Temkin & Krahé, 2008) over hours or even days (e.g., Konradi, 2007). This means written complainant evidence used in experiments lacks ecological validity (Kerr, 2017), so examining the effect of stimuli modality also permits an investigation of whether more ecologically valid experimental stimuli (video complainant evidence) influences the emotional victim effect. We therefore investigated the effect of stimulus modality, whether observers read about or saw the complainant, as a moderator in this meta-analysis.…”

“…In a meta-analysis, it is critical to consider whether study validity may cause an effect to be over- or underestimated (Higgins & Altman, 2008; Lakens, Hilgard, & Staaks, 2016). By examining whether the stimulus format moderates the effect-size estimate, we can also assess how accurately the experimental paradigm captures the real-life context of the effect of complainant emotional demeanor on credibility (sometimes referred to as ecological validity; Kerr, 2017). Similarly, we can examine a specific aspect of the external validity—whether the results will generalize to other populations—by exploring sample type as a moderator (Krauss & Lieberman, 2017).…”