Eyewitness testimony by individuals who stammer: Evidence, experience and perceived credibility · Maras K, Park S, Grafton P et al.

STUDY 3

Study 3 aimed to experimentally confirm findings from the qualitative analysis in Study 2, which indicated that providing information about stammering would improve credibility ratings of PWS.

Study 3: Method

Design

Utilizing a between‐subjects design, mock juror participants viewed a video of a witness with a stammer and were randomly assigned to one of two conditions: (i) ‘information’, in which they were informed that the witness had a stammer and were provided information about stammering; or (ii) ‘no information’, in which they received no information about stammering. Dependent variables were mock jurors' ratings of perceived credibility across four subscales: confidence, likeability, trustworthiness and capability. We also included a measure of participants' knowledge of stammering to confirm that the provision of information about stammering improved knowledge about the condition in the first instance.

Participants

An a priori power analysis using G*Power 3.1 (Faul et al., 2007) indicated that to detect a medium‐large effect size (based on Hobson et al., 2022; Maras et al., 2018), a minimum sample size of 126 was necessary to achieve Cohen's (1988) recommended power of .80. An opportunity sample of 162 mock juror participants eligible for UK jury service (18–75 years old and UK citizen) was recruited to take part. Recruitment took place through advertisement on messaging and social media platforms, as well as through the University of Bath's Research Participation Scheme. Data from seven participants were subsequently excluded for failure to select the correct response on an attention check question, resulting in a final sample of 155. There was no difference between experimental groups in terms of age, t(151) = 0.71, p = .225, d = 0.20, or gender distribution, Fisher's exact test, p = .572 (Table 6).

Table. Mock juror demographics for Study 3. (See the original article for the full table.)

Ethical approval was granted by the University of Bath's Psychology Research Ethics Committee (reference number: 3928‐3947).

Materials and procedure

Mock juror participants completed the study online via QuestionPro. After completing the screening and demographic questions, they were randomly assigned to either the information condition or the no information condition.

Information about stammering

Participants in the information condition were informed that the witness video they were about to view was of a PWS, and they were provided information about stammering (see Study 2 and Supplementary Material F for details).

Eyewitness testimony

Next, mock juror participants read the written transcript of the free recall component of a male witness' testimony before viewing a video of him being questioned by an advocate. To create the eyewitness testimony video, a 73‐year‐old white male volunteer with a stammer was recruited from STAMMA. He viewed a video of a staged theft lasting 1 min 36 s before, 2 h later, completing a written account of what he witnessed in the video. His written account was then shared with an experienced legal advocate who prepared a line of questions with which to test his account. Four days after watching the video, the witness was questioned by the advocate via video call during which he was first asked to provide a verbal free recall of the events before being cross‐examined on this account. The free recall component was transcribed, and the cross‐examination was presented as a 5‐min video.

Witness credibility scale

Mock jurors then rated the witness' credibility using the Witness Credibility Scale (Brodsky et al., 2010). See Study 2 and Supplementary Material D for details.

Knowledge of stammering questionnaire

Finally, jurors completed a knowledge of stammering measure, adapted from Campbell and Barger's (2010) ‘Knowledge of Autism Questionnaire’. Statements such as, ‘Stammering can involve someone repeating sounds or words’, were developed in collaboration with STAMMA (see Supplementary Material G). Participants responded ‘True’, ‘False’ or ‘Unsure’ (an option added to avoid forced responses). Cronbach's alpha indicated acceptable reliability (α = .65).

Study 3: Results

Knowledge of stammering

An independent‐samples t‐test examined whether the provision of information about stammering improved participants' knowledge of the condition. Participants who received information about stammering scored significantly higher on the knowledge of stammering questionnaire (M = 9.82, SD = 2.44) than those who did not receive information about stammering (M = 8.83, SD = 2.46), t(149) = 2.21, p = .014, d = 0.36.

Provision of stammering information on credibility ratings

To examine if perceptions of credibility differed between the stammering information condition and the no information condition, a one‐way MANOVA was run with the four credibility variables: confidence, likeability, trustworthiness and capability. There was a significant multivariate effect of information provision on perceived credibility, F(4, 150) = 3.02, p = .020, h 2 p = 0.07. Univariate tests indicated that juror participants who were informed that the witness had a stammer and were given information about stammering rated him as more likeable, F(1, 153) = 10.04, p = .002, h 2 p = 0.06, trustworthy, F(1, 153) = 10.69, p = .001, h 2 p = 0.07 and capable, F(1, 153) = 4.44, p = .037, h 2 p = 0.03 than participants who were not informed about his stammer. There was no significant difference between conditions in terms of perceptions of his confidence, F(1, 153) = 4.45, p = .063, h 2 p = 0.02 (Table 7).

Table. Mean credibility ratings by information condition (standard deviations are in parentheses). (See the original article for the full table.)

Study 3: Discussion

As predicted, jurors who were informed that the witness had a stammer and were provided with information about stammering rated him as more credible overall, and more likeable, trustworthy and capable specifically. However, perceptions of the witness's confidence were unaffected by information provision.