To assess participants’ evaluative racial beliefs and beliefs about race-related oppression and privilege, participants perform an Evaluative Race Implicit Association Test (E-IAT; Greenwald, Nosek, & Banaji, 2003) and a Stereotype Implicit Association Test (S-IAT; Uhlmann, Brescoll, & Paluck, 2006). The order of the IATs is counterbalanced within both conditions (perspective-taking and objective focus). Detailed instructions state participants have to categorise words and pictures as quickly and accurately as possible into categories with the keyboard buttons ‘s’ (left) and ‘k’ (right). The category labels are visible in the upper corners of the screen during the tasks. Furthermore, before each block, participants are provided with detailed instructions for each specific block. The E-IAT used for this study follows a standard procedure for an IAT with seven blocks. The categories in which the word stimuli need to be categorized are positive and negative for the E-IAT, and thinking (denken) and doing (doen) for the S-IAT. The word stimuli of the E-IAT consist of ten positive and ten negative words that were randomly chosen. The word stimuli of the S-IAT consist of ten words related to doing and ten words related to thinking. The categories in which the picture stimuli need to be categorized are African American and European American. The picture stimuli consist of photographs with faces of ten African American and ten European American males and females, and were randomly chosen from the Chicago Face Database (Ma, Correll, & Wittenbrink, 2015).
Each participant completes seven blocks of randomized trials, of which blocks 1, 2, 3, 5, and 6 are practice blocks and blocks 4 and 7 are target blocks. In the first block (20 trials), participants categorise only words, in which positive is left and negative is right. In the second block (20 trials), participants categorise only pictures of European American (left) and African American (right) males and females. In the third (20 trials) and fourth blocks (40 trials), the first two block are combined and participants categorise both positive (left) and negative (right) words, and pictures of European American (left) and African American (right) males and females. In the fifth block (20 trials), participants categorise only pictures of European American and African American males and females, however, the categories corresponding with the response keys are switched (European American is now right/‘s’ and African American is now left/‘k’). The sixth (20 trials) and seventh (40 trials) blocks again combine the categorisation of both words and pictures, where participants categorise positive (left) and negative (right) words, and pictures of African American (left) and European American (right) males and females. When participants do not respond accurately, an error message is shown, and participants have to give the correct response as quickly as possible to continue to the next trial.
IAT D-scores. For both IATs, D-scores are calculated according the scoring algorithm that was created by Greenwald, Nosek, and Banaji (2003). The steps for calculating D-scores can be found in Table 1. For each participant, this results in two D-scores, namely one for the E-IAT and one for the S-IAT.
Table 1. Steps for calculating IAT D-scores using the improved algorithm.
Eliminate trials with latencies > 10,000 ms and eliminate subjects for whom more than 10% of the trials have latencies less than 300 ms, further use all trials from the blocks.
Compute the mean of correct latencies for each block.
Compute one pooled SD for all trials in blocks 3 and 6, and another for blocks 4 and 7.
Replace each error latency with the block mean (computed in Step 3) + 600 ms.
Average the resulting values for each of the four blocks.
Compute the difference between blocks 6 and 3, and between blocks 7 and 4.
Divide each difference by its associated pooled-trials SD from Step 4.
Average the two quotients from Step 8.
Note. ms = milliseconds, SD = standard deviation.
The experiment ends with a questionnaire, assessing participants’ identification with the character (Cohen, 2001), narrative transportation into the story (Appel, Gnambs, Richter, & Green, 2015), film and character liking, demographic information (age, gender, first language), and television use (amount of hours watching linear and on demand television per week).
Narrative transportation is assessed through the Short Transportation Scale (Appel, Gnambs, Richter, & Green, 2015), of which the original items can be found in Table 2. For this research, the items were translated into Dutch, of which the translated items can be found in Table 3. A mean score of the response scores on the items is calculated, in which higher mean scores reflect a higher degree of narrative transportation. Narrative transportation serves as a control variable in the analyses.
Table 2. Items of the Transportation Scale – Short Form (TS-SF).
I could picture myself in the scene of the events portrayed in the movie.
I was mentally involved in the narrative while watching it.
I wanted to learn how the narrative ended.
The narrative affected me emotionally.
While watching the movie I had a vivid image of Aibileen/Cecil.
Note. Items were presented with seven-point response scales from 1 (not at all) to 7 (very much). Item numbers TS correspond to those provided by Green and Brock (2000, Table 1).
Table 3. Items of the Transportation Scale – Short Form (TS-SF) in Dutch.
Ik kon mezelf inbeelden in de gebeurtenissen uit The Help/The Butler.
Tijdens het kijken van The Help/The Butler was ik in gedachten betrokken bij het verhaal.
Ik wilde weten hoe het verhaal van The Help/The Butler afliep.
Het verhaal van The Help/The Butler raakte me op emotionele wijze.
Tijdens het kijken van The Help/The Butler had ik een levendig beeld van Aibileen/Cecil.
Note. Items were presented with seven-point response scales from 1 (not at all) to 7 (very much).
Character identification is assessed through the identification measure as described by Cohen (2001), of which the original items can be found in Table 4. For this research, the items were translated into Dutch of which the translated items can be found in Table 5. A mean score of the response scores on the items is calculated, in which higher mean scores reflect a higher degree of identification with the media character. Character identification serves as a control variable in the analyses.
Table 4. Items of Identification with media characters.
While viewing The Help/The Butler, I felt as if I was part of the action.
While viewing The Help/The Butler, I forgot myself and was fully absorbed.
I was able to understand the events in The Help/The Butler in a manner similar to that in which Aibileen/Cecil understood them.
I think I have a good understanding of Aibileen/Cecil.
I tend to understand the reasons why Aibileen/Cecil does what he or she does.
While viewing The Help/The Butler I could feel the emotions Aibileen/Cecil portrayed.
During viewing, I felt I could really get inside Aibileen’s/Cecil’s head.
At key moments in The Help/The Butler, I felt I knew exactly what Aibileen/Cecil was going through.
While viewing The Help/The Butler, I wanted Aibileen/Cecil to succeed in achieving his or her goals.
When Aibileen/Cecil succeeded I felt joy, but when he or she failed, I was sad.
Table 5. Items of Identification with media characters in Dutch.
Tijdens het kijken van The Help/The Butler voelde het alsof ik onderdeel was van het verhaal.
Tijdens het kijken van The Help/The Butler vergat ik mezelf en ging ik compleet op in het verhaal.
Ik was in staat om de gebeurtenissen in The Help/The Butler op dezelfde manier te begrijpen zoals Aibileen/Cecil dat deed.
Ik denk dat ik Aibileen/Cecil goed begrijp.
Ik begrijp de redenen van Aibileen/Cecil om zich te gedragen zoals zij/hij deed.
Tijdens het kijken van The Help/The Butler kon ik de emoties van Aibileen/Cecil voelen.
Tijdens het kijken van The Help/The Butler had ik het gevoel dat ik in Aibileen’s/Cecil’s hoofd zat.
Op belangrijke momenten in The Help/The Butler wist ik precies wat Aibileen/Cecil doormaakte.
Tijdens het kijken van The Help/The Butler wilde ik dat Aibileen/Cecil slaagde in het bereiken van haar/zijn doelen.
Toen Aibileen/Cecil slaagde in het bereiken van haar/zijn doelen was ik blij, maar toen zij/hij faalde, voelde ik me verdrietig.
Film and character liking are assessed by asking to what extent participants liked watching the film, and liked the main character. Film and character liking serve as control variables in the analyses.
In hoeverre vond je het kijken naar de film The Help/The Butler leuk?
In hoeverre vond je het hoofdpersonage Aibileen/Cecil leuk?
Also, a manipulation check is included by asking to what extent participants tried to take the perspective of the main character by imagining his/her thoughts, feelings and experience and to what extent participants tried to be objective and emotionally detached.
In hoeverre probeerde je je voor te stellen wat Aibileen/Cecil mogelijk dacht, voelde en beleefde?
In hoeverre probeerde je zo objectief en emotioneel afzijdig te blijven?
Last, the demographic variables age, gender and first language are assessed. Furthermore, television use is assessed by asking how many hours a week participants watch linear and on demand television and some additional questions regarding the experiment are asked.
Wat is je leeftijd?
Wat is je geslacht?
Wat is je moedertaal?
Gemiddeld gezien, hoeveel uur per week kijk je naar televisie?
Gemiddeld gezien, hoeveel uur per week van die kijktijd is on demand (bijv. via een digitale aanbieder zoals Netflix)?
Heb je tijdens het onderzoek met (een van) de andere deelnemers gepraat over de film?
Appel, M., Gnambs, T., Richter, T., & Green, M. C. (2015). The Transportation Scale-Short Form (TS-