Abstract
Limited research examines how people of color view Jesus’s race and the influence of culture on such a view. We investigated the relationship between views of Jesus’s race and implicit bias when viewing images of Jesus (e.g., positive words associated with White Jesus, negative words associated with non-White Jesus) among Asian American students (N = 103) from a Christian university. We predicted that (a) a small percentage of participants would view Jesus as White; (b) participants would demonstrate implicit bias; and (c) beliefs about Jesus’s race, racial colorblindness, ethnic identity, and internalized model minority stereotype would predict implicit bias. Participants were more likely to view Jesus as non-White but favor images of White Jesus over images of non-White Jesus, especially participants who endorsed White Jesus. Racial colorblindness was associated with beliefs about Jesus’s race and implicit bias. These findings highlight how inaccurately portraying prominent figures’ race as White perpetuates White standards such that race is obliviated and biases are internalized.
It is historically inaccurate to depict Jesus as a White person. Despite this, Jesus has been and continues to be portrayed as a White male in artistic renderings (e.g., see Warner Sallman’s 1940 painting, Head of Christ; McFarlan, 2020). Given this trend, there is a small but emerging body of studies reporting what individuals might believe about the racial and ethnic identity of Jesus. For instance, Howard et al. (2021) found that 47.20% of their U.S. college sample believed that Jesus was White. Similarly, Jackson et al. (2018) reported that one of the prominent ways American Christian participants viewed God was as a White person. This research trend is consistent with a recent YouGovAmerica survey of the American public that found that 60% of participants were accepting of Jesus being depicted as White (Smith, 2020). Unsurprisingly, this belief that Jesus is White is correlated with outcomes such as colorblind racial ideology and anti-Black attitudes (Howard et al., 2021).
Our empirical study addresses questions that remain unanswered in the current literature. One, prior studies on Jesus’s race featured racially diverse samples with the majority of participants identifying as White (Howard et al., 2021; Jackson et al., 2018). However, an alternative approach might be to intentionally recruit a non-White sample that will lend itself to important empirical queries, such as the role of culture-specific processes in relation to views about Jesus’s race. Second, as a correlate of one’s beliefs about Jesus’s race, Howard et al. (2021) utilized an Implicit Association Test (IAT; Greenwald et al., 1998) which involved the pairing of various words with Black and White male faces; Howard et al. (2021) reported that there was no statistically significant difference in the IAT scores between those who believed Jesus was White and those who believed that Jesus was non-White. However, it would be compelling to conduct an IAT that displays faces of Jesus Himself from various racial backgrounds, instead of generic White and non-White male faces. In sum, the present study adds to the current empirical literature by examining beliefs about Jesus’s racial identity; implicit associations of words (favorable vs. unfavorable) with images of Jesus portrayed as White or non-White; and how these beliefs might be correlated with racial color-blind ideology (Neville et al., 2000), ethnic identity (Phinney & Ong, 2007), and internalized model minority stereotype (Yoo et al., 2010) among Asian American students. Such an empirical effort has the potential to both broaden (i.e., with an Asian American sample) and deepen (i.e., cultural correlates of implicit bias) the understanding of how culture can shape views of Jesus’s race. In turn, this can inform preventive and intervention efforts to portray Jesus in church and church-affiliated settings in a manner that is more historically accurate and not reliant on a “White standard” of Christianity.
Christianity and Asian Americans
In the United States, the largest religious group among Asian Americans is Christianity (42%; Pew Research Center, 2012). Moreover, the numbers also indicate that Asian Americans place a strong emphasis on religious practices. For instance, compared to the general U.S. population, Asian Americans report a higher level of church attendance and are more likely to say that living a religious life is important to them (Pew Research Center, 2012). Given these trends, it is not surprising that Asian Americans are represented on Christian college campuses. For example, in a recent report based on IPEDS (the Integrated Postsecondary Education Data System), the Council for Christian Colleges and Universities (CCCU, 2021) reported that 4.0% of its students identified as “Asian/Native Hawaiian or Other Pacific Islander.” In sum, Asian Americans make up a significant percentage of American Christians, and the Christian higher education setting is one setting in which Asian Americans are found.
Beyond the numbers, Christianity has important functions and consequences in Asian American lives. On the one hand, religion can promote psychological health. For example, both God support and religious community support have been shown to increase life satisfaction and increase psychological well-being, suggesting that this unique type of support may be impactful beyond general social support (P. Y. Kim, 2017; G. H. H. Yi & Bjorck, 2014). Support from those who share similar experiences (i.e., discrimination and microaggressions) and identities (i.e., Christian and Asian) may explain why religious support has particularly salient effects (P. Y. Kim, 2017). Beyond these effects, religious support has also been found to decrease emotional distress. For example, religious attendance can serve as a protective factor for various mental health distress, by reducing the likelihood of major depression among Asian Americans (Ai et al., 2013).
Conversely, scholars have also written about how religion, including Christianity, might hurt. White Christian nationalist views run deep throughout the United States shaping how current events are viewed, and they promote racist and xenophobic views against Asian Americans, including during the COVID-19 pandemic (Perry et al., 2020). Similarly, Asian American students on Christian campuses experience marginalization and discrimination in different ways due to the religious and racialized campus climate that contrasts the campus climate of those on secular campuses. For example, Christian campuses may lack diversity of individuals and perspectives in the classroom, deny racism, and perpetuate the “missionary mentality” (C. L. Kim et al., 2010). Thus, Asian Americans’ racialized experiences in Christian higher education settings may look different compared to the experiences of their counterparts at secular institutions. Nevertheless, research on Christianity and its influence on perspectives regarding psychosocial issues, however, are sparse. To address this gap, we decided to examine how Christianity, and the White standards it promotes, influence individuals’ view of Jesus’s race. These beliefs will shed light on how Asian Americans’ social attitudes, such as implicit bias and colorblind racial ideology, are shaped by religious beliefs.
Beliefs About Jesus’s Race
One preliminary empirical question that could be asked is regarding what Asian Americans believe about the racial identity of Jesus. Given the centrality of Jesus in the Christian faith, it is important that Jesus’s followers have a factually accurate understanding of Jesus, including His racial background. In particular, seeing Jesus as White can be conceptualized as an internalization of a “White standard”—a standard that upholds White Christianity as an idealized expression of Christianity (Blay, 2011; Howard et al., 2021). Given this theorizing, it is not surprising that the belief that Jesus was White has been associated with more prejudice against Blacks, agreement with social hierarchy based on groups and colorblind racial beliefs (Howard et al., 2021).
In the present study, we were especially interested in the percentage of Asian American participants that would endorse Jesus’s race as White. Although Howard et al. (2021) found that 47.20% of their majority White college sample reported that Jesus was White, we anticipated that a lower percentage of our Asian American sample will endorse Jesus as a White person, based on a couple of reasons. One, prior public research suggests that non-White individuals (e.g., non-White individuals in Britain) tend to be less accepting of White depictions of Jesus compared to their White counterparts (Smith, 2020). Thus, our sample of non-White (i.e., Asian American) college students are more likely to view Jesus as non-White. Second, given that our sample consisted of students enrolled in a Christian university, they might have a more accurate understanding of Jesus’s race compared to someone not as familiar with Christianity. Therefore, we predicted that a smaller percentage of our sample will report that Jesus was White compared to Howard et al. (Hypothesis 1).
Implicit Association Test
In addition to explicitly asking participants about their beliefs regarding Jesus’s race, researchers can also tap into implicit associations made between Jesus’s race and descriptive words. The examination of implicit attitudes is rooted in the assumption that some psychological processes (e.g., attitudes) happen without conscious awareness, including implicit cognition that might contradict explicit beliefs within the same individual (Greenwald & Krieger, 2006). Translated to our study, this might mean that even for those who are consciously aware of the inaccuracy of portraying Jesus’s race as White, unconsciously, the same individuals might still view such depiction of Jesus in a more favorable light than a non-White rendering of Jesus. The IAT (Greenwald et al., 1998) is considered a reliable and valid measurement of one’s automatic mental associations. It has been used extensively in studies with race-related themes such as how genetic counselors’ pro-White implicit biases predict negative communication patterns with non-White patients (Schaa et al., 2015) and how multicultural counseling education predicts reduced implicit racial prejudice among counseling graduate students (Castillo et al., 2007). The IAT is also prevalent in the psychology of religion literature and has been used to examine associations between Christian and humanist beliefs among Christians (Ventis et al., 2010) and associations between America and Christianity (Butz & Carvalho, 2015). More central to the current study, the IAT has also been used to examine the intersection of religious and race-related themes. For example, Howard and Sommers (2019) found that priming White participants with images of White Jesus led to more implicit anti-Black attitudes. Likewise, Black participants who were primed with images of White Jesus were also more prone to implicit anti-Black attitudes (Howard & Sommers, 2017). Based on these empirical inquiries and their findings, it seems reasonable to conclude that IATs can capture race-related biases in relation to religious themes.
More recently, Howard et al. (2021) utilized the Race-IAT (R-IAT; Greenwald et al., 1998) in their study of beliefs about Jesus’s race, which involves the pairing of faces (Black and White) with positive or negative words; they reported null findings involving the R-IAT (e.g., no difference in how quickly White and Black faces were paired with descriptive words, null correlations with variables such as racism). However, one might rightly wonder if an IAT that utilizes Jesus’s pictures, instead of pictures of unknown males, might yield different findings. This need for an IAT using images of Jesus seems especially compelling in light of previous studies by Howard and Sommers (2017, 2019) that found a differential effect of priming based on images of White Jesus versus images of a White male (e.g., Black participants demonstrated anti-Black and pro-White bias in their IAT scores only if they were primed with images of White Jesus but not with pictures of a White male; Howard & Sommers, 2017). Generalizing from this priming finding, we reasoned that using images of Jesus in our IAT, instead of the traditional R-IAT tests of generic male images, would be a promising research endeavor. Consistent with prior IATs that reported implicit favorable views of White faces and unfavorable views of non-White ones (e.g., Greenwald et al., 1998; Maina et al., 2018; McConnell & Leibold, 2001), we predicted that our IAT will show that, on average, Asian American participants will be quicker to associate White Jesus with more positive words and non-White Jesus with negative ones (Hypothesis 2). Moreover, the belief that Jesus is White will be correlated with the IAT scores (Hypothesis 3).
In addition to explicit beliefs about Jesus’s race as a correlate of IAT scores, we also turned to cultural constructs that might predict implicit bias scores: racial colorblindness, ethnic identity, and internalized model minority stereotype. Below, we briefly define these constructs and articulate how they might be associated with implicit bias scores.
Cultural Constructs in Relation to IAT Scores
Racial colorblindness
Racial colorblindness is an ideology claiming that race does not matter in our society, and it consists of two components: (a) color-evasion, or emphasizing the sameness of racial experiences, and (b) power evasion, or emphasizing equal access to opportunities to succeed and the lack of structural racism barriers (Neville et al., 2000). Although perhaps intended to reflect fairmindedness and respect for the individual regardless of cultural backgrounds, these attitudes perpetuate a distorted view of non-Whites and the way race functions in everyday life (Neville et al., 2000). Moreover, recent conceptualizations of racial microaggressions have identified statements communicating racial colorblindness as microaggressive in nature because they invalidate the experiences of people of color (Sue et al., 2007). It is not surprising, then, that racial colorblindness can inflict significant harm psychologically and interpersonally (Neville et al., 2013) with wide-ranging deleterious outcomes such as gender intolerance (Neville et al., 2000), lower multicultural counseling awareness and knowledge (Neville et al., 2006), reduced perception of microaggressions that have occurred (Offermann et al., 2014), and more disagreement with institutional decisions to abandon an offensive campus mascot (Neville et al., 2011).
More central to the current study, racial colorblindness, especially the color-evasion dimension, might be thought to decrease racial prejudice because it ignores racial differences; but the literature suggests that colorblindness predicts racially insensitive behavior, including racial prejudice (Apfelbaum et al., 2008; Holoien & Shelton, 2012; McConnell & Leibold, 2001; Norton et al., 2006). Racial prejudice is an evaluative attitude toward a particular social group based on their race, and it commonly takes the form of aversive prejudice where one unknowingly holds negative attitudes toward people based on their ethnic group membership, therefore, reflecting implicit bias (Dovidio & Gaertner, 2004). Given this argument that ignoring racial differences can actually cause racial insensitivity, we reasoned that racial colorblindness would be associated with other forms and measures of implicit bias, like IAT scores. Increased racial colorblindness was related to an increased likelihood of Jesus being viewed as White (Howard et al., 2021). Howard et al. (2021) speculated that those who acknowledge and understand the way race operates in society and everyday interactions (i.e., low on racial colorblindness) were more likely to recognize Jesus as non-White because of their experience recognizing the influence of race. Therefore, at least in one study (Howard et al., 2021), racial colorblindness has been linked to an explicit bias of believing that Jesus was White. However, no prior studies have examined the association between racial colorblindness and implicit biases regarding Jesus. We reasoned that those who are racially colorblind will more quickly link positive words with images of White Jesus and negative words with images of non-White Jesus, based on the argument that those who do not regularly think about race will likely default to the stereotypical and normalized presentation of Jesus and view that presentation as inherently positive. In sum, we predicted that racial colorblindness will be related to an easier association of White Jesus with positive words and non-White Jesus with negative words (Hypothesis 4a).
Strength of ethnic identity
Ethnic identity is the multifaceted sense of belonging in an ethnic group membership. Achieving a high ethnic identity requires people to question their ethnic group, its importance to their self-concept, how they perceive their membership, and how it is viewed by others (Cheryan & Tsai, 2007). Two factors that significantly influence the structure of ethnic identity are commitment and exploration. Commitment is a strong attachment to an ethnic group, while exploration involves an ongoing process of learning about one’s ethnicity through relevant information and experiences (Phinney & Ong, 2007). For Asian Americans, the strength of ethnic identity can be described by one’s emotional attachment to their group membership, connection with their country of origin, and self-perception (Lien et al., 2003). For example, forming a strong ethnic identity is associated with daily happiness and less daily anxiety among Chinese American adolescents (Kiang et al., 2006). Similarly, Chae and Foley (2010) found that high ethnic identity is related to psychological well-being among Chinese American, Korean American, and Japanese American participants. Although no prior studies have explored the association between ethnic identity and implicit beliefs about Jesus’s race in an Asian American sample, we reasoned that a stronger ethnic identity would predict less implicit bias. At the strongest level of ethnic identity achievement, “people become interested in learning about a variety of cultures and come to value the diversity of ethnic groups in their society” (Linnehan et al., 2006, p. 422). In other words, someone with a stronger ethnic identity might not only favor ingroup members but also develop a positive, open posture toward outgroup ones (Phinney, 1989). Extending this theorizing, someone with a healthy sense of ethnic identity is more likely to value diversity and inclusion, and therefore they might be more sensitive toward societal and religious elements that go against diversity and inclusion (e.g., use of White images in Christian contexts). As such, we predicted that a stronger ethnic identity will be associated with less favorable view of White Jesus, and more positive beliefs about non-White Jesus. Therefore, we hypothesized that a stronger ethnic identity will be associated with less implicit bias (Hypothesis 4b).
Internalized model minority stereotype
The model minority stereotype refers to the generalization that Asian Americans have achieved significant academic and vocational success in the United States and that they experience fewer barriers to their success compared to their non-Asian racial minority counterparts (Yoo et al., 2010). In addition, the model minority stereotype includes a comparison component in which Asian Americans are held as the standard that other groups of color should emulate (Yoo et al., 2010). Internalized model minority stereotype, then, is when Asian Americans endorse this belief about themselves (Yoo et al., 2010). Previous studies have found that internalized model minority stereotype is associated with adverse outcomes such as unfavorable help-seeking attitudes (P. Y. Kim & Lee, 2014) and poorer mental health (Atkin et al., 2018). Beyond mental health and help-seeking domains, internalized model minority stereotypes can also shape how one views social issues. In a recent study of internalized model minority stereotypes, Matriano et al. (2021) reported an inverse correlation between the belief that Asian Americans are successful due to their high education and intelligence levels, and support for Black liberation. Likewise, J. Yi and Todd (2021) reported that the internalized model minority stereotype was associated with more anti-Black sentiments and disagreement with affirmative action, suggesting that this endorsement of the stereotype might have consequences on views related to race. Although internalized model minority stereotype is internally directed (i.e., focused on one’s own Asian American identity) and racial colorblindness is externally directed (i.e., focused on how one views others), internalized model minority stereotype operates similarly to racial colorblindness. For example, internalized model minority stereotype is disguised as positive sentiments but can cause adverse outcomes for those who internalize it; we are curious if implicit bias can also be associated with internalized model minority stereotype. We are not aware of studies that have examined this association among Asian Americans. But drawing from the current literature reviewed earlier, it seems logical that a strong endorsement of internalized model minority stereotypes can influence implicit bias. Specifically, internalized beliefs about the high-achieving nature of Asian Americans may impact views related to Jesus’s race. As such, we predicted that those who have internalized the model minority stereotype will associate White Jesus with more positive words, and non-White Jesus with more negative ones (Hypothesis 4c).
Study hypotheses
Method
Participant characteristics
Participants were 103 self-identified Asian American undergraduate students (Mage = 20.26, SD = 1.82; 65 female, 33 male, 5 non-binary/other) from a Christian university located in the Pacific Northwest region of the United States. Participants had the option of reporting more than one Asian ethnicity; the following ethnicities were reported: Chinese from mainland China (n = 26), Filipina/o (n = 24), (Japanese n = 17), Korean (n = 15), Hong Kongese/Hong Konger/Chinese from Hong Kong (n = 11), Vietnamese (n = 11), Taiwanese/Chinese from Taiwan (n = 7), Indian (n = 5), Cambodian (n = 3), Thai (n = 3), Indonesian (n = 1), Laotian (n = 1), and others (n = 6). Eighty-six participants reported being born in the United States, and participants reported living in the United States for an average of 18.88 years (SD = 4.24). Participants self-identified as Christian (Catholic, non-denominational, Orthodox, Pentecostal, or Protestant; n = 78), “no religion” (n = 13), Buddhist (n = 4), agnostic (n = 3), Sikh (n = 2), Hindu (n = 1), and Muslim (n = 1); 1 participant wrote in “Believe in God, but not part of any religion.”
Recruitment and procedures
Emails announcing the study were sent out to Asian American students enrolled at the current institution through the Registrar’s office. Participants who wished to participate could click on the study link, which then took them to the online survey in the Qualtrics survey platform consisting of demographic questions, study measures, and the IATs of the present study. For participation in the current study, participants were entered into a drawing for a single prize of $300 worth of Amazon.com gift card. All participants were treated according to the ethical standards of the American Psychological Association.
Although 126 individuals at least began the Qualtrics survey, the final dataset included 103 participants after the following procedures. First, we removed 16 participants who did not provide consent for participation. Second, we removed two participants who did not identify as Asian American. Third, we removed one participant who was not a student enrolled at the institution where data collection took place. Finally, we removed four participants with one or more measures missing.
Measures
Beliefs about Jesus’s race
Howard et al. (2021) used a multiple-choice question to ask participants about Jesus’s race, and this question was adapted from Jensen (2009)’s gendered conception of God coding. We utilized this question in our study. The question read, “Based on your personal understanding of Jesus, do you think of Jesus as _______,” and the choices included White, Black, something else, and do not know.
Implicit Association Test
We utilized IATs (Greenwald et al., 1998) to assess the mental connections made about Jesus based on His physical appearance. Specifically, the IAT assessed the speed of association between targets (Target A: White Jesus; Target B: non-White Jesus) and categories (negative and positive words). Non-White Jesus images included Compassionate Jesus by John Giuliani, Light of Life by Melissa T. Kamba, and Black Jesus by John Campbell. White Jesus images included Graven Upon My Hands by Sandy Freckleton Gagon, Head of Christ by Warner Sallman, and Grace and Truth by Simon Dewey. The positive (e.g., caring, supportive, just) and negative (e.g., cruel, hating, harsh) words were drawn from the Positive/Negative God Go/No-go Association Task (Pirutinsky et al., 2017). We created our IATs using iatgen (Carpenter et al., 2019), which is an online tool researchers can utilize to create a survey-based IAT that is administered in Qualtrics. Our IATs included seven blocks (three practices, four trials). Block 1 and Block 2 included practices of sorting the target images and category words, respectively. Participants then completed two combined blocks (Block 3 + Block 4; Block 6 + Block 7) that required the pairing of the targets with categories (“compatible” pairing: White Jesus with positive words, non-White Jesus with negative words; “incompatible” pairing: non-White Jesus with positive words, White Jesus with negative words). The counterbalancing procedure resulted in four different combinations that participants began Block 3 with: compatible pairing with White Jesus appearing on the right; incompatible pairing with White Jesus appearing on the left; compatible pairing with White Jesus appearing on the left; and incompatible paring with White Jesus appearing on the left. Finally, between the Block 3 + Block 4 and Block 6 + Block 7 combinations, there was a practice block for participants to practice the category in reverse, intended to eliminate the practice effect from the initial combined block. For additional details on iatgen, see Carpenter et al. (2019). The estimated internal consistency, based on split-half reliability with Spearman-Brown correction (see Carpenter et al., 2019; De Houwer & De Bruycker, 2007), was .88. A positive difference score (D) on the IAT measure indicated that the compatibility associations were made more quickly, whereas a negative D meant that the incompatible associations were made more quickly. We used the iatgen online tool (Carpenter et al., 2019) to analyze the IAT data. Out of the 103 participants, four participants were dropped for excessive speed (i.e., more than 10% of trials were faster than 300 ms). Trials that took longer than 10,000 ms were also dropped, with the proportion of trials dropped due to this reason being .0009. The error rate (i.e., the proportion of trials in which errors were made) was .086.
Colorblind racial attitudes
The Colorblind and Racial Attitudes Scale-14 (CoBRAS-14; Neville et al., 2007) is a shortened version of the Colorblind and Racial Attitudes Scale (Neville et al., 2000), and it includes 14 items that assess the degree of endorsement of racially colorblind attitudes. Each item is rated by the participants on a 6-point Likert-type scale (1 = strongly disagree to 6 = strongly agree). Example items are “Racism may have been a problem in the past, but it is not an important problem today” and “Social policies, such as affirmative action, discriminate unfairly against White people” (Neville et al., 2017). The CoBRAS-14 demonstrated good reliability in our study, α = .82. A higher score on this measure indicated a stronger agreement with racial colorblind ideology, and we used the mean score for analysis.
Ethnic identity
The Multigroup Ethnic Identity Measure-Revised (MEIM-R; Phinney & Ong, 2007) is a revised version of the Multiethnic Identity Measure (Phinney, 1992), and it assesses the degree to which one explores and has committed to one or more ethnic identities. MEIM-R includes six items, and each item is rated on a 5-point Likert-type scale (1 = strongly disagree to 5 = strongly agree). A higher score indicates a stronger sense of ethnic identity. The measure has been used with Asian American samples (e.g., Tuazon et al., 2019; Zhang, 2015). A sample item found in the measure is “I have spent time trying to find out more about my ethnic group, such as its history, traditions, and customs” (Phinney & Ong, 2007). The measure demonstrated good internal consistency in our study, α = .89. A higher score on this measure indicated a stronger ethnic identity, and we used the mean score for analysis.
Internalized model minority stereotype
The Internalization of Model Minority Myth Measure (IM-4; Yoo et al., 2010) is a measure that assesses the degree to which individuals perceive Asian Americans as more successful. This measure includes two subscales: (1) the Model Minority Myth of Achievement Orientation (MM–Achievement) which focuses on the endorsement of Asian American values of achievement and hard work and (2) the Model Minority Myth of Unrestricted Mobility (MM–Mobility) which reflects a belief that there is a lack of barriers to success for Asian Americans. In the current study, we used the MM-Achievement subscale only. The MM-Achievement includes five items that rate participants on a 7-point Likert-type scale (1 = strongly disagree to 7 = strongly agree). A sample item is “Asian Americans have stronger work ethics” (Yoo et al., 2010). This measure has been validated with an Asian American sample (Yoo et al., 2010). In the current study, MM-Achievement demonstrated high internal consistency with a score of α = .91.
Results
Preliminary analysis
Table 1 displays the bivariate correlations, means, and standard deviations of the study variables. Implicit bias scores were significantly and positively correlated with colorblind racial attitudes, r = .23, p = .02. Colorblind racial attitudes were significantly and inversely correlated with ethnic identity, r = −.22, p = .03.
Bivariate Correlations, Means, and SDs.
p < .05.
Explicit beliefs about Jesus’s race
Hypothesis 1 predicted that the percentage of participants reporting Jesus’s race as White will be lower than the 47.20% reported by Howard et al. (2021). In response to the question that inquired about their belief regarding Jesus’s race, 19 participants reported believing that Jesus was White, three participants reported that Jesus was Black, 66 participants reported that Jesus’s race was something else, and 15 participants reported being unsure of Jesus’s race. As such, only 18.45% of participants reported Jesus’s race as White, supporting Hypothesis 1.
IATs and correlates
Hypothesis 2 stated that it will be easier for participants to link White Jesus with positive words and non-White Jesus with negative words. As noted earlier, a positive D score indicated a tendency to associate White Jesus with positive words, and non-White Jesus with negative words. Figure 1 displays the density plot of the D-scores (M = 0.24, SD = 0.41). Results revealed that D scores significantly differed from zero, t (98) = 5.82, p < .001, Cohen’s d = .59. Thus, participants were quicker to associate White Jesus with positive words and non-White Jesus with negative ones, and this association was statistically significant. Hypothesis 2 was supported.

Density plot of implicit bias (D-scores).
Next, we conducted a one-way ANOVA to see if there was a significant effect of views about Jesus’s race on the D scores. This was to test Hypothesis 3, which predicted that participants’ beliefs about Jesus’s race will be associated with their implicit associations; those who view Jesus as White will more quickly associate positive words with White Jesus and negative words with non-White Jesus. The Shapiro–Wilk test of normality suggested that assumption of normality was not violated, W = .99, p = .742. Likewise, the homogeneity of variance assumption (Levene’s test) held, F (3, 95) = .03, p = .995. ANOVA results revealed that there was a significant overall effect of views about Jesus’s race on IAT scores, F (3, 95) = 4.96, p = .003. Post hoc analysis using the Tukey method indicated that the statistically significant contrast was between participants who reported Jesus as White (M = 0.54, SD = 0.36) and those who reported “something else” (M = 0.16, SD = 0.39). Figure 2 summarizes the D-scores across the four responses. In sum, Hypothesis 3 was supported.

Implicit bias (D-scores) by belief in Jesus’s race.
Hypothesis 4 examined whether our three cultural correlates (racial colorblindness, ethnic identity, and internalized model minority stereotype) might be predictive of IAT scores. We used hierarchical regression analyses to test these research questions. Specifically, in Step 1, we entered racially colorblind attitudes and ethnic identity as the first set of predictors, given that these are established cultural constructs that can generalize across racial groups. In Step 2, we entered internalized model minority stereotype as a construct that is specific to the Asian context. Table 2 displays the regression results. In Step 1, racial colorblindness was a statistically significant predictor of the D scores, β = .20, t = 2.03, p = .045. Ethnic identity was not a significant predictor, β = −.14, t = −1.39, p = .17. Step 1 explained a significant proportion of the variance, R2 = .07, F = 3.80, p = .03. In Step 2, the added predictor of internalized model minority stereotype (β = .10, t = 1.01, p = .32) was not a significant predictor. Racial colorblindness remained a significant predictor (β = .20, t = 2.01, p = .047), and ethnic identity remained a nonsignificant predictor (β = −.14, t = −1.38, p = .17). Step 2 explained a significant proportion of the variance, R2 = .08, F = 2.87, p = .04. In sum, the more participants endorsed a racially colorblind approach, the more likely they were to associate White Jesus with positive words and non-White Jesus with negative descriptors. Thus, Hypothesis 4 was partially supported as the racially colorblind approach emerged as a significant predictor but not racial identity and internalized model minority stereotype.
Implicit Bias (D-Scores) Regressed on Racially Colorblind Attitudes, Ethnic Identity, and Internalized Model Minority Stereotype.
Note. Statistically significant predictors are in bold.
Discussion
The present study was an empirical investigation of (a) beliefs about Jesus’s race, (b) implicit bias that involved pairing images of White and non-White Jesus with positive and negative descriptors, and (c) relations between implicit bias and beliefs about Jesus’s race, internalized model minority stereotype, strength of ethnic identity, and racial colorblindness in a sample of Asian American college students from a Christian university. 18.45% of our participants reported the belief that Jesus was White which, as we predicted, was a smaller percentage than reported in the study by Howard et al. (2021) study (47.20%), which was a much lower number compared to what was reported in a general college sample (47.20%). At the same time, participants overall were more likely to favor White Jesus images over non-White Jesus images, as evidenced by a quicker association of positive words with White Jesus and negative words with non-White Jesus. Furthermore, implicit bias was predicted by both beliefs about Jesus’s race and racial colorblindness. Contrary to our prediction, internalized model minority stereotypes and strength of ethnic identity were not significantly related to implicit bias.
Explanation of findings
Beliefs about Jesus’s race and implicit bias
In contrast to prior findings (e.g., Howard et al., 2021), our participants were much less likely to report that Jesus was White. It is likely that students enrolled in Christian higher education (current sample) have been more frequently exposed to accurate educational materials that depict Jesus as non-White compared to those in nonreligious settings. Another explanation is that our sample’s racial identification as non-White might indicate more awareness of Jesus’s racial identification as a non-White being compared to a majority White sample.
However, our participants were also likely to link favorable words with White Jesus and unfavorable words with non-White Jesus (i.e., implicit bias) on a measure of implicit association. Not surprisingly, this implicit bias was especially strong for those who answered that Jesus was White compared to those who believed Jesus’s race to be “something else.” By using White and non-White images of Jesus in our IAT instead of generic male faces of prior literature (e.g., Howard et al., 2021), we sought to capture biases about Jesus specifically. Our finding is consistent with the perspective that in society and the church, Jesus continues to be portrayed as White and people are generally accepting of such depictions—in fact, more accepting of such depiction compared to Jesus being portrayed as non-White (e.g., Smith, 2020). For those who believe that Jesus was a White male, a portrayal of Jesus that is consistent with their perspective will be viewed more favorably (i.e., White Jesus), and a depiction that is inconsistent with their belief will be viewed more unfavorably (i.e., non-White Jesus). Another possibility is that both the Christian context of Christian higher education and individual perspectives of our majority Christian sample contribute to an internalized view of Jesus as a White person.
Racial colorblindness and implicit bias
The present study found that racial colorblindness significantly predicted implicit bias. This finding is consistent with existing literature demonstrating that a higher level of racial colorblindness is associated with negative attitudes toward people of color (McConnell & Leibold, 2001; Yogeeswaran et al., 2017). In addition, racial colorblindness has been linked to an increased likelihood of believing that Jesus is White (Howard et al., 2021). However, our study is the first to report a significant finding of racial colorblindness in relation to an IAT involving images of Jesus and descriptive words.
We posit four possible explanations for the significant empirical relation between racial colorblindness and implicit bias, each rooted in the argument that the “status quo” racial colorblindness espouses can actually function as an endorsement of Whiteness as the standard. Racial colorblindness, especially the power-evasion dimension, reflects contemporary expressions of racism like prejudice and racist attitudes that tend not to be expressed explicitly (Neville et al., 2013). Because it is not acceptable to outwardly express prejudice and racist attitudes, racially colorblind individuals might avoid intergroup contact in an effort to hide these attitudes (Neville et al., 2013). That is, lack of contact with outgroup members might be a common response of someone who tends to hold a racially colorblind perspective. However, contact with other racial groups is vital, according to the contact hypothesis, to developing positive racial attitudes toward members of different races, and lack of contact can sustain prejudice and negative attitudes (Allport, 1954). For example, individuals with less outgroup contact are more likely to have inaccurate, negative attitudes toward the outgroup because they have not had positive interactions to counteract their socialized attitudes and assumptions about them. It makes sense that unless these attitudes are challenged, over time, negative biases will become internalized so that they turn into deeply held implicit biases. For example, Aberson et al. (2004) found that those with close outgroup friendships showed less implicit bias on IAT measures compared to those without interethnic friendships, suggesting that contact can reduce unconscious biases. By interacting with outgroup members, individuals gain a true understanding of outgroup perspectives and behaviors that can encourage them to reevaluate and change their inaccurate, negative attitudes and assumptions about the outgroup. Without contact, however, individuals’ negative feelings toward outgroup members will not be challenged and will be continually internalized, resulting in the formation of implicit biases. What is intriguing here is the possibility that the lack of contact impacts biases unevenly, such that White Jesus images are viewed more favorably despite the outgroup identity of the White identity for Asian American students, whereas non-White Jesus images (also outgroup members, as all of the images were non-Asian) are viewed more unfavorably. Such an uneven influence of contact on implicit bias can be perhaps understood against the backdrop of the well-theorized moderators of interpersonal contact (see Allport, 1954); as one example, if structural or institutional systems advantaged certain types of contact (e.g., with Whites) over others (e.g., with non-Asian racial minorities), biases toward Whites could be more impacted compared to biases related to non-Asian racial minorities. However, because we did not directly assess contact and its theorized moderators, we posit these thoughts and extensions with ample caution.
Our second explanation draws from the possible overlap between Asian cultural values and racial colorblindness and how this might be associated with favorable views of ingroup members and unfavorable views of outgroup ones. Colorblind racial ideology emphasizes sameness and equality across racial groups, and at the surface level, seems to be a well-intentioned effort to promote diversity and manage racial prejudices (Neville et al., 2013). These similar values of sameness and minimization of differences are also emphasized in some Asian cultures and align with collectivist values of interconnectedness and group harmony (e.g., B. S. K. Kim et al., 2005). For those who adhere to collectivist values, group interests and well-being, maintenance of group morale, and an emphasis on social harmony might be prioritized. If individuals do not understand the limits and potential harms of racial colorblindness, they might be motivated by its appeal of emphasizing sameness to maintain the ingroup cohesion and ingroup interconnectedness that is valued in some collectivist cultures (B. S. K. Kim et al., 2005).
Third—and related to cultural values—collectivist worldviews are associated with favoring ingroup members to achieve a more positive self-image for their group, according to the social identity theory (Mackie & Goethals, 1987). Because these individuals possess an affinity for their ingroup members, it makes sense that this will eventually lead to negative attitudes, and therefore, bias toward outgroup members to protect and preserve the identity and integrity of their group. Consistent with this notion, strong family and cultural ties have been shown to predict ingroup bias (Heine & Lehman, 1997). In sum, those who view their own group more favorably will naturally hold more negative attitudes and biases toward those not in their group—including implicit biases. However, given that we did not directly assess collectivistic values, we provide these explanations with caution.
Finally, our fourth explanation describes how racial colorblindness and the egalitarian views it promotes, prompt individuals to internalize views of their social status and blame minorities for this perceived lower status. Yogeeswaran et al. (2017) found that those with lower social dominance orientation are more likely to hold colorblind attitudes that predict negative attitudes toward minorities (outgroup). Similarly, both racial colorblindness and low social dominance orientation posit egalitarian values such that the sameness of ethnic groups and experiences are emphasized, and the structural factors of social and racial inequalities are ignored (Neville et al., 2013; Pratto et al., 2000). Because of their beliefs about equal opportunity and access, we posit that racially colorblind individuals are more likely to internalize society’s standards about their social standing. Coupled with blame and resentment of outgroups and the tendency to less favorably view minorities, it makes sense that colorblind individuals will hold more negative views toward minority outgroups. Furthermore, since our study employed non-White, non-Asian images of Jesus, it makes sense that racially colorblind individuals would view non-White Jesus more negatively, thus associating Him more quickly with negative words. Moreover, individuals who hold high colorblind attitudes may more easily blame non-White individuals for their position in society. This, in turn, can foster negative feelings toward non-White groups and lead to implicit biases toward them.
Implications
Our significant findings have implications for Christian higher education settings. Our findings indicated that Asian American students viewed images of non-White Jesus more negatively compared to images of White Jesus, and this was especially true for those who endorsed a racially colorblind perspective. More broadly, we posit that this reflects internalized White supremacy on the part of Asian American students. We contend that one way to counter this is to increase the frequency of exposure to non-White Christianity. At the instructional level, faculty can be more intentional about revising their curriculum to include more non-White depictions of Christianity, whether that is actual images of Jesus, scholars cited, scholars assigned, and other ways in which Christianity might be portrayed. At the institutional level, administrators should support institutional efforts to recognize, name, and alter White standards masquerading as status quo or racially colorblind approaches to Christianity. Christian higher education institutions have the responsibility to foster open discussion around race and its intersection with religion and to be incisive in confronting the inadequate ways of integrating conversations about race and ethnicity in many Christian contexts. Indeed, our study implies that a continued promotion or tolerance of White Jesus depictions can strengthen internalized biases (e.g., an implicit bias that views White Jesus more favorably) on the part of Asian American students. Leaders should be clear in their messaging that White iconography in Christian settings can be harmful, leading to outcomes such as increased anti-Back attitudes (Howard & Sommers, 2019).
Our study also has important implications for research. In particular, we used images of Jesus instead of generic male faces (Howard et al., 2021) to tap into biases related to Jesus. Our significant findings related to the IAT suggest that this is a worthwhile methodological variation, and future studies should continue to explore how this kind of religious IAT can capture the psychological and spiritual experiences of Asian Americans. While not discounting the importance of the traditional IAT depicting racially diverse generic faces (e.g., Greenwald et al., 1998) as a research tool in multicultural psychology, we contend that our IAT involving images of Jesus is a novel and creative way to tap into yet another form of implicit bias. A future study, for example, might consider including both images of Jesus and generic male faces as a way to see how these two IAT scores might correlate with one another in addition to how they might differentially predict outcomes.
Limitations
This study has some limitations which can be used to inform future research. First, because there were no randomly assigned conditions, neither causality nor sequence can be established between participants’ racial colorblindness and implicit associations of Jesus as White, only that they are correlated. Future studies employing different research designs, such as an experimental design, are needed to gain a clearer picture of the causal relationship of the variables examined in this study. Second, three out of the top four Asian ethnicities represented were from East Asian settings (Chinese from mainland China, Japanese, and Korean). Different Asian ethnicities come from a diverse range of cultures, worldviews, beliefs, and practices that influence their views of Jesus’s races, and any two Asian Americans can have entirely different views of Jesus. This study did not consider how one’s diverse Asian American background could uniquely influence their views of Jesus. Future research is needed to examine specifically how one’s ethnic background and cultural experiences uniquely affect their views of Jesus. In addition, future research should strive to include a wider range of Asian ethnicities, especially South and Southeast Asian ethnicities. Third, although consistent with the gender breakdown at the institution from which data collection took place, our study included majority females. We encourage future researchers to be more strategic about recruitment across genders so that clearer conclusions can be drawn about how gender socialization might influence views related to Jesus’s race and their relation to cultural variables such as racial colorblindness. Fourth, although this study was conducted in a Christian higher education context, not all participants identified as Christian. Given that believing in Jesus can influence the way one views Him, identifying as a Christian can play a significant role in the way they perceive Jesus’s race. Future studies should include an entirely Christian sample to examine if findings are replicated and if believing in Jesus is an important moderating variable. However, we also recognize that colonial, white supremacy, and patriarchal ideals are deeply ingrained into Christian culture, and they may unconsciously influence our findings and how Jesus’s race is perceived; we contend that the Christian context of Christian higher education that guided our study’s sampling is also an important one.
Conclusion
Asian American students in our study generally viewed Jesus as non-White, but they also associated positive words with White Jesus and negative words with non-White Jesus. Racial colorblindness was identified as an important predictor of implicit bias. We hope that our study can serve as a call for educators and administrators to accurately represent Jesus in Christian higher education settings and for researchers to continue to explore the various representation of Jesus in IAT studies.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data collection for this article was supported by a Faculty Research and Scholarship Grant from the School of Psychology, Family, and Community at Seattle Pacific University.
