Are Brenda, Juanita, and Latoya More Feminine Than Jia and Neha? Gendered Evaluations of Racialized Names

Names can be easily implemented in controlled lab experiments as well as large-scale audit studies (Gaddis, 2018, 2019; Gaddis et al., 2021), particularly when the attachment of facial photos may be unusual or unfeasible (e.g., resumés typically do not include photos in the United States). Researchers have used names to signal a variety of identities such as gender (Moss-Racusin et al., 2012), race (Kenthirarajah et al., 2023), religion (Lajevardi, 2020), age (Newman et al., 2018), and nationality (Oreopoulos, 2011). When controlling for all other information, manipulation of names powerfully affects job callbacks (Bertrand & Mullainathan, 2004), email responses (Milkman et al., 2012), student evaluations (Zhao & Biernat, 2017), romantic desirability (Gebauer et al., 2012), harassment (Yan & Bernhard, 2023), criminal sentencing (Kenthirarajah et al., 2023), peer review (Huber et al., 2022), and even dog adoption (Quadlin & Montgomery, 2022).

Despite their ubiquity and usefulness, names are not without methodological challenges. Certain names are perceived as more attractive and youthful, which can (unintentionally) influence research outcomes (Kasof, 1993). Computational tools that draw gender inferences from names have particularly high error rates for Asian names (Lockhart et al., 2023). Importantly for this research, racialized names signal other identities such as social class (Gaddis, 2019; Simonsohn, 2015). Stereotypically Black and Hispanic names are seen as less educated and having lower income relative to Asian and White names (Crabtree et al., 2022).

Joining the call to critically examine the usage of names (Gaddis, 2018; Gaddis et al., 2021), this research investigates the perceived gender attributes of racialized names, with a focus on Chinese and Indian names. The gender of racialized names in research is often determined by societal norms, census popularity, and pilot testing (e.g., Gaddis, 2019; Milkman et al., 2012; Slepian & Galinsky, 2016). Nevertheless, participants’ perceptions may differ from researchers’ intentions. This poses a challenge when participants may not accurately infer the intended identity of names.

Names signal identities, presumably by activating stereotypes associated with the identity in question (Biernat et al., 2024; Kenthirarajah et al., 2023). In the United States (US), race and gender are highly intertwined to produce intersectional perceptions (for a review, see Lei et al., 2023). Gendered race theory argues that in the US, Asian men are perceived to be more feminine and less masculine than White and Black men while Asian women are hyper-feminized (Galinsky et al., 2013; Hall et al., 2015; K. L. Johnson et al., 2012; Schug et al., 2017; Wilkins et al., 2011). East Asian men are perceived to be more feminine and less masculine than South Asian men as well as Black and White men (Goh & Trofimchuk, 2023). Asian women are seen as highly feminine and they represent the prototype of womanhood (Lei et al., 2022). However, most research in this area used facial cues to signal race, with one notable exception: Hall and colleagues (2015) used names along with demographic forms to signal race and gender, and they found that Asian applicants (regardless of gender) were seen as particularly suited for feminine careers while Black applicants were suited for masculine careers. If names alone do activate racial-gender stereotypes of Asians, then Asian male and female names would be perceived as more feminine and less masculine than other racialized names in accordance with the gendered race theory (Hall et al., 2015; K. L. Johnson et al., 2012; Schug et al., 2017; Wilkins et al., 2011).

The initial aim of this research sought to examine if Asian names can activate gendered stereotyping. Study 1 was conducted with the pre-registered predictions that Asian male names would be seen as more feminine and less masculine than male names of other racial groups; in contrast, Asian female names would be seen as more feminine and less masculine than other racial groups. Unexpectedly, Study 1 found that Chinese and Indian male and female names were considered less gender-stereotypical than all other racial groups. This led to two additional studies to further determine the gendered evaluations of Asian ethnic names. In Studies 1 and 2, participants rated how feminine and masculine they perceived male and female names of five racial groups (i.e., Chinese, Indian, Black, Hispanic, and White). Study 3 asked participants to categorize the gender of each name and rate their confidence in knowing the gender.

Five hundred and one Prolific participants in the US completed the study. Thirty-three participants were excluded for missing an attention check or indicated that they responded randomly. With 468 participants (M _age = 32.21), 287 identified as women, 160 identified as men, and 21 identified as non-binary or preferred another term. There were 33 Black Americans, 313 White Americans, 26 East Asians, 17 South Asians, 20 Southeast Asians, 26 Hispanic/ Latinx people, 2 Native Hawaiians/ Pacific Islanders, 3 Middle Easterners, 25 multiracial individuals, and 3 indicated otherwise.

Participants saw a list of 30 names from audit studies on the housing market (Gaddis et al., under review; Gaddis & Ghoshal, 2020; Lu et al., 2021). There were three female and three male names for each of the five racial-ethnic groups (i.e., Chinese, Indian, White, Black, and Hispanic). Participants rated how feminine and how masculine they perceived each name to be (1 = strongly disagree; 5 = strongly agree). Participants also rated each name on perceived foreignness but this variable was not analyzed.

Because all interaction terms were significant, only the simple effects were reported here (see Figure 1). For details on the main effects, see Appendix B. Table 1 shows the descriptive statistics for each condition. All post-hoc tests used Bonferroni correction for multiple comparisons and examined differences in target race at each level of target gender.

Perceived femininity was analyzed with a 5 (target race: Chinese, Indian, Black, Hispanic, and White) x 2 (target gender: female vs. male) within-subjects ANOVA. The interaction was significant, F(3.11, 1453.71) = 1067.53, p < .001, partial η² = .70.¹

There was a significant effect for female names, F(4, 464) = 166.46, p < .001, partial η² = .59. Participants rated Chinese female names as the least feminine compared to Indian (p < .001, d_z = .44), Black (p < .001, d_z = 1.01), Hispanic (p < .001, d_z = .93), and White (p < .001, d_z = 1.16) names. Indian names also received lower femininity ratings than Black (p < .001, d_z = .67), Hispanic (p < .001, d_z = .57), and White (p < .001, d_z = .82) female names. Black and Hispanic female names did not differ (p = .141, d_z = .11), but both Black (p < .001, d_z = .31) and Hispanic (p < .001, d_z = .34) names were less feminine than White female names.

There was a significant effect for male names, F(4, 464) = 560.77, p < .001, partial η² = .83. Indian male names received the highest femininity ratings compared to Chinese (p < .001, d_z = .45), Black (p < .001, d_z = 1.57), Hispanic (p < .001, d_z = 1.72), White (p < .001, d_z = 1.97) names. Chinese male names were also rated as more feminine than Black (p < .001, d_z = 1.19), Hispanic (p < .001, d_z = 1.36), and White (p < .001, d_z = 1.57) male names. Black male names were considered more feminine than Hispanic (p = .018, d_z = .15) and White (p < .001, d_z = .52) names; Hispanic male names were rated as more feminine than White names (p < .001, d_z = .31).

There was also a significant interaction for perceived masculinity, F(3.19, 1487.43) = 998.69, p < .001, partial η² = .68. To probe the interaction, simple effects tests were conducted on the effect of race at each level of gender.

The effect of race was significant for women, F(4, 464) = 215.75, p < .001, partial η² =. 65. Chinese female names received higher masculinity ratings than Indian (p < .001, d_z = .43), Black (p < .001, d_z = 1.11), Hispanic (p < .001, d_z = 1.03), and White (p < .001, d_z = 1.32) female names. Indian female names were also rated as more masculine than Black (p < .001, d_z = .70), Hispanic (p < .001, d_z = .60), and White (p < .001, d_z = .85) names. Hispanic female names were considered more masculine than Black (p = .002, d_z = .17) and White names (p < .001, d_z = .35), while Black female names were rated more masculine than White names (p < .001, d_z = .24).

There was a significant effect for male names, F(4, 464) = 368.17, p < .001, partial η² =. 76. Indian male names were rated as the least masculine, compared to Chinese (p < .001, d_z = .42), Black (p < .001, d_z = 1.39), Hispanic (p < .001, d_z = 1.62), and White (p < .001, d_z = 1.53) male names. Chinese male names were similarly rated as less masculine compared to Black (p < .001, d_z = 1.00), Hispanic (p < .001, d_z = 1.18), and White (p < .001, d_z = 1.16) male names. Black and Hispanic male names did not differ (p = .130, d_z = .12). White male names were rated as more masculine than Black (p < .001, d_z = .34) and Hispanic (p = .010, d_z = .15) male names.

Despite the multiple comparisons, a clear trend emerged. Both Chinese and Indian names were considered less stereotypically gendered than other racialized names, hovering near the mid-points for perceived femininity and masculinity. Chinese and Indian female names were perceived as less feminine and more masculine than Black, Hispanic, and White female names. This was contrary to the pre-registered predictions that Asian female names would be rated as more feminine given what the gendered race theory would predict (Lei et al., 2023). Asian male names showed the reverse pattern as they were seen as more feminine and less masculine than male names from other racial groups. Although there are some differences between Chinese and Indian names as well as differences among Black, Hispanic, and White names, these differences are fairly small relative to the medium and large effect sizes observed in the comparisons between Asian names against Black, Hispanic, and White names for both men and women. The results could be due to the specific set of names used in this study, so a second set of names was selected to further examine the perceived femininity-masculinity of Asian names.

Because Study 1 found an unexpected pattern for Asian female names, a second study was conducted using a different set of names. Study 2 had the same pre-registered predictions that Asian male and female names would be seen as more feminine and more masculine than other racialized names.

Study 2 recruited 300 Prolific participants in the US and excluded 14 who missed an attention check or said they responded randomly. Of the 286 remaining participants (M _age = 31.83), 147 identified as women, 134 identified as men, and 5 identified as non-binary or preferred another term. There were 32 Black Americans, 176 White Americans, 21 East Asians, 9 South Asians, 11 Southeast Asians, 20 Hispanic/ Latinx people, 3 Native Hawaiians/ Pacific Islanders, 1 Middle Easterner, and 13 multiracial individuals.

Study 2 was a close replication of Study 1, which used the same materials and procedure with two exceptions. First, Study 1 included the perceived foreignness of the names as an exploratory variable that was not analyzed, but this variable was not included in Study 2. Second, Study 2 used a different set of names. Study 2 used 20 names from Milkman et al. (2012, 2015), and there were two male and two female names for each of the five racial groups (i.e., Chinese, Indian, Black, Hispanic, and White). Participants rated each name’s perceived femininity and masculinity as two separate scales (1 = strongly disagree; 5 = strongly agree). See Appendix A for all the names.

All interactions were significant and only the simple effects were described here (see Figure 2). See Appendix B for the main effects. Table 1 shows the descriptive statistics for each condition. All post-hoc tests used Bonferroni correction for multiple comparisons.

Study 2 used a separate list of names, analyzed with a 5 (target race: Chinese, Indian, Black, Hispanic, and White) x 2 (target gender: female vs. male) within-subjects ANOVA. The interaction was significant, F(2.49, 708.69) = 308.32, p < .001, partial η² = .52.

There was a significant effect for female names, F(4, 282) = 102.78, p < .001, partial η² = .59. Chinese and Indian female names did not differ (p = .172, d_z = .14). Chinese female names were rated as less feminine compared to Black (p < .001, d_z = .87), Hispanic (p < .001, d_z = .99), and White (p < .001, d_z = 1.05) names. Indian female names also received lower femininity ratings than Black (p < .001, d_z = .84), Hispanic (p < .001, d_z = .98), and White (p < .001, d_z = 1.02) names. Black female names were considered less feminine than Hispanic (p = .001, d_z = .23) and White (p < .001, d_z = .25) female names, while Hispanic and White female names did not differ (p >.999, d_z = .03).

The effect of race was significant for male names, F(4, 282) = 89.66, p < .001, partial η² = .56. Chinese male names were considered the most feminine compared to Indian (p < .001, d_z = .34), Black (p < .001, d_z = .78), Hispanic (p < .001, d_z = .91), and White (p < .001, d_z = 1.05) male names. Likewise, Indian male names were more feminine than Black (p < .001, d_z = .50), Hispanic (p < .001, d_z = .63), and White (p < .001, d_z = .74) names. Black male names received higher femininity ratings than Hispanic (p = .002, d_z = .22) and White (p < .001, d_z = .32) names. Hispanic names were rated more feminine than White male names (p = .015, d_z = .19).

There was a significant interaction for perceived masculinity, F(2.75, 784.58) = 385.22, p < .001, partial η² = .58. The simple effect of race was significant for female names (F(4, 282) = 108.483, p < .001, partial η² = .61) and male names (F(4, 282) = 105.65, p < .001, partial η² = .60).

Chinese female names were rated as the most masculine compared to Indian (p = .001, d = .23), Black (p < .001, d_z = .97), Hispanic (p < .001, d_z = 1.12), and White (p < .001, d_z = 1.08) female names. Indian female names were seen as more masculine than Black (p < .001, d_z = .79), Hispanic (p < .001, d_z = .96), and White (p < .001, d_z = .90) female names. Black female names were also more masculine than Hispanic names (p = .002, d_z = .22), while not significantly different from White names (p = .057, d_z = .17). Hispanic and White female names did not differ significantly (p > .999, d_z = .04).

Chinese male names were rated as the least masculine compared to Indian (p < .001, d_z = .31), Black (p < .001, d_z = .91), Hispanic (p < .001, d_z = 1.09), and White (p < .001, d_z = 1.10) male names. Indian male names were considered less masculine compared to Black (p < .001, d_z = .64), Hispanic (p < .001, d_z = .83), and White (p < .001, d_z = .81) male names. Black male names were perceived as less masculine than White male names (p = .001, d_z = .23) but not significantly different than Hispanic male names (p = .051, d_z = .17). White and Hispanic male names did not differ (p > .999, d_z = .05).

Contrary to the pre-registered predictions but replicating Study 1, Study 2 used a different set of racialized names to show that Chinese and Indian female names were consistently perceived as more masculine and less feminine than female names of other racial groups (i.e., Black, Hispanic, and White female names). Chinese and Indian male names were also considered less masculine and more feminine than the other three racial groups. The consistent patterns observed for both Studies 1 and 2 may reflect participants’ unfamiliarity with Asian names, which was investigated in the next study.

Studies 1 and 2 reliably found that Chinese and Indian names were considered less gender-stereotypical than names from other racial groups. The effects were consistent for both male and female names. One possibility might be due to participants’ unfamiliarity with Chinese and Indian names. Study 3 asked participants to guess the gender of each name (with an “Unsure” option) and rate their confidence in their guesses. It was predicted that participants would be more likely to select the “Unsure” option and expressed lower confidence for Chinese and Indian names than Black, Hispanic, and White names (for both male and female names).

Study 3 recruited 349 US participants through the CloudResearch Connect platform. Connect is a newer and more affordable participant-recruitment platform (Hartman et al., 2023), and it shows comparable strong data quality with Prolific (Douglas et al., 2023). Three participants were excluded for missing an attention check or said they responded randomly. Of the 346 remaining participants (M _age = 41.10), 171 identified as women, 173 identified as men, and 2 identified as non-binary or preferred another term. There were 26 Black Americans, 267 White Americans, 16 East Asians, 4 South Asians, 4 Southeast Asians, 16 Hispanic/ Latinx people, 3 Native Americans/ Indigenous people, 2 Middle Easterners/ North Africans, 7 multiracial people, and 1 identified with a category that was not listed.

Participants saw a list of 50 names, combining all the names from Studies 1 and 2 (see Appendix A). Within each of the five racial groups, there were five male and five female names. Participants first guessed the gender of each name with one of three options: Male, Female, and Unsure. Afterward, participants rated how confident they were in knowing the gender of each name (1 = Not confident at all; 5 = Extremely confident).

As shown in Table 2, participants indicated greater uncertainty (i.e., selecting “Unsure”) for Chinese and Indian names as well as selecting another gender than the intended gender of the names (e.g., guessing a male name as female). Using Chi-Square goodness-of-fit for female names, the observed frequency of “Unsure” responses across the five racial groups differed significantly from the expected value (based on the total “Unsure” responses of 542 for female names, divided by 5), χ² = 611.40, p < .001. The observed frequency for male names was also significantly different compared to the expected value (based on the total “Unsure” responses of 425 for male names, divided by 5), χ² = 521.84, p < .001.

To examine confidence in knowing the gender of each name, a 5 (target race: Chinese, Indian, Black, Hispanic, and White) x 2 (target gender: female vs. male) within-subjects ANOVA was conducted. The interaction was significant, F(3.22, 1111.49) = 55.36, p < .001, partial η² = .14. There was a significant simple effect of race for female names (F(4, 342) = 247.78, p < .001, partial η² = .74) and for male names (F(4, 342) = 252.46, p < .001, partial η² = .75). See Table 2 for descriptive statistics. See Appendix B for the main effects.

Decomposing by gender, the only non-significant pairwise comparison for female names was between the two Asian groups (p = .057, d_z = .15); all other pairwise comparisons differed significantly at ps < .001. Participants felt less confident about the gender of Chinese female names relative to Black (d_z = 1.53), Hispanic (d_z = 1.54), and White (d_z = 1.60) female names. Likewise, participants were less confident about the gender of Indian female names relative to Black (d_z = 1.51), Hispanic (d_z = 1.50), and White names (d_z = 1.56). Participants were less confident about Hispanic female names compared to Black (d_z = .23) and White names (d_z = .60), and less confident about Black than White names (d_z = .40).

For male names, all racial groups differed significantly (all pairwise ps < .001). Participants expressed less confidence about the gender of Chinese male names relative to Indian (d_z = .64), Black (d_z = 1.48), Hispanic (d_z = 1.62), and White names (d_z = 1.69). Similarly, participants felt less confident about Indian male names relative to Black (d_z = 1.24), Hispanic (d_z = 1.36), and White names (d_z = 1.40). Participants were less confident about Black male names compared to Hispanic (d_z = .37) and White names (d_z = .67), and less confident about Hispanic than White names (d_z = .44).

When offered the option to select “Unsure” in identifying the gender of racialized names, participants expressed greater uncertainty regarding Chinese and Indian names. Participants also expressed less confidence in knowing the gender of Chinese and Indian names relative to other racialized names. These patterns were observed for male and female names. Study 3 built on the previous two studies to suggest that American perceivers may be less familiar with the (researcher-intended) gender of Asian ethnic names.

Three studies examined the gendered evaluations of Chinese, Indian, Black, Hispanic, and White names. Consistently, Chinese and Indian names were considered less gender-stereotypical than Black, Hispanic, and White names. Contrary to the gendered race theory that predominantly used facial stimuli to show the hyper-feminization of Asian women (e.g., Galinsky et al., 2013; Hall et al., 2015; K. L. Johnson et al., 2012; Lei et al., 2022; Schug et al., 2017), Chinese and Indian female names were perceived to be less feminine and more masculine than Black, Hispanic, and White female names. Chinese and Indian male names were considered more feminine and less masculine than male names from all other racial groups. Participants also expressed greater uncertainty and lower confidence about the gender of Chinese and Indian names. Taken together, these studies suggest that the intended gender of Asian ethnic names may be unintentionally downplayed, which raises methodological concerns.

Research using facial photographs has consistently demonstrated that (East) Asian women are seen as particularly feminine and they embody the prototype of womanhood (Lei et al., 2022, 2023). However, Asian female names alone cannot elicit such gendered stereotyping (Bailey et al., 2024), suggesting that intersectional stereotyping of Asian women may require the processing of facial cues and/or more elaborate verbal cues such as in combination with a demographic form (Hall et al., 2015). Although Asian male names may be consistent with the gendered race theory by showing that both Chinese and Indian male names were seen as effeminate, it is more likely the case that Asian male names were considered less gender-stereotypical and more gender-ambiguous than other racialized male names. In both Studies 1 and 2, Chinese and Indian names scored near the mid-points (for both male and female names). Study 3 found that participants were more uncertain and less confident about the (researcher-intended) gender of Chinese and Indian names. This may be driven by (predominantly White) American participants’ unfamiliarity with Asian ethnic names. It is possible that clarifying the gender of these Asian names may trigger gendered stereotyping consistent with facial cues.

Gendered evaluations of racialized names are likely anchored on prototypically White, Anglicized names. White female names were seen as the most feminine and least masculine, while White male names were seen as the most masculine and least feminine. White Americans are seen as the prototypical American (Devos & Banaji, 2005; Zou & Cheryan, 2017) and Whiteness is the default racial assumption in the US (Purdie-Vaughns & Eibach, 2008; Thomas et al., 2014). As such, evaluations tend to be anchored around White Americans. East Asian men are seen as effeminate and Black men are considered hyper-masculine because they are judged in relation to White men who are considered to embody the ideal American manhood (Lei et al., 2023). Unfamiliarity with Asian ethnic names is likewise driven by greater familiarity and popularity of Anglicized White names that are deemed more normative, particularly among the predominantly White samples across all three studies.

Gender stereotyping is attenuated for Asian names relative to other racialized names, which holds implications beyond research methodology. One (mis)interpretation of the findings could be that Asian women face less gender bias. However, women who appear less feminine in their job applications receive penalties for breaking prescriptive gender stereotypes (He & Kang, 2021). Audit studies that used these exact Asian names have further demonstrated that Asian men and women receive fewer responses than inquiries from White male names (Milkman et al., 2012, 2015). Even if Asian women do not face as much gender bias in specific instances that only involve their names, they would still face penalties along the racial dimension, which is likely the case in these audit studies (Lu et al., 2021; Milkman et al., 2012). Nevertheless, evaluations are based on multiple cues beyond names in the real world. For instance, in-person interviews are typically held after evaluations of resumés, which may provoke discrimination via cues such as phenotype, accent, attire, and more (Kang et al., 2016; Lai & Banaji, 2020). It is unlikely that names could prevent Asian women from the prevalence and intersection of gender and racial bias (Mukkamala & Suyemoto, 2018).

Unlike previous research that has found simultaneous signaling of multiple identities (Crabtree et al., 2022), this research found that an intended identity is attenuated. These findings raise potential methodological concerns and considerations for researchers who wish to signal the gender of Asian ethnic names. Two potential solutions are offered with caveats to consider: The first solution uses Anglicized names, and the second uses gender pronouns and titles.

First, researchers could use Anglicized first names with racialized last names to signal gender and race more clearly (e.g., Bruce Lee instead of Jun Lee). After all, gender attribution is restricted to first names and not last names; last names do not typically convey information about gender, but they do signal race and ethnicity (Crabtree & Chykina, 2018; Lauderdale & Kestenbaum, 2000). Gender signals could be clearer and more aligned with other racialized groups if researchers used Anglicized first names for Asians. This does have ecological validity as well, given that Asian Americans and Asian international students in the US often adopt Anglicized names (Fang & Fine, 2020; Zhao & Biernat, 2018) and Asian job applicants engage in “White-washing” of their resumés by adopting Anglicized names (Kang et al., 2016).

Although this method may distinguish gender more clearly, it may signal other identities, such as assumptions of immigration status and English proficiency (Gaddis & Ghoshal, 2020). Anglicizing Asian names increases response rates and generates more favorable evaluations relative to Asian ethnic names (Biernat et al., 2024; Zhao & Biernat, 2017). Although using Anglicized first names may signal racial-ethnic and gender information more clearly, it introduces other important methodological elements to consider and it may fundamentally shift participants’ responses. Furthermore, there may be ethnic differences among Asian Americans in the adoption of Anglicized names, as evidence suggests that South Asians prefer using ethnic names over Anglicized names (Cila et al., 2021). There could disproportionately be more East Asians than South Asians with Anglicized first names. This case is illustrated in the movie Harold and Kumar Go to White Castle, in which Harold is a Korean American character and Kumar is an Indian American character. If researchers Anglicized South Asian names, this may not accurately represent the population in question.

Another solution for cueing gender more clearly is to supplement Asian names with gender pronouns (e.g., “she/ her/ hers”) or titles (e.g., “Mr./ Ms./ Mrs.”). The inclusion of gender pronouns is increasingly common in email signatures, websites, and resumés. This solution is easy to implement in audit studies using resumés and emails.

However, these cues may signal other (unintended) identities as well. Gender pronouns on company websites signal inclusivity for LGBTQ+ members (I. R. Johnson et al., 2021), but pronouns have also become a point of contention in an increasingly polarized political climate in the US (Izaguirre, 2023). Including gender pronouns on a resumé may signal progressive political orientation or Democratic Party affiliation. Additionally, relationship status is signaled through gendered titles for women (i.e., Ms. vs. Mrs.). Single people are stigmatized more and receive less social support than married or coupled people (DePaulo & Morris, 2005; Girme et al., 2022). As such, gender pronouns and titles introduce additional methodological elements that need to be taken into consideration and merit further research on their effectiveness.

Although the presented solutions may not be foolproof, this demonstrates the complexity that names carry. As others have advocated, more research is needed to better understand this valuable research tool (Gaddis, 2018). These studies also demonstrate the value of more extensive pilot testing to understand how participants perceive names. If the goal is to use the most gender stereotypical names, then multiple variables should be considered in the pilot-testing such as perceived femininity-masculinity alongside confidence level. Additionally, Study 3 introduced the “unsure” option to capture participants’ uncertainties in gender categorization rather than using a male-female binary force-choice, and this method could also be used to assess participants’ certainty and confidence in categorizing the race or ethnicity of these names.

Ultimately, race is bundled and intersectional (Sen & Wasow, 2016). It may be difficult and perhaps even misguided to isolate race from other related or intertwined identities (Crabtree et al., 2022). The uncertainties that people have about Asian names do reflect the prejudice and microaggressions that Asian Americans regularly confront in the real world (Kohli & Solórzano, 2012). Removing the uncertainty that people have may affect the research question itself. Signals may never be entirely clear even if we use stimuli like facial photos, which could be affected by a myriad of factors such as skin tone, facial hair, and hairstyle. Like any other research tool, names carry potential confounds that need to be carefully considered.

It is important to note that all three studies used two specific sets of names from prior audit studies (Gaddis et al., under review; Lu et al., 2021; Milkman et al., 2012, 2015). These two sets were chosen because both contained male and female names from the same five racial-ethnic groups. Future studies could test racialized names from other studies. Furthermore, these studies were limited to Chinese, Indian, Black, Hispanic, and White names. Future research could examine a greater variety of racial and ethnic groups (e.g., Arab, Japanese, Vietnamese, and Russian names).

Additionally, all three studies had predominantly White American participants, which is consistent with previous studies on gendered race theory (Goh & Trofimchuk, 2023; Hall et al., 2015; Wilkins et al., 2011). This may explain why participants were more unsure and less confident about the gender of the Asian names. Future research should examine if Asian Americans are more accurate and confident in categorizing the gender of Asian names. Chinese Americans would likely be more familiar with the gender of Chinese names, and they could rate the names more gender-stereotypically. It would be interesting to see if this familiarity is generalizable in cross-ethnic judgments. For instance, how would Indian Americans rate the femininity and masculinity of Chinese ethnic names? Asian Americans are not a monolith, and it is important to consider their diversity in terms of both stimuli and participants (Goh & Trofimchuk, 2023; Vinluan & Remedios, 2024).

This research is conducted on participants in the US. As such, the results here can only speak to how racialized names are perceived within the US context. Unfamiliarity with Asian names likely would not generalize to participants residing in certain Asian countries. Participants in Asian countries, likewise, may not be familiar with the gender of Black and Hispanic names.

However, White names may show a certain generality given the popularity of Western media around the world, which could help people associate certain names with a particular gender. Furthermore, colonization and militarization could shape different countries’ familiarity with Western cultures (Readsura Decolonial Editorial Collective, 2022). The United Kingdom had colonized much of the world and the vestiges of its imperial power (i.e., coloniality) may enforce familiarity with Anglicized names among people in the Commonwealth (e.g., Malaysia, India, and South Africa). Similarly, American military activities in Asia (e.g., the Korean War and Vietnam War) and the colonization of the Philippines could have imprinted familiarity with Anglicized, White names. Filipino participants, in such a case, may also be familiar with Hispanic names, given the history of Spanish colonization. In fact, many Filipinos have Hispanic names (Ocampo, 2016). As such, participants in many parts of the world could likely identify the gender of stereotypically White names like Brenda and Steven. But the asymmetry of (White) Americans’ unfamiliarity with Asian names reveals the unbalanced nature of societal power.

Names are commonly used in social science research to examine identity-based discrimination (Gaddis et al., 2021), but these names do not always reflect what the researchers intend (Crabtree et al., 2022; Gaddis, 2019). Three studies found that people perceived Indian and Chinese names (both male and female names) to be less gender-stereotypical than Black, Hispanic, and White names, likely because participants expressed greater uncertainty about the gender of Asian names. Names may activate identities that researchers may not intend, and they may also attenuate the identities that researchers intend. Despite such methodological concerns, this research does not advocate for the retirement of names from our methodological toolkit. It is a valuable and effective method that is easy to implement. Rather, greater consideration should be made in the study design and caution should be exercised in the data interpretation.

The author thanks Vlada Trofimchuk and Taylor Douglas for their assistance on the paper.

The author has no conflicts of interests to disclose.

All studies received approval from a university Institutional Review Board prior to data collection.

See Appendix A for all the names used in the three studies that were drawn from prior research; all names were selected based on Census popularity and extensively pilot-tested by the original authors (for details, see Gaddis et al., under review; Lu et al., 2021; Milkman et al., 2012, 2015). The sample size for all studies was determined by the availability of funding but ensured to exceed 80% power using a conservative metric in power calculation (N required = 218 given effect size f = 0.1, alpha = .05, power = .80, 2 groups and 5 measurements, and correlation of .10 among repeated measures). Power was calculated using G*Power (Faul et al., 2007). The data, pre-registrations, and materials for all three studies are available on Open Science Framework: https://osf.io/uzkw8/

There was a significant main effect of gender, F(1, 467) = 4057.69, p < .001, partial η² = .90. Female names (M = 4.07, SE = .02) were perceived to be more feminine than male names (M = 1.94, SE = .02).

There was a significant main effect of race, F(3.29, 1536.47) = 114.25, p < .001, partial η² = .20. Indian names (M = 3.37, SE = .03) were more feminine than all other racialized names, all ps < .001. Chinese (M = 2.99, SE = .03) names were less feminine than Indian names, p < .001. Chinese names did not differ from Black names (M = 2.94, SE = .02), p > .999. Chinese names were more feminine than Hispanic (M = 2.84, SE = .02; p < .001) and White names (M = 2.89, SE = .02, p = .018). White names did not differ from Black (p = .129) and Hispanic names (p = .438). Black names were considered more feminine than Hispanic names (p = .004).

There was a significant main effect of gender, F(1, 467) = 2974.44, p < .001, partial η² = .86. Female names (M = 1.87, SE = .03) were perceived as less masculine than male names (M = 3.91, SE = .02).

There was a significant main effect of race, F(3.04, 1417.12) = 78.74, p < .001, partial η² = .14. Chinese (M = 2.93, SE = .03) names were more masculine than Indian (M = 2.56, SE = .03) names, p < .001. Chinese names did not differ from Black names (M = 2.94, SE = .02; p > .999) and White names (M = 2.96, SE = .02; p > .999). Chinese names were less masculine than Hispanic (M = 3.04, SE = .02; p = .001). Indian names were less masculine than all other racialized names, all ps < .001. Black names were considered less masculine than Hispanic names (p < .001) but not different from White names (p > .999). Hispanic names were more masculine than White names (p = .034).

There was a significant main effect of gender, F(1, 285) = 1694.53, p < .001, partial η² = .86. Female names (M = 4.26, SE = .03) were rated as more feminine than male names (M = 1.59, SE = .04).

There was a significant main effect of race, F(2.69, 766.29) = 13.66, p < .001, partial η² = .05. Chinese names (M = 2.89, SE = .04) did not differ significantly from Indian (M = 2.81, SE = .03; p = .157), Black (M = 2.98, SE = .02; p = .188), Hispanic (M = 2.99, SE = .02; p = .086), or White (M = 2.96, SE = .02; p = .366) names. Indian names were considered less feminine than Black, Hispanic, and White names, all ps < .001. Black, Hispanic, and White names all did not differ significantly from one another, ps > .999.

There was a significant main effect of gender, F(1, 285) = 2266.80, p < .001, partial η² = .89. Female names (M = 1.70, SE = .03) were rated as less masculine than male names (M = 4.33, SE = .03). There was not a significant main effect of race, F(2.69, 766.35) = 1.92, p = .132, partial η² = .01.

There was a significant main effect of gender, F(1, 345) = 91.21, p < .001, partial η² = .21. Participants were less confident about female names (M = 3.82, SE = .03) than male names (M = 3.95, SE = .03).

There was a significant main effect of race, F(1.66, 572.11) = 793.35, p < .001, partial η² = .70. Participants were the least confident about the gender of Chinese names (M = 2.73, SE = .06) compared to Indian (M = 3.06, SE = .05), Black (M = 4.45, SE = .03), Hispanic (M = 4.48, SE = .03), or White names (M = 4.70, SE = .03), all ps < .001. There was less confidence for Indian names than Black, Hispanic, and White names, all ps < .001. Black and Hispanic names did not differ in confidence level, p > .999. Participants were more confident about White names than Black and Hispanic names, all ps < .001.