We Want Black Students, Just Not You.pdf

792579 research-article2018 SREXXX10.1177/2332649218792579Sociology of Race and EthnicityThornhill Diversity Ideologies and Realities We Want Black Students, Just Not You: How White Admissions Counselors Screen Black Prospective Students Sociology of Race and Ethnicity 2019, Vol. 5(4) 456­–470...

792579 research-article2018 SREXXX10.1177/2332649218792579Sociology of Race and EthnicityThornhill Diversity Ideologies and Realities We Want Black Students, Just Not You: How White Admissions Counselors Screen Black Prospective Students Sociology of Race and Ethnicity 2019, Vol. 5(4) 456­–470 © American Sociological Association 2018 https://doi.org/10.1177/2332649218792579 DOI: 10.1177/2332649218792579 sre.sagepub.com Ted Thornhill1 Abstract Most historically and predominantly white institutions (HPWIs) now desire some number of black students on their campuses. However, recent theoretical scholarship suggests that HPWIs’ desire for and willingness to embrace black students is predicated on their racial palatability. The theory of intraracial discrimination stipulates that white gatekeepers are increasingly inclined to screen blacks to “weed out” those they perceive as too concerned with race and racism. In this study, the author assessed whether there was evidence of intraracial discrimination within the HPWI admissions regime. The data were derived through an audit of 517 white admissions counselors, employed at the same number of institutions, who received inquiry e-mails from fictitious black high school students who presented as more or less racially salient. The findings reveal that white admissions counselors are more responsive to black students who present as deracialized and racially apolitical than they are to those who evince a commitment to antiracism and racial justice. These findings provide convincing support for the theory of intraracial discrimination within the HPWI admissions regime. The author concludes with a brief discussion of the implications of these findings. Keywords racism, antiracism, racial discrimination, black college students, college admissions, higher education Over the past four decades, federal court rulings (e.g., Regents of the University of California v. Bakke, Gratz v. Bollinger, Grutter v. Bollinger, and Fisher v. Texas I and II), public referenda (e.g., California and Michigan), state law (New Hampshire), and executive order (Florida) have constrained the ability of colleges and universities to legally consider race as a factor in admissions. Yet these prohibitions do not eliminate informal considerations of race or white racist practices within the historically and predominantly white institution1 (HPWI) admissions regime. Much scholarship has found that the United States remains a highly racialized society, sustaining and generating manifold disparate outcomes by race (e.g., Bonilla-Silva 2001; Brown et al. 2003; Feagin 2014). As racial actors within a “racialized social system” (Bonilla-Silva 1997, 2016), admissions personnel and prospective students and applicants ensure that race remains salient within the HPWI admissions regime, irrespective of racebased affirmative action policies. Recent critical race scholarship posits that racist practices do not hinge solely on racial ascription and traditional in-group/out-group racial dynamics. Whites can also discriminate against blacks and other people of color intraracially, on the basis of their perceptions about the extent to which race informs a given individual’s self-concept (i.e., their racial salience). Therefore, how people of color 1 Florida Gulf Coast University, Fort Myers, FL, USA Corresponding Author: Ted Thornhill, Florida Gulf Coast University, 10501 FGCU Boulevard South, Fort Myers, FL 33965, USA Email: [email protected] Thornhill live, think, and adorn their racialized bodies could result in disparate treatment (e.g., in college admissions, hiring, and promotion) on the basis of whites’ intraracial evaluations (Carbado and Gulati 2013; Thornhill 2015). Although the theoretical scholarship discussed below is highly suggestive, as of this writing, screening along the lines of racial salience, particularly within the HPWI admissions regime, has not been tested empirically. As such, in this study, I examine whether there is evidence of white admissions counselors’ engaging in intraracial discrimination vis-à-vis black prospective students. Specifically, I use the correspondence method to answer the following research question: do white college admissions counselors intraracial screen black high school students who betray a commitment to antiracism by responding to their inquiry e-mails less frequently than those black students who reveal no interest in racial matters? Intraracial Discrimination In the “postracial” era, when interracial discrimination is illegal and socially unacceptable, attempts to completely exclude blacks from organizations are largely impracticable. Instead, many organizations now desire the presence of at least some blacks (Embrick 2011; Herring 2009). However, this does not mean that all blacks, whose qualifications should make them competitive candidates for jobs or college admission, are deemed acceptable by white employers or admissions personnel. Legal scholars Devon Carbado and Mitu Gulati (2013), using employers as an exemplar, stated that most prefer “good blacks” [who] will think of themselves as people first and black people second (or third or fourth); they will neither “play the race card” nor generate racial antagonism or tensions in the workplace; they will not let white people feel guilty about being white; and they will work hard to assimilate themselves into the firm’s culture. The screening of African Americans along these lines enables the employer to extract a diversity profit from its African American employees without incurring the cost of racial salience. The employer’s investment strategy is to hire enough African Americans to obtain a diversity benefit without incurring the institutional costs of managing racial salience. (P. 2) Carbado and Gulati (2013) argued that in addition to standard interracial conceptions of discrimination, whites also discriminate against blacks 457 intraracially. Their central argument is that whites’ cognitive radars can be tuned to detect blacks’ ideological and political commitments, particularly as they concern racial matters, to evaluate whether they are congruent with whites’ color-blind expectations. Perceived deviation from the color-blind imperative can then become a basis for whites to discriminate against racially salient blacks. Carbado and Gulati contended that these distinctions extend beyond skin tone (Hannon 2015), social class (Lacy 2007), and ethnicity (Watts Smith 2014). Buttressed by both historical evidence (Thornhill 2015) and the scholarship on color-blind ideology (Bonilla-Silva 2018), Carbado and Gulati’s theory of intraracial discrimination is persuasive. The term intraracial discrimination can call to mind the familiar concept of colorism. Hence, a point of clarification is in order. The phenomenon whereby darker skinned people of color experience greater stigmatization and discrimination than their lighter skinned counterparts has been well documented (Hannon 2015; Hunter 2007). However, Carbado and Gulati’s (2013) contention is that there are other ways that whites, and conceivably other people of color (see Thornhill 2015), can discriminate among blacks (or other people of color). The differing perspectives and proposed strategies of W.E.B. Du Bois and Booker T. Washington (both light-skinned blacks) vis-à-vis blacks’ station in U.S. society provide a useful and illustrative starting point for thinking about whites’ preference for manifestations of blackness that are independent of racial physicality and centered instead on racial ideology and racial politics. Both Du Bois and Washington wanted greater life chances for blacks. However, they embraced radically different philosophies and approaches for the realization of this shared goal. The historical record shows many whites’ support for Washington’s focus on vocational training and industrial development for blacks and a depreciation of black political equality (see Harlan 1972). In contrast, Du Bois came to endorse a vision of black equality that demanded immediate and unmitigated access to the franchise and higher education as well as the moral imperative for blacks to speak out against all forms of white racism (Du Bois [1903] 2005), a scenario anathema to nearly all whites of the period. Whites’ support of Washington’s program and opposition to Du Bois’s was not a function of skin tone; whites’ belief in hypodescent rendered them both black. It was Washington’s accommodationism that garnered his support among whites, not his complexion (Harlan 1972). Given whites’ desire to maintain 458 their dominant position in the social order, it is unlikely that they would have been less supportive of Washington’s accommodationism if he had a darker complexion. Racial politics can supersede color as a criterion used by whites to evaluate blacks intraracially. For a contemporary example, one need only consider the case of former National Football League quarterback Colin Kaepernick, a lightskinned black American, whose national anthem protest against police brutality and racial injustice resulted in pariah status among team owners and condemnation from millions of whites. Then, as now, being black and light-skinned did not, and does not, appear to garner greater white sympathy for one’s antiracist politics. Yet although intraracial discrimination seems likely to exist independent of color, the scholarship on intersectionality suggests that gender is more closely linked. Drawing on Kimberlé Crenshaw’s (1989) pioneering work on intersectionality, Carbado and Gulati (2013) argued that intraracial discrimination is also gendered. They contended that whites’ intraracial evaluations of blacks incorporate considerations of gender whereby they can (1) assess black women relative to black men and (2) make judgements about suitability among black women. Therefore, whether and the extent to which white admissions counselors engage in intraracial screening of black prospective students could be a function of student gender. Moreover, given that white admissions counselors are also gendered, this too could affect how they screen black prospective students. Hence, I take gender into account in both my research design and the analysis of my data. Despite strong theorization and suggestive historical evidence, whether whites treat racially salient blacks more negatively than their less racially salient counterparts remains an open question. An empirical assessment can provide us with greater understanding as to the existence and extent of this alternative form of white racism. In this study, I test two hypotheses suggested by Carbado and Gulati’s (2013) theory of intraracial discrimination: Hypothesis 1: Non–racially salient black students will receive a higher level of responsiveness from white college admissions counselors than will racially salient black students. Hypothesis 2: Racially salient black students whose interests and extracurricular involvements challenge the validity of color-blind ideology will experience the lowest level of responsiveness from white college admissions counselors. Sociology of Race and Ethnicity 5(4) White Racism on Campus: The Experience of Black Students There is much rigorous and empirically substantiated race theory that has documented the structural and systemic character of racism (Bonilla-Silva 1997; Brown et al. 2003; Feagin 2006). This research shows that the legacies of state-sponsored racial oppression, and a now hegemonic colorblind ideology, have enabled white racist practices to persist in formal organizations across social institutions (Bonilla-Silva 2018; Feagin 2014). HPWIs are not immune to this state of affairs. Studies have found that black students at HPWIs routinely experience white racism in classrooms, residence halls, faculty and academic advisers’ offices, libraries, and other public spaces on and adjacent to campus (Feagin, Vera, and Imani 1996; Harwood et al. 2012; Johnson-Ahorlu 2012; Smith et al. 2016; Solórzano, Ceja, and Yosso 2000; Suarez-Balcazar et al. 2003). Sometimes the perpetrators are students. In other instances, faculty members, staff members, campus police, or administrators are responsible. Also implicated are the institutional policies, practices, and traditions that contribute to racially inequitable outcomes that penalize and marginalize black students. Collectively, these environments can make it difficult for black students to realize, in an equitable manner, the opportunities (e.g., on-campus jobs that provide career-related experience, internships, research opportunities) and resources (e.g., at on-campus career centers, tutoring and writing centers, academic advising offices, and counseling and psychological services centers) that increase the likelihood of graduation, earning an advanced degree, and/or securing employment that requires a four-year college degree (Feagin et al. 1996; Harper, Smith, and Davis 2016; Johnson-Ahorlu 2012). It is within this context that student-led protests erupted on college campuses nationwide during the 2015–2016 academic year (see Dickey 2016). In response, several of these HPWIs publicly acknowledged students’ grievances, with some agreeing to one or more of the following: working to further diversify the student body and/or hire more faculty members of color (e.g., the University of Missouri); modifying the curriculum (e.g., Brown University); requiring implicit bias training for campus police or safety officers (e.g., Syracuse University); creating dedicated spaces on campus for students of color to study, congregate, and decompress (e.g., Ithaca College); hiring one or more diversity officers (e.g., Thornhill the University of Alabama); removing or modifying racist iconography from campus buildings (e.g., the University of Kentucky); and/or removing white racists’ names from eponymous campus buildings (e.g., Georgetown University). If fully realized, these changes could contribute to more racially hospitable and equitable environments on these campuses. However, history testifies to the uncanny ability of white supremacy to adapt to changes that are disruptive to its ideological and structural stranglehold (Bonilla-Silva 2001; Omi and Winant 2015). Therefore, skepticism seems warranted, especially considering recent national survey data on college and university presidents’ perceptions of the state of “race relations” in higher education. White Racism on Campus: Delusional Administrators Despite overwhelming evidence highlighting the depth and ubiquity of white racism on campuses across the country, most senior administrators remain in denial. This can be observed in the results of an annual survey of college and university presidents conducted by Gallup and Inside Higher Ed in spring 2016, a time when dozens of antiracist student protests were under way on campuses across the country (see Dickey 2016). These administrators were asked a series of questions about “race relations” on their campus and campuses nationwide. Eighty-four percent of college presidents indicated that “race relations” on their campus were either “excellent” or “good” (the number was 81 percent in 2015). Yet when asked about “race relations” on college and university campuses nationwide, fewer than 1 percent of presidents surveyed indicated that they were “excellent,” and only 24 percent thought they were “good” (the two numbers totaled 42 percent in 2015). Furthermore, the number of college presidents who thought “race relations” on campuses nationwide were “better than [they were] five years ago” was halved between 2015 and 2016, from 30 percent to 15 percent, respectively (Jaschik and Lederman 2015, 2016). College presidents’ views on the overall state of “race relations” on college and university campuses have become increasingly negative. Yet they remain confident that the racial climate on their own campuses is “excellent” or “good.” Indeed, studies have shown that administrators at HPWIs often describe their campuses as racially hospitable environments that are only infrequently 459 sites of isolated “racial incidents” at variance with their institutional values of diversity, equity, and inclusion (Davis and Harris 2016; Embrick, Brunsma, and Thomas 2016). There is a clear disjuncture between many administrators’ delusional and racism evasive statements and the profusion of empirical research, news reports, and personal testimonies that highlight the presence of much white racism at HPWIs. Within this context, it would be unrealistic, if not naive, to expect HPWIs to take seriously the grievances of student groups, largely black, and capitulate to their demands for antiracist policy interventions (such as those noted above) in anything but isolated cases, and even then, only in a tepid, fleeting, fragmented, and disingenuous manner. White capitulation in the face of demands for racial justice has not figured prominently in U.S. history; violence, threats, recrimination, retaliation, and intransigence have been more common responses (Feagin 2012, 2014; Omi and Winant 2015). Still, even in this racial climate, HPWIs must admit some black students in order to sustain their fictional, postracial campuses. Diversifying White College Campuses In the mid-1960s, small private colleges began actively recruiting students of color. This resulted in modest but growing numbers of black and Latino students on their campuses (Duffy and Goldberg 1998). However, a resurgent political conservatism in the late 1970s dampened this enthusiasm, slowing and in some cases reversing these gains. This regressive pattern halted in the mid-1980s as colleges resumed their effort to recruit students of color (Duffy and Goldberg 1998; Rubin 2014). Today, to the extent allowable by law, many HPWIs, especially the most selective ones, continue to develop their incoming classes in a race-conscious way (Rubin 2014). These institutions remain committed, to some extent, to achieving and/or maintaining a racially diverse student population. I offer three sources of evidence in this regard. First, consider the number of higher education associations and HPWIs, collectively representing more than 2,000 institutions of higher education, that filed amicus curie briefs in support of the University of Texas’s race-conscious admissions policy in the Texas v. Fisher cases.2 Although each of these institutions may have had different motivations, in formalizing their support in this manner 460 they signaled a desire to maintain their ability to use race as a factor when evaluating students of color for admission. Second, although what HPWIs mean by “diversity” can at times be unclear and racially problematic (see Embrick 2011), a casual exploration of most of their Web sites will reveal ample evidence of these institutions’ stated commitments to diversity. HPWIs’ desire to project some measure of racial diversity was documented empirically by Pippert, Essenburg, and Matchett (2013) in their analysis of college viewbooks. They concluded that the consistency at which institutions of higher education presented misleading depictions of racial diversity leads us to the understanding that it is intentional and near universal. It is clear that racial diversity is being used as a commodity in the marketing of higher education and presenting an image of diversity is more important than accurately portraying the student body. (P. 275) This inaccuracy is largely a consequence of the visual overrepresentation of black student enrollment. This study revealed that the representation of black students in college marketing materials was, on average, 104 percent greater than the actual representation of black students on campus. Pippert et al. pointed out that HPWIs typically explain the overrepresentation of black students as aspirational and/or a signal that they are welcoming. The authors also noted the seeming conflation of black students with diversity, as Hispanic and Native students are underrepresented in HPWI viewbooks. This study clearly highlights HPWIs desire for not simply diverse student bodies en masse but specifically the bodies of those racialized as black. Third, a 2014 report in the Chronicle of Higher Education titled “Shaping the Class: How College Enrollment Leaders View the State of Admissions and Their Profession” shows that at hundreds of HPWIs, there is a serious effort to enroll students of color (recall that this often means black students). Indeed, 86 percent of enrollment leaders at private, not-for-profit institutions indicated a goal of enrolling “diverse students.” At public institutions, the number was 95 percent. Moreover, at both private and public institutions, enrollment leaders reported that this goal was either met or exceeded 90 percent of the time. It appears that most HPWIs profess to value, seem to desire, and make at least some effort at increasing or maintaining the number of black students and, to a lesser Sociology of Race and Ethnicity 5(4) extent, non-black students of color, on their respective campuses. However, HPWIs’ “commitment” to diversity appears to be undermined by the theoretical scholarship suggesting that at least some of these institutions screen black students in a way that could be disadvantageous for those whose lived experience and academic and cocurricular interests attest to the continuing significance of race and white racism (Carbado and Gulati 2013). When coupled with the research documenting whites’ color-blind convictions (Bonilla-Silva 2018; Brown et al. 2003) and recent student protests on college campuses across the United States (see Dickey 2016), some white admissions counselors could be inclined, if not encouraged, to screen out prospective students perceived as too concerned with race and racism. There is ample reason (historical, theoretical, and empirical) to suspect that a nontrivial number of college administrators and the white staff members under their charge may be angry and/or frustrated with student protestors who have the perceived temerity to challenge white authority and white racism on campus. Prospective students and applicants of color may be picked through, sorted, measured, and weighed in an extra-academic manner by admissions counselors to identify those deemed racially palatable or sufficiently deracialized. I contend that for many HPWIs, it is these racially palatable, deracialized, or nondescript prospects and applicants of color who are perceived as the crème de la crème, or what might be termed desirable diversity. Making more granular distinctions between and among prospective students of color, to ensure they are the right “fit” for the institution, could increasingly be one of their responses. College Admissions and the Question of “Fit” Admissions evaluations at many HPWIs emphasize applicant “fit.” According to the “2014 State of College Admissions,” published by the National Association for College Admission Counseling, private nonprofit colleges and universities and smaller institutions are more likely to evaluate applicants holistically than their public and larger counterparts (Clinedinst 2015). Specifically, “private colleges assigned greater importance than public colleges to . . . essay/writing sample, interview, counselor and teacher recommendations, extracurricular activities, portfolios, state graduation scores, and demonstrated interest” (p. 29). Collectively, these factors provide students with an 461 Thornhill opportunity to demonstrate to admissions counselors that they would be a good “fit” at their institution. One recent study found that selective colleges consider institutional fit second only to academic merit as an admissions factor (Rubin 2014). Consistent with Bowen and Bok’s (1998) seminal book, The Shape of the River, on race-conscious admissions policies, Rubin (2014) found that admissions offices at many selective institutions continue to use four criteria to sort applicants into pools, “including students who are non-white U.S. citizens, students who can pay full tuition, legacy applicants, and students who have a talent that is seen as being desirable, or needed, by the institution” (p. 11). Once applicants are sorted in this manner, they are compared against others within their respective pools. “The majority of colleges [initially] differentiate between students by their level of academic merit. But, after that, it is about class-building and selecting individuals who will further the college’s mission and enrich its campus [emphasis added]” (p. 14). Although Rubin’s study provides a revealing portrait of the often hidden process of race-conscious admissions, it does not indicate whether colleges further subdivide students of color by race and/or ethnicity (e.g., black, Asian, Latino) once they are sorted into the “nonwhite U.S. citizens” pool. This is an important omission because it has implications for how white admissions counselors could be engaging in intraracial (i.e., within-group) screening practices. There are at least four reasons to suspect that admissions counselors do subdivide applicants of color. First, by separating students of color from white students and placing them in a “nonwhite U.S. citizens” pool, admissions counselors demonstrate that they are instructed, practiced, and primed to formally evaluate applicants interracially. Second, “white” and “nonwhite” is a descriptively inadequate and insufficiently complex expression of racial stratification in the United States; it does not accurately reflect most individuals’ mental schemas vis-à-vis race. That is, Americans more commonly make distinctions such as black, white, Asian, and Latino/a. Third, admissions counselors, approximately 80 percent of whom are white (Phair 2014), are not exempted from the data that show most whites subscribe to at least some negative racial stereotypes about and discriminate against members of specific racial or ethnic groups, not about or against a collective nonwhite group (Chou and Feagin 2015; Feagin and Cobas 2014; Houts Picca and Feagin 2007). Fourth, colleges could have a stated or unstated goal or desire to increase the number of students from a specific racial or ethnic group, thereby encouraging them to differentiate between and among students of color. For these four reasons, and on the basis of the scholarship reviewed above, it seems likely that admissions counselors do not perceive applicants of color as a racially undifferentiated amalgam but that they are instead attuned to their “differences” and evaluate them both interracially (e.g., blacks vs. Asians vs. Latino/as) and intraracially (e.g., blacks vs. other blacks). Research Design The data for this study come from an audit of the e-mail practices of 517 white college admissions counselors employed at the same number of institutions. Specifically, I use the correspondence approach to observe how responsive white admissions counselors are to inquiry e-mails from black high school students who evince different degrees of racial salience. The inquiry stage was selected over the application stage for two reasons: (1) logistical practicality and (2) the ethics of applying for scarce seats in an incoming class. Still, given the experimental nature of the correspondence method, if white admissions counselors respond to different “types” of black prospective students at significantly different rates, then we would observe the type of intraracial discrimination theorized by Carbado and Gulati (2013). My study proceeded in three stages. First, I developed four narrative templates that represented different degrees of racial salience or racial “types” that would serve as the body of the e-mails used in the audits. Second, I constructed and tested a list of racialized names and then created corresponding e-mail addresses. Third, I developed a sample of admissions counselors who would receive e-mails from the fictitious black students. I elaborate on these three procedures in the following paragraphs. Construction of Narratives The narratives used in the audit are presented in Table 1. Each narrative contains blank spaces for personalization; an indication that the student learned of the institution from a friend, teacher, or guidance counselor; extracurricular involvement (including leadership experience); and a question asking whether the student would be a good fit for the institution on the basis of his or her interests and involvements. Narratives 1 and 2 were constructed to represent black prospective students who were 462 Sociology of Race and Ethnicity 5(4) Table 1. E-mail Templates Used in Audit. Narrative 1 Dear (admissions counselor name), I have recently begun my college search and my high school guidance counselor recommended that I look at (college). I am from the (city) area and will be a senior next year. I have always enjoyed and excelled in English and Math. Therefore, I am considering pursuing these majors in college. In addition to my coursework, I serve as the Assistant Editor of my high school’s newspaper and am a member of the marching band. Moreover, I started volunteering at my community library as a math tutor, where I primarily assist middle school students with geometry and algebra. As you can see, I am highly involved in a variety of activities that help me to further explore my academic interests. I just wanted to know whether you think someone with my interests would fit well at (college)? (student name) Narrative 2 Dear (admissions counselor name), My name is (student name) and I am in the process of looking at colleges. One of my teachers suggested that I consider (college). My academic interests mainly include the natural sciences such as biology and environmental science. As a junior, I have been involved in many extracurricular activities. In addition to serving as the President of Students for Environmental Awareness, I am one of three students on my high school’s greenhouse planning committee. These two activities have strengthened my commitment to new sustainability efforts. Additionally, I have recently joined local community efforts in (city) to “Take Back The Tap” and hope to develop a similar campaign at my high school before I graduate next year. I hope to continue my involvement in issues related to the environment in college. Do you believe (college) would be a good fit for me given the types of extracurricular activities that I’ve been involved in? (student name) Narrative 3 Dear (admissions counselor name), My name is (student name) and I recently learned about (college) from a friend while I was researching colleges. I’m writing to you to find out whether someone with my interests would fit well at (college). This year, I have dedicated more time for activities outside my schoolwork and recently started a student organization devoted to educating students at my high school about African American culture and history. Though the group’s focus is on the African American experience it also emphasizes cross-cultural understanding. I am excited to continue my involvement in this organization as a senior next year. Also, every Sunday and Thursday, I play the keyboard and piano in my school’s Jazz Band and my church’s Gospel Choir. Although I participate in a wide range of activities that require a great deal of my time, I hope to stay just as involved in college. Do you believe (college) would be a good fit for me? (student name) Narrative 4 Dear (admissions counselor name), I recently learned about (college) from my guidance counselor. I am writing to find out whether you believe your school would be a good fit for someone with my interests. I am from (city) and I have a strong passion for issues of racial justice and African American culture and politics. These interests have led to my involvement in several extracurricular activities including the Black Student Organization and the Anti-Racism Student Alliance at my school that I co-founded this past fall. I am also the youngest member of a group in my community that is planning to host a workshop on the topic of race and politics at a local community center next year. One of the goals is to empower community members by educating them about topics and issues that affect them every day but that they might not know much about. Some of the topics that the workshop will cover include white privilege, affirmative action, colorblind racism, racial microaggressions, and institutional racism (e.g. in the education system and the criminal justice system). I am really excited to be part of such an important event and hope to continue my involvement after I graduate. Who knows, if the event is a success and I learn enough, I might be able to develop a similar workshop for the students at the college that I end up attending. (student name) 463 Thornhill not racially salient (i.e., they revealed no indication that they might have interests or involvements visà-vis race and racism). As such, an admissions counselor might reasonably perceive that race was not central to the student’s identity. The nondescript narrative expresses academic interests in English and math, extracurricular involvements in the high school’s newspaper and marching band, and volunteer service outside of school. The environmental sustainability narrative expresses academic interests in biology and environmental science, extracurricular involvement in a school greenhouse planning committee, and community activism around the issue of sustainability. Narratives 3 and 4 were intended to represent black prospective students who were racially salient (i.e., they betrayed an interest in race-based activities). The racial unity narrative reveals race consciousness but evinces no direct interest in racial politics or concern about racial inequality. In Carbado and Gulati’s (2013) formulation, it “express[es] a ‘soft’ commitment to colorblindness” (p. 124). This narrative includes extracurricular involvements in a jazz band and a student club dedicated to African American history and culture, participation in a church gospel choir, and no explicitly stated academic interests. The antiracist narrative unequivocally acknowledges the continued significance of race and white racism for blacks’ life chances. It also reveals direct involvement in community-based, antiracist education. Although these four narratives contain very similar features, they are not identical templates. Instead, they are meant to convey different representations of “blackness” that white admissions counselors could encounter in their communications with, or evaluations of, black prospective students or applicants. Furthermore, the four narratives, or racial “types,” can also be viewed as existing along a continuum of racial salience. The nondescript and environmental sustainability narratives suggest the author of the e-mail is likely not a racially salient black person and is therefore most likely to be perceived as racially palatable (i.e., sufficiently deracialized so as not to discomfort most whites). The racial unity narrative is moderately racially salient. Although the author of the e-mail is race conscious, which violates the color-blind imperative, the statement suggesting the importance of “cross-cultural understanding” serves to moderate their racial salience, thus increasing the student’s racial palatability. The antiracist narrative is the most racially salient, and therefore its author is likely to be perceived as the least racially palatable. Additionally, I selected four cities with different proportions of black residents (Detroit, Atlanta, Chicago, and Seattle) and randomly assigned one of them to three of the four narratives prior to sending them to admissions counselors. This was done for two reasons. First, as admissions counselors tend to have regional assignments, this was done to increase the likelihood that the fictitious students would be communicating with their assigned counselors. When two or more counselors were assigned to the same region, I randomly selected one. When regional assignments were not provided, I randomly selected a counselor listed on the Web site. Second, I suspected that the racialized nature of these four cities could interact with students’ degree of racial salience such that it might affect counselors’ levels of responsiveness. The decision to target counselors by region proved logistically beneficial. However, I did not detect a significant region or region-racial salience interaction effect on counselors’ responsiveness. Identities of Prospective Students Next, I developed a pool of racialized names that would signal to admissions counselors that the author of the e-mail was black. To generate first names (traditionally female and male) I drew on several sources, including Social Security Administration data, several states’ departments of public health, names used in previous audit studies, and several we thought would be read as black by white counselors. Last names were derived from the U.S. census, which reports the most popular last names by race. I took the two separate lists of names and randomly assigned them to each other to produce a single list of first and last names. Next, I created five different surveys with 16 names on each that asked the respondents to select the race/ ethnicity that best represented each name. These surveys were distributed to approximately 900 predominantly white, middle- to upper-middle-class college students attending a highly selective liberal arts college in the Midwest. The eight names used in the study (four female and four male) represent those where 90.8 percent or more of the survey respondents indicated they were “black” names (see Table 2). The average for these eight names was 94.8 percent (94.6 percent for female names and 95.1 percent for male names). Only 2.9 percent of respondents thought these names were “white.” These names were strong cues to admissions counselors that the authors of the e-mails were black. 464 Sociology of Race and Ethnicity 5(4) Table 2. First and Last Names Used in Study. Black Male Names Jamal Jackson Tyrone Reid Quashawn Williams Keshawn Grant Total Perception Perception Black White Perception Perception All Other Black Female Perception Perception All Other Races Names Black White Races 97.3 1.3 1.4 95.1 3.5 1.4 94.2 1.4 4.2 93.6 5.0 1.4 95.1 2.8 2.1 Sampling Admissions Counselors and Sending Inquiry E-mails The sample of white admissions counselors was derived in the following manner. First, I used the U.S. News & World Report 2013 Best Colleges Rankings to compile a list of non-research-intensive, private, four-year HPWIs in the United States that had academic programs mentioned in the narratives or related to interests in the narratives (i.e., majors in the arts and sciences and education). Next, I developed a random sample of admissions counselors, one per institution. This was accomplished by using the profile pictures of the admissions counselors listed on the colleges’ Web sites. If the randomly selected counselor did not have a profile picture, I used a Google search to locate a picture of the counselor through alternative Web sites such as Facebook and LinkedIn. Once the list of counselors was complete, I used an interrater reliability analysis (with the support of an undergraduate research assistant) to classify admissions counselors as white, black, or neither white nor black. There was a 100 percent correspondence between our racial ascriptions of the white admissions counselors. The findings reported in this paper are based on the sample of white admissions counselors. Finally, to further militate against detection, each admissions counselor was sent only two of the four narratives, with an average lag time of one month between the first and second e-mails. Although admissions counselors’ work is cyclical in nature, on the basis of the academic calendar, which could potentially affect their overall responsiveness to inquiry e-mails, there is no reason to suspect that counselors engage in racial screening practices more or less intensively at different points during the admissions cycle. There were 97 cases in which I received a response from an admissions Lakisha Lewis Shaquana Coleman Tanisha Richardson Destiny Brown Total 97.2 0 2.8 96.4 1.4 2.1 93.9 3.4 2.7 90.8 7.1 2.1 94.6 3.0 2.4 counselor (or student admissions assistant) who was not the intended recipient of one of the e-mails. This scenario had three versions: (1) the original counselor no longer worked in the admissions office, (2) a different counselor indicated that he or she was forwarded the e-mail because he or she worked with students from the city mentioned in the e-mail, or (3) we received the e-mail from a different counselor (or student admissions assistant) with no explanation (this was most likely because of reason 1 or 2). These trials were excluded from the study to maintain the uniformity and rigor of the audit. Although it is possible that a nonresponse could have been due to the first two scenarios, there is no reason to believe that this would have occurred at significantly different rates across narratives. Furthermore, some counselors could have forgotten to respond to an e-mail. This, however, is something that could also be expected to occur across narrative types. The findings reported below are based on the 517 remaining audits. Given the research design, I am unable to speak directly to institutional policies and practices at specific HPWIs. Whether the patterns I discovered are formally directed or tacitly condoned is an important question. However, it is a separate question and one this study cannot answer. Still, irrespective of institutional intentionality, my findings have consequences for black high school students with college aspirations as well as implications for how white admissions counselors communicate with them. Findings I divide my findings into two parts. First, I consider the data in aggregate (i.e., 1,034 e-mails). I then focus on the 517 audits that represent six matchedpairs subaudits. Table 3 shows that white admissions counselors responded to non–racially salient 465 Thornhill Table 3. Response Rates by Racial Salience. Non–Racially Salient 65.50 [317] Racially Salient Percentage Difference Overall χ2 p Value 55.27 [304] 10.23 60.06 [621] .001 Note: χ2 = Σ(fo – fe)2/fe. Table 4. Response Rates by Counselor Gender, Student Gender, and Narrative. Counselor Gender Student Environmental Gender Nondescript Sustainability Racial Unity Antiracist Male Male Female Female Male Female Overall 62.50 [65] 57.78 [26] 66.10 [39] 64.44 [87] 59.38 [38] 69.01 [49] 63.60 [152] 64.89 [61] 56.86 [29] 74.42 [32] 68.87 [104] 61.25 [49] 77.46 [55] 67.35 [165] 63.16 [72] 63.08 [41] 63.27 [31] 62.11 [100] 59.74 [46] 64.29 [54] 62.55 [172] 40.57 [43] 43.33 [26] 36.96 [17] 52.66 [89] 45.45 [35] 58.70 [54] 48.00 [132] Overall χ2 p Value 57.66 [241] 55.20 [122] 60.41 [119] 61.69 [380] 56.38 [168] 66.67 [212] 60.06 [621] .001 .16 .01 .02 .08 .17 .001 Note: χ2 = Σ(fo – fe)2/fe. black students’ e-mails (nondescript and environmental sustainability narratives) 65 percent of the time, whereas racially salient black students (racial unity and antiracist narratives) garnered a response rate of only 55 percent. This 10 percentage point difference is highly statistically significant (p < .001) and provides initial support for Hypothesis 1. Examining the response rates for the four narratives individually (see Table 4) shows the much lower response rate for the antiracist narrative relative to the nondescript, environmental sustainability, and racial unity narratives. On average, black students whose inquiry e-mails revealed an antiracist ethic experienced a response rate that was 17 percentage points lower than black students presenting one of the other three narratives (p < .001). Furthermore, the differences in the response rates between and among the nondescript, environmental sustainability, and racial unity narratives are not statistically significant. That the racial unity narrative is statistically indistinct from the two non– racially salient narratives is an important finding that will be discussed later. At this point, these findings provide qualified support for Hypothesis 1, which states that non–racially salient black students will receive a higher level of responsiveness from white college admissions counselors than will racially salient black students. The qualification is necessary because the results shown in Table 4 indicate that the low level of responsiveness to the antiracist narrative is largely responsible for depressing the racially salient response rate illustrated in Table 3. Therefore, the data offer support for Hypothesis 2, which states that racially salient black students whose interests and extracurricular involvements challenge the validity of color-blind ideology (i.e., the antiracist narrative) will experience the lowest level of responsiveness from white college admissions counselors. Next, I assess whether and the extent to which gender moderates these findings. The effect of counselor gender on the likelihood of responding to each of the four narratives is also reported in Table 4. Both male and female counselors reveal a statistically significant preference for the nondescript, environmental sustainability, and racial unity narratives over the antiracist narrative. This pattern is stronger among male counselors, as they were nearly 25 percent less likely than female counselors to respond to e-mails containing the antiracist narrative. As it concerns student gender, black female students were significantly more likely than black male students to receive responses to their inquiries irrespective of counselor gender and narrative type (64.3 percent vs. 55.9 percent, p < .01, on the basis of a separate analysis). Focusing on the interaction of counselor and student gender by narrative type reveals several notable findings. First, white female counselors are considerably more responsive than white male counselors to black female students who present the antiracist narrative (a 22 percentage point difference). Second, although female counselors’ response rates for male and female students continue to demonstrate a depreciation of those students presenting the antiracist narrative, the differences across narratives are 466 not statistically significant in the case of both black male (p = .08) and black female (p = .17) students. Third, in the case of male counselors, although we observe the durable pattern vis-à-vis the depreciation of the anti-racist narrative, there is no statistically significant difference in the response rate of white male counselors to black male students by narrative (p = .16), though this is perhaps due in part to their already relatively low response rates. However, the differences in white male counselors’ response rates to the four narratives for black female students are statistically significant (p < .01). Fourth, and particularly noteworthy, is the 37.5 percentage point advantage in male counselors’ response rate for black female students interested in environmental sustainability over those interested in antiracism. This is the largest difference I documented, with the actual response rate for black women presenting the antiracist narrative (36.9 percent) barely approaching it. I derived the foregoing findings by analyzing counselors’ response rates using the entire data set in aggregate. An additional and more refined way to examine the data is to use McNemar’s test to analyze matched pairings. Recall that each counselor received two different e-mails separated by an average of one month. When paired in this manner, the data can be disaggregated and evaluated as six separate audits. McNemar’s test, the results of which are derived through an analysis of discordant pairs, allows me to answer the question of whether counselors, when presented with e-mail correspondence from two different black students representing two different racial “types,” treat these students equally or whether their behavior betrays a preference for one over the other. The results of this analysis are presented in Table 5. In all six audits the most frequent outcome was one of equal treatment (i.e., responding to or ignoring both e-mails). Nevertheless, I observed strong evidence of intraracial discrimination. In two of the three audits that included the antiracist narrative (antiracist vs. nondescript and antiracist vs. racial unity), counselors revealed a statistically significant preference for the alternative narrative. Audit 1 shows that counselors who received the nondescript and antiracist narrative were nearly ten times more likely to favor black students who presented the nondescript narrative than they were to favor those who submitted the antiracist narrative (p < .001). The third audit (antiracist vs. racial unity) highlights counselors’ preference, by a ratio of nearly three to one (p < .01), for those racially salient black students who express an apolitical, “‘soft’ commitment to colorblindness” (Carbado and Gulati 2013:124) over their explicitly Sociology of Race and Ethnicity 5(4) race-political counterparts who revealed an antiracist praxis. In the second audit (antiracist vs. environmental sustainability), counselors revealed a preference for the environmental sustainability narrative over the antiracist narrative by a ratio of nearly two to one, though it did not meet the conventional threshold for statistical significance. This is likely due to the gendered nature of counselors’ response to female and male students presenting the environmental sustainability narrative as well as sample size considerations. The three remaining audits (audits 4–6) also revealed no statistically significant differences in counselors’ response rates. Discussion My findings indicate a clear pattern whereby white admissions counselors are more likely to ignore black high school students’ inquiry e-mails if they betray an acknowledgment of the continuing significance of white racism. Although this does not necessarily mean that white admissions counselors will treat these students’ applications unfavorably should they decide to apply, my findings suggest that they may be treated unfairly if they evince a rejection of color-blind ideology. If admissions counselors cannot bring themselves to provide these students with the courtesy of a response to their e-mail inquiries, whether genuine and substantive or curt and boilerplate, why should we expect these counselors to become responsive, helpful, courteous, and unbiased should these inquirers become applicants? I contend that these findings are likely a conservative estimate of the costs borne by more racially salient black students navigating the college admissions process. The fact that moderately racially salient black students whose e-mails included the phrase “crosscultural understanding” did not experience less responsiveness from white admissions counselors is an important finding. Overall, these black students received e-mail responses at roughly the same rate as black students who betrayed no interest in racialized involvements. Hence, it was not simply race consciousness among black students that depressed the level of responsiveness among white admissions counselors. Rather, it was critical race consciousness, which acknowledges the continuing significance of race and white racism for blacks’ life chances, that penalized some black students. One limitation of my study is that the three narratives with the highest response rates indicated that the prospective student learned of the school from a teacher or guidance counselor. It is possible that the differences in the response rates between 467 Thornhill Table 5. Distribution of Audit Responses. Audit 1: antiracist vs. nondescript Equal treatment “Equal” treatment (both ignored) Antiracist narrative favored  Nondescript narrative favored Audit 2: antiracist vs. environmental Equal treatment “Equal” treatment (both ignored) Antiracist narrative favored Environmental narrative favored Audit 3: antiracist vs. racial unity Equal treatment “Equal” treatment (both ignored) Antiracist narrative favored Racial unity narrative favored Audit 4: racial unity vs. nondescript Equal treatment “Equal” treatment (both ignored) Racial unity narrative favored  Nondescript narrative favored Audit 5: racial unity vs. environmental Equal treatment “Equal” treatment (both ignored) Racial unity narrative favored Environmental narrative favored Audit 6: environmental vs. nondescript Equal treatment “Equal” treatment (both ignored) Environmental narrative favored  Nondescript narrative favored Total Percentage 30 21 2 19 72 41.67 29.17 2.78 26.38 100% 36 26 11 21 94 38.30 27.66 11.70 22.34 100% 41 24 12 32 109 37.61 22.02 11.01 29.36 100% 36 23 12 20 91 39.56 25.27 13.19 21.98 100% 42 11 9 13 75 56.00 14.67 12.00 17.33 100% 38 14 15 9 76 50.00 18.42 19.74 11.84 100% Percentage Difference Ratio 23.62*** 9.5 10.64 1.9 18.35** 2.7 8.79 1.7 5.33 1.4 7.90 1.7 Note: McNemar’s χ2 = (b – c)2/b + c. **p < .01. ***p < .001. these three narratives and the antiracist narrative are, at least in part, attributable to the latter being initiated from a friend as opposed to a teacher or guidance counselor. Perhaps admissions counselors are more responsive to e-mail inquiries that result from teacher or guidance counselor recommendations. Yet an argument could also be made that admissions counselors would take seriously the recommendation from a friend because perhaps that friend attends the institution, was recently accepted to it, or has a family member who attended it. Unfortunately, there is no way for me to assess whether the source of prospective students’ referral contributed to the response rate for the four different narrative types, as the three sources of referral did not vary randomly; they were connected to the narrative templates. Although this could have been an interesting variable to consider, these referrals 468 were used solely to make them more appealing and thereby increase the likelihood they would generate a response. Still, if a teacher or guidance counselor recommendation were sufficient to produce an e-mail response, the three non-antiracist narratives could be expected to have an average response rate well above 65 percent. In sum, the lack of variation in the source of referral within narratives would be highly unlikely to vitiate the central findings of the study, namely, that white admissions counselors racially screen black students. Relatedly, although statistically equal response rates across narrative types would be highly suggestive in the direction of equal treatment, this is not guaranteed. Future research should investigate the substance of what counselors communicate to students who evince different degrees of racial salience. It could be the case that some “types” of black students receive enthusiastic responses, offers of application fee waivers, information about scholarships, and/or invitations to meet with a counselor in person or visit campus, while others are more likely to receive curt or boilerplate responses that neither offer nor mention any of the above. Another potential intervening variable that could affect counselors’ responsiveness to different “types” of black students is enrollment. Many HPWIs are struggling financially, largely because of declining enrollment or an inability to grow enrollment. It could be argued that responsiveness would be higher among those schools experiencing greater financial and enrollment pressures. To investigate this issue in Part, I assessed whether intraracial screening varied as a function of enrollment. My findings revealed no statistically significant relationship between intraracial screening practices and enrollment, neither in the overall sample nor within the four narrative types. Some readers may note that some white counselors did respond to the most racially salient black students. That is, not all white admissions counselors screen black students in a way that disadvantages those who are most racially salient. My response to this position is threefold. First, though all whites benefit to varying degrees from their privileged position in the racial hierarchy, not all actively engage in new racism practices that sustain this system (i.e., some whites are antiracist). Second, while ignoring the inquiry e-mails of racial justice-oriented black prospective students is clear evidence of a disregard for or depreciation of their collegiate aspirations, responding to their e-mails does not necessarily indicate the opposite. It takes little time and effort for an admissions counselor to Sociology of Race and Ethnicity 5(4) respond to an e-mail. And the fact of a response is not the same as an enthusiastic and supportive response. Third, claims that most admissions counselors operate in a race-neutral and meritocratic manner requires an absence of empirical evidence to the contrary. In contrast, claims that some white admissions counselors operate in a racially biased manner only requires documented evidence that they do exactly that, not evidence of absolute racial exclusion, total racial domination, or the absence of exceptions. Conclusions My study provides perhaps the first empirical test of Carbado and Gulati’s (2013) theory of intraracial discrimination. The findings contained herein provide convincing evidence in support of this theory. White admissions counselors were, on average, 26 percent less likely to respond to the inquiry e-mails of black high school students who evinced a concern about the continuing problem of white racism. White male counselors were 37 percent less likely to respond to these students. And when black female students sent these e-mails, white male counselors were almost 50 percent less likely to respond. These findings are disturbing, though not entirely surprising. One of the primary factors that likely contributes to this type of discrimination is the autonomy and discretion afforded to admissions counselors. There may be an institutional expectation that admissions counselors respond to all inquiries from prospective students, perhaps even allowing for some delay because of counselors’ recruitmentrelated travel schedules. However, for many counselors, this is not their practice. Approximately 40 percent of the 1,034 inquiry e-mails we sent to white admissions counselors went unanswered, and for certain subgroups of black students that percentage jumped to nearly two thirds. Furthermore, given the prevailing color-blind ideology and HPWIs’ rhetoric of diversity and inclusion, most white admissions counselors would not admit to intraracially screening black prospective students, often ignoring the e-mails of those committed to antiracism. My data show that many do. Indeed, my findings challenge the frequent claims of private HPWIs that they are or aspire to become sites of diversity, equity, and inclusion. Instead, some HPWIs employ admissions counselors who screen out black students who are arguably best equipped to identify and challenge raciall