EXPLORING THE IMPACT OF APPLICANTS ’ GENDER AND RELIGION ON PRINCIPALS ’ SCREENING DECISIONS FOR ASSISTANT PRINCIPAL APPLICANTS

In this experimental study, a national random sample of high school principals (stratified by gender) were asked to evaluate hypothetical applicants whose resumes varied by religion (Jewish, Catholic, nondenominational) and gender (male, female) for employment as assistant principals. Results reveal that male principals rate all applicants higher than female principals and that the gender and religion of applicants failed to negatively or positively affect principals’ evaluations. These results suggest that discrimination based on an applicant’s gender and religion failed to be manifested during the pre-interview stage of the selection process. The paper concludes with a theoretical discussion of the distinction between explicit and implicit prejudice, and encourages future researchers to investigate the potential impact of prejudice on employment selection decisions and to consider whether schools should promote diversity in leadership positions.

Bon, S. C. (2009).Exploring the impact of applicants' gender and religion on principals' screening decisions for assistant principal applicants.International Journal of Education Policy and Leadership 4 (1).Retrieved [DATE] from http://www.ijepl.org.Society has become extremely diverse, particularly in terms of race, culture, ethnicity, disability, language, religion, and sexual orientation.While these changes in composition are noticeable in many of the public spaces of society, it is the classrooms, hallways, and playgrounds of U.S. schools where the magnitude of our increasingly diverse society has been acutely felt.To adequately meet the educational needs of diverse learners, educators must be able to adopt differentiated models of instruction, create inclusive learning environments, alter assessment practices, and-perhaps most important-commit to making a difference in the lives of children.Furthermore, these efforts are mandated by state and federal policies aimed at providing educational opportunities for all learners.
In light of legal mandates, the complexity of student needs, and an increasingly diverse student body, it is no surprise that the educational community has embraced the importance of teacher quality and teacher diversity.These goals for quality and diversity are, or at least should be, similarly desired for educational administrators.Thus, the purpose of this study is to explore the impact of diversity, specifically the gender and religion of an assistant principal applicant, during the employment selection process.
The selection of gender as a variable is in response to lingering concerns regarding gender equity for female school administrators (Riehl & Byrd, 1997); religion was included because it has long been a contentious issue in schools and because "religious freedom has historically been considered a fundamental right in American society" (Fasman, 1978, p. 358).Although an interaction effect was not necessarily anticipated between gender and religion, researchers in women' s studies have suggested that there may be an interaction between gender and ethnicity (Hyde, 2007).Specifically, Hyde encourages additional research to explore "whether 'well-established' gender phenomena are similar or different across ethnic groups" (p.262).If gender phenomena differ across ethnic groups, it may also be worth exploring whether or not gender differences will appear across religious groups.Finally, the following discussion further explains the reasons for the selection of these two variables.

Gender Equity
Throughout the 1980s and 1990s, researchers expressed concerns about the status of females in school leadership positions.Whitaker and Lane (1990) noted that " [w]omen are not moving rapidly up the career ladder into administration of public schools" (p.8).Early researchers sought to explain the disproportionate representation of females in school administration by focusing on gender bias.For example, Shakeshaft (1987) documented unfavorable attitudes toward female administrators and Yeakey, Johnston, and Adkison (1986) examined the negative impact of gender-role stereotypes and concluded that this was one of several factors that limited the entry of females into administrative positions.
In an experimental study of the selection process, Reis, Young, and Jury (1999) hypothesized that gender discrimination negatively affected prospective female applicants for the assistant principal position.However, the results of the study revealed the opposite: that female applicants received significantly higher evaluations than male applicants.Reis et al. cautiously concluded that "there may be an intentional effort to increase the number of females in all administrative positions" as a result of growing legal awareness about calls for gender equity (p.241).Mertz and McNeely (1994) drew a similar conclusion following their review of the status and condition of women in administrative positions in education.Specifically, they suggested that Title IX "called public attention to the situation" and led to increased scrutiny and to an increase in the percentage of female administrators in urban school districts (p.367).Nonetheless, Mertz and McNeely speculated that because urban school districts serve "larger numbers of poor and minority students" (p.369), these administrative positions are less appealing to male administrators and so were readily available for the influx of female administrators.Riehl and Byrd (1997) investigated "whether women are treated inequitably with regard to administrative hiring" (p.45).They found that gender had no impact on the selection process, which led them to encourage females to acquire part-time administrative experience in order to increase their preparation for and likelihood of entering administrative positions.Riehl and Byrd also highlighted the importance of including gender equity goals as a part of the administrative selection process in light of the disproportionate number of females in administrative positions compared to teaching positions.
Recent efforts to document the status of females in administrative positions reveal that there is "a growing proportion of women in the ranks of public school administrators" (Andrews & Ridenour, 2006, p. 36) and that public attitudes toward women as leaders are changing.Andrews and Ridenour suggest that these two realities may positively affect the number of women in leadership positions.These observations seem consistent with the most recent public school principal survey, which reports the percentage of male and female principals for 2003-04 as 50.3 percent and 49.7 percent, respectively (NCES, 2006).Despite the appearance of gender equity suggested by the fairly equal proportions of male and female principals, the percentage of women in leadership positions (49.7 percent) is relatively low when compared to the percentage of women in teaching positions (79 percent).
Although early studies explored the status of females in school administration by focusing on gender bias, that is, these studies sought to determine whether or not females are discriminated against during the employment selection process.Recent studies have shifted the focus to gender equity.The definition of gender equity, however, is unclear and demands closer examination.Specifically, is gender equity achieved when males and females are equally represented in leadership positions?Alternatively, is equity achieved when "women become administrators in the same proportion as they are teachers" (Riehl & Byrd, 1997, p. 61)?In light of this latter interpretation of gender equity, it may be premature to assume that gender equity has truly been achieved among public school administrators.Thus, it is important to continue to examine the impact of gender on decisions made during the employment selection process.
Collectively, past efforts to examine the impact of gender bias or discrimination during the employment selection process have provided valuable insight regarding gender equity in school leadership positions.In addition, the increasing number of female principals provides further evidence that gender equity concerns may no longer be legitimate.Despite this evidence, it is premature to abandon research efforts to examine gender bias or discrimination as long as the proportion of females who are teachers is significantly greater than the proportion of females who are administrators.Finally, if, in fact, attitudes toward females in leadership positions are changing and schools are intentionally selecting females as school leaders, this study may add value by demonstrating whether gender positively affects female applicants for an assistant principal position.

Religious Diversity
Religious beliefs and traditions are a hallmark of the freedoms embraced in the United States.Because the First Amendment of the U.S. Constitution states that "Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof," freedom of religion is clearly a protected right.The importance of tolerance for and respect of religious freedoms may be extremely relevant today, in part, as a result of increasing religious diversity in the United States.In addition, ongoing conflicts in the Middle East and significant tensions between religious and ethnic groups around the world underscore the prominence of religious freedoms.Blumenfeld (2007) identifies Christian privilege as a widely occurring phenomenon in U.S. public schools and society, and he asserts that it negatively affects the religious freedoms of marginalized faith groups and individuals who are nonbelievers.Blumenfeld claims that schools have institutionalized the norms of Christian privilege, thus promoting the marginalization of non-Christians, nonbelievers, and even "minority Christian denominations, for example, African American and Latino/a churches, Amish, Mennonites, Quakers, Seventh-Day Adventists, Jehovah' s Witnesses, adherents to Christian Science and to the Church of Jesus Christ of Latter Day Saints, and in some quarters, to Catholics" (p.196).He suggests several ways to recognize diversity, promote freedoms, and overcome marginalization, including revising curriculum, recruiting staff from a variety of backgrounds in order to provide positive role models, and providing training for school personnel that focuses on religious diversity and equality of treatment.
In light of Blumenfeld' s suggestion to promote religious diversity among public school faculty and staff, it may be helpful to explore whether or not religion influences, negatively or positively, pre-interview decisions about a prospective employee.Furthermore, the extent to which religious affiliation may or may not influence an individual' s opportunity for employment has not been fully explored in empirical studies (Sheridan, 2006).Thus, the present study seeks to fill a void in existing research by including religious affiliation as a variable within the context of a study about employment selection decisions.
The employment selection discussion that follows provides a cursory overview of past research on the impact of bias and discrimination during the selection stage of the employment process, revealing that efforts have focused on gender, race, age, and disability status but have failed to examine religious discrimination and bias.Given the paucity of research about religious discrimination and contradictory reports about gender discrimination, as well as the focus of many research efforts on teacher selection rather than administrator selection, this study will expand the current literature by examining the possible impact of gender and religion on an applicant who is seeking an assistant principal position in a public high school.
Employment Selection Process Young and Oto (2004) have characterized educational personnel selection research studies as policy-capturing studies that focus on the compliance of public school administrators with numerous federal laws: the Age Discrimination in Employment Act (ADEA) of 1967 and Title VII of the Civil Rights Act of 1996 (Young & Fox, 2002), Title VII of the Civil Rights Act of 1964 (Reis, Young, & Jury, 1999), and the Americans with Disabilities Act of 1990 (Young & Prince, 1999).These and other state and federal laws, such as Title IX and the Equal Educational Opportunity Act, govern the employment selection process by protecting individuals' rights and restricting discriminatory practices by employers.In each of the research studies, a specific characteristic of the hypothetical applicant was manipulated, holding all other information constant, in an attempt to isolate the influence, negative or positive, of legally protected applicant characteristics.The various employment protections will be discussed further in the legal framework section of this study.Young and Fox (2002) examined the impact of age and national origin on the evaluator' s pre-interview review of applicants for teacher and assistant principal positions.On the basis of evaluations of materials during this screening stage, they concluded that age discrimination and national origin negatively affected the evaluation.Reis et al. (1999) examined gender of the evaluator and of the applicant in an effort to determine whether there was an interaction effect, relying on the similarityattraction paradigm (Graves & Powell, 1995) to support their hypothesis that female principals would favor female applicants for an assistant principal position on the basis of gender.Although the results failed to support an interaction effect, female applicants were rated significantly higher than male applicants.Finally, Young and Prince' s (1999) study examined the effect of a hypothetical teacher applicant' s physical disability status on principals' pre-interview evaluation; no significant impact was found.
The selection process contains three distinct points at which an applicant has an opportunity to make an impression on a potential employer.These have been characterized as the pre-interview stage, interview stage, and postinterview stage (Dipboye, 1992;Macan & Dipboye, 1990).Initial impressions are made at the preinterview stage, making this stage critical for applicants.If they are not viewed favorably at this point, they are unlikely to gain an opportunity to interview; if they do not progress to the interview stage, it is unlikely that they will be offered employment.
This study focuses on decisions made by an evaluator about an applicant' s qualifications for an assistant principal position during the pre-interview stage of the selection process.As indicated earlier, researchers have examined numerous factors for possible influence on decisions made by evaluators during this period (e.g., Young & Allison, 1982;Reis, Young & Jury, 1999;Young & Prince,1999;Young & Fox, 2002).In each of these studies, key legal principles were examined and discussed in an attempt to reveal the numerous protections that prohibit discrimination against individuals during the employment selection process.

Legal Framework
The impact of discrimination and bias on decisions made during the employment process has been the focus of numerous federal statutes and regulations.These statutes and regulations provide legal protections for individuals who seek redress in the courts in response to the employers' discriminatory conduct.For purposes of this study, key federal legislation is examined to clarify the existence of basic protections that must be afforded to individuals who qualify on the basis of certain unalienable rights.Because the focus of this study is on gender and religious discrimination during the employment selection process, the following discussion provides an overview of key federal legislation related to these two types of discrimination.

Gender Discrimination
The Civil Rights Act of 1964 prohibits discrimination in hiring, firing, and other terms and conditions of employ-ment.Under Title VII, an employer may not discriminate against an individual with respect to "compensation, terms, conditions, or privileges of employment, because of such individual' s race, color, religion, sex, or national origin" (Title VII of the Civil Rights Act of 1964).Lee (2005) refers to Title VII as "the federal court mainstay in the struggle for equal employment opportunity for the female workers of America" (p.678).In a similar fashion, Pope (2006) indicates that Title VII has been "the biggest contributor to the nation' s labor docket" (p.925) for the past 40 years.
Sex discrimination is also prohibited under Title IX: No person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance (Title IX of the Education Amendments of 1972).
Litigation under Title IX has focused predominantly on the rights of students to be free from discrimination based on sex.Litigation under Title IX has focused predominantly on the rights of students to be free from discrimination based on sex, although there is a perception that it has had an impact on discrimination against females seeking school leadership positions.However, the vast majority of cases and subsequent legal analyses reveal that Title IX is infrequently invoked during litigation of sex discrimination complaints in employment.Despite the perceived impact of Title IX on sex discrimination against females seeking leadership positions in schools, the vast majority of cases and subsequent legal analysis reveal that Title IX is infrequently invoked when litigating complaints of sex discrimination in employment.Instead, the emphasis has been on Title VII.
The following section provides an overview of Title VII as it relates to religious discrimination.Included is a discussion of recent data reported by the U.S. Equal Employment Opportunity Commission (EEOC), which was created under Title VII and empowered to investigate and prevent unlawful employment practices.Specifically, the EEOC "indicates that employers may not treat employees or applicants less-or more-favorably because of their religious beliefs or practices" (U.S. Equal Employment Opportunity Commission, n.d., Introduction).

Religious Discrimination
As noted earlier, the possible influence of religious discrimination on decisions made during the employment selection process has not been examined.Fasman (1978) examined the legislative history of Title VII and concluded that little attention was given to religious discrimination issues because the focus was on racial discrimination.In 1972, Congress amended Title VII and added an affirmative duty to "reasonably accommodate" an employee unless it imposes an "undue hardship" on the employer (section 701 [j]).Much of the case law on religious discrimination that has followed the passage and subsequent amendments to Title VII has focused on issues related to "reasonable accommodation" (Fasman; see e.g.Heller v. EBB Auto Co., 1993).
Although there has been a steady increase in religious discrimination charges over the past 10 years, overall, only 3 percent of EEOC charges involved "claims of religious discrimination, while sex discrimination was claimed in 30 percent" of EEOC charges (Pope, 2006, p.925).According to the EEOC, "2,541 charges of religious discrimination were made in 2006," and 2,387 of those charges were resolved (U.S. Equal Employment Opportunity Commission, 2007).However, these statistics do not include specific details revealing whether the religious discrimination charges arose out of pre-or postemployment practices.
Based on the relatively few charges reported by the EEOC, researchers may have concluded that religious discrimination is simply not prevalent enough to merit further examination.Although there may be some legitimacy to these possible conclusions, given the importance of religious beliefs and traditions in the United States and growing concerns about the marginalization of non-Christians, nonbelievers, and minority Christian denominations (Blumenfeld, 2007), the possible impact of religious discrimination should merit further examination.
An exhaustive literature review of social science and legal databases confirmed suspicions about the lack of empirical research on religious discrimination.Specifically, the literature review revealed only one article in which the author examined court cases; the cases focused primarily on religious freedoms of teachers (Beezer, 1982).Beezer' s extensive review covered cases that addressed the impact of teacher employment policies in areas including (a) leave of absence for religious observance, (b) religious beliefs and nature of job, and (c) physical appearance and attire.These areas were addressed because of their prevalence and possible impact on employee rights.Since Beezer' s review, law review articles have been the primary source of information regarding recent developments in religious discrimination jurisprudence in employment, and empirical studies have generally overlooked religious discrimination issues (Sheridan, 2006).Sheridan concludes that "investigations of the dynamics of religious discrimination require our immediate attention" (p.335).Sheridan (2006) explored the incidence in the United Kingdom of religious discrimination against Muslims as a result of "Islamaphobia."She identifies an important issue, which influenced a major decision to not include Islamic religions in the present study: that it is difficult to distinguish between religious and ethnic discrimination because many individuals are unaware of the distinctions between Muslim and Islam (p.320).Furthermore, Sheridan indicates that most people are unable to clearly separate an individual' s religious identity (e.g., Islamic religion) from an individual' s ethnic identity (e.g., Middle Eastern or Arabic).Thus, it would be extremely difficult to isolate the occurrence of religious rather than ethnic discrimination using an experimental design, which was the research method selected for this study.
In light of the blurred distinction between religion and ethnicity, discrimination on the basis of Islamic religion was not explored in this study.The focus of this study, then, is religions (Catholic, Jewish) that are not unique to an identifiable and visible ethnic group and that have in the past been identified as sources of religious bias.Specifically, as highlighted in the discussion below, concerns about the existence of anti-Catholic bias (see Blumenfeld, 2007;McCarraher, 2003;and Perl & Bendyna, 2002) and anti-Semitism (see Dasgupta, 2004;Djupe & Sokhey, 2003;and Leets, 2002) continue to be expressed in the literature.

Anti-Catholicism.
Recent research (Perl & Bendyna, 2002) has examined the extent to which adult Catholics perceived anti-Catholic bias and how their perceptions were related to political party identification.Perl and Bendyna argue that Catholicism is an important denomination for study because of historical and present hostilities against Catholics.According to Perl and Bendyna, Catholics have long been considered a minority religion.They claim that "recent concerns about anti-Catholic attitudes have arisen in a political context" (p.654).For example, during the 2000 Republican primary campaign, President George W. Bush appeared at Bob Jones University amid deep criticism about established school policies and remarks made by the school' s leadership.At this time, the school had a policy that banned interracial dating, and the school leaders had criticized Catholicism and described it as a "Satanic cult" (Kornblut, 2000, A10).
Numerous editorial reviews of several recent books about anti-Catholicism conclude "that fundamentalistevangelical Protestantism, not 'secular humanism,' has been and remains the largest, most toxic reservoir of anti-Catholicism in contemporary America" (McCarraher, 2003, p. 40).Although Catholicism is viewed as a minority religion, Catholics make up "a quarter of the general population in the United States," and are regularly subjected to anti-Catholicism in the media according to Woodward (as cited in Massa, 2004).

Anti-Semitism.
Although anti-Semitic acts are reportedly decreasing in present U.S. society, the Jewish community continues to view anti-Semitism as a defining issue (Djupe & Sokhey, 2003).In part, this concern about anti-Semitism is legitimized by the long history of violence toward Jews and Judaism in Europe (Sutcliffe, 2004).The Jewish community represents about two percent of the U.S. population; thus, being Jewish in the United States qualifies an individual as a member of a minority group.
The potential for bias against someone who is Jewish may possibly be greater in light of the general tendency by individuals to prefer groups that are socially advantaged and valued by mainstream society (Dasgupta, 2004), and generally, "Jews are recognized as an outgroup" in society (Leets, 2002, p. 343).Mellott and Schwartz (1999) examined the disadvantages experienced by those who are considered to be outgroup members of society, and they found that the distinction between ingroup and outgroup membership was likely to affect both implicit and explicit bias.
According to a 2004 report by the Federal Bureau of Investigation (FBI), bias disproportionately leads to violent acts and hate crimes against minority groups, particularly Jews, African Americans, and homosexuals.Nearly 68 percent of 1,586 total religious bias offenses were anti-Jewish.The remaining sources of religious bias were 13 percent anti-Islamic, 4 percent anti-Catholic, 3 percent anti-Protestant, less than 1 percent anti-atheism or agnosticism; approximately 13 percent of cases were against other unspecified religions and miscellaneous antireligion groups.Such evidence of religious bias suggests that further investigation of discrimination may be warranted, particularly in light of the potential rights at risk if perceived religious affiliation negatively influences employment selection decisions.

Theoretical Framework
Various studies have examined the employee selection process in an attempt to determine the degree of compliance by public school administrators with specific federal laws.However, the studies have not considered whether both gender and religion have influenced decisions made by public school administrators during the screening stage of the selection process.The potential impact of bias on hiring decisions has been examined using several different theoretical frameworks to explain how screening decisions are made (Young & Oto, 2004).
In the present study, the difference between implicit and explicit prejudice theories is examined as a possible explanation for screening decisions made by potential employers during the selection process.
Prejudice is believed to be an explicit orientation that "is routinized to the point where it is automatically accessed in the presence of objects" (Rudman, 2004, p. 133).Yet changes to laws and social values have led to a decreased likelihood that individuals will deliberately reveal their prejudices.Thus, researchers have focused on the distinction between implicit and explicit forms of prejudice, and several studies have revealed that the difference may be identifiable through specific models of testing.According to Rudman, Greenwald, Mellott, and Schwartz (1999), implicit prejudice is "conceptually distinct from explicit prejudice" (p.438) because it can function unconsciously.Although few people would readily admit that they are biased, "psychological research routinely exposes counterintentional, unconscious biases" (Banaji, Bazerman, and Chugh, 2003, p. 1).
Both implicit (unconscious) and explicit (conscious) forms of prejudice may affect decision making at the screening stage of the selection process.Banaji and Greenwald (1995) focused on distinguishing between implicit and explicit prejudice, which eventually led to the development of the Implicit Association Test (IAT).To reveal explicit prejudice, participants were asked to respond on a Likert scale to statements about gender equality such that a high overall score represented sexism and a low score represented acceptance of gender equality.The implicit measure of gender bias was assessed by calculating the number of times participants associated male names, correctly or incorrectly, with fame, rather than female names.Banaji and Greenwald explain that when participants associate male names more readily with fame than female names, this choice reveals an unconscious bias in favor of males and against females.Finally, their failure to detect a correlation between the explicit responses about gender equality and the implicit responses favoring males over females led Banaji and Greenwald to conclude that explicit gender stereotypes (sexism) were unrelated to implicit gender bias.
Implicit measures of bias are believed to be more valid estimates than explicit measures because it is difficult to control implicit biases, which are thought to be automatic (Rudman, 2004).Furthermore, implicit biases "can (unintentionally) guide our thoughts and actions" (pp. 137-38).In light of the potential negative impact of both implicit and explicit forms of prejudice, Rudman encourages policymakers to seek ways to override these biases, such as the adoption of affirmative action policies.
Rudman ( 2004) also reports that implicit stereotypes have resulted in discrimination against female job applicants seeking managerial positions.She seeks to raise awareness about the overall negative impact of biases on "evaluative (e.g., hiring) decisions" (p.138) and calls upon society to address this problem.One solution she proposes is to increase the representation of women in leadership roles, which she claims can lead to the dismantling of gender hegemony (Rudman & Kilianski, 2000).
The examination of two potential sources of bias, gender and religion, may provide useful guidance to policymakers as they consider what types of affirmative action policies may be necessary in order to overcome potential biases.In the present study, the focus is primarily on the selection process, employing an experimental design to explore the impact of gender and religion on the evaluation of and likelihood of extending an interview invitation to a hypothetical assistant principal applicant.Although the theoretical distinction between explicit and implicit prejudice is included and offered as a lens through which the selection process and employment decisions may be better understood, the IAT was not used in the study.

Method Participants
Public high school principals throughout the 48 contiguous United States constituted the population for this investigation.From this population, a list of 400 participant names was generated, using a purely random stratified selection process by Market Data Retrieval in 2004.The sample was stratified by gender, such that 200 of the participants were male and 200 were female.The sample size was determined according to an analysis of statistical power suggested by Cohen (1977) and was based on anticipated return rates as reported in past studies.A total of 274 administrators responded, representing a 64 percent response rate, but, for a variety of reasons, not all of the responses were usable.
First, participants who did not respond to all questions were eliminated from the respondent pool.Second, in order to achieve a balanced data set as suggested by Zar (1985), 10 male principals and 10 female principals were randomly selected from the respondent group for each experimental condition (total participants = 120 respondents).Although this further reduced the number of usable responses, a balanced data set (that is, having all cells of the same sample size) is a legitimate way to ensure independence of statistical interactions and is recommended when using the Analysis of Variance (ANOVA) statistic.
The average age of participants was 51 years old.Sixty male respondents and 60 female respondents were included.Participants reported unanimously that they had previous experience as teachers (on the average 18 years of teaching experience); and on average, the participants reported that they had 13 years of administrative experience (see Table 1).

Materials
Each principal was mailed a set of credentials representing a hypothetical applicant for an assistant high school principalship.The credentials included (a) a letter introducing the research study, which did not indicate the nature of the manipulations; (b) a resume for the hypothetical applicant; (c) a reference letter for the applicant; (d) a job description for the position; (e) an applicant evaluation form; (f) a demographic questionnaire for the reviewing principal to self-report; and (g) a stamped, preaddressed envelope for returning the applicant evaluation form and demographic questionnaire.Participants were asked to review the applicant' s credentials for an assistant principalship position in his or her building.
A job description for an assistant principalship was provided to each participant in an attempt to maintain continuity across all experimental conditions.Participants overwhelmingly reported that they used paper credentials during the screening stage of the selection process.This high percentage provides a form of validity for the use of hypothetical applicant materials to simulate the screening stage of the selection process.

Procedures
Participants were randomly selected on the basis of gender and then were randomly assigned to one of six experimental conditions.To avoid introduction of a systematic bias and to meet statistical assumptions, male and female principals were randomly assigned in equal proportions to one of six information packets.The six packets reflected all possible combinations of the independent variables, applicant gender, and applicant religion.

Independent Variables.
Three independent variables were examined in this study: gender of the evaluator, religion of the applicant, and gender of the applicant.Gender of the evaluator, and religion (Jewish, Catholic, nondenominational) and gender of the hypothetical applicants were each varied, resulting in a completely crossed 2x3x2 factorial design.The information packets sent to all of the subjects contained the operationalization of each of the independent variables.
The names of the hypothetical applicant varied slightly to reveal gender.Thus, the surname for all applicants was Gold, but the male applicants were named Robert Gold and the female applicants were named Rebecca Gold.The appropriate pronouns were also used throughout the reference letters; that is, when discussing Rebecca' s skills, the letter referred to "her" ability and, for Robert, "his" ability.
Religious affiliation was indicated in two places in the hypothetical applicant' s packet of information: (1) on the resume, by including "Newman Catholic Community Center," "Hillel Jewish Community Center," or "Nondenominational Faith Community Center" under the section heading "Community Activities" and (2) in the reference letter, by including a comment about the applicant' s leadership roles at one of those same three centers.

Dependent Variables.
Participants were requested to evaluate the applicants as if they were screening for an assistant principalship within their building and to rate the applicants on a total of six separate but related criteria.The six criteria were compiled to yield a candidate evaluation form for use in screening applicants for the position of assistant principal (Reis, Young, & Jury, 1999).The evaluation form comprised two dependent variables.One criterion was a Likert-type scale ranging from 1 through 10 to indicate the probability of being granted an interview, with higher ratings indicating greater probability.The reliability of this variable has been assessed in several studies, according to Young & Oto (2004).In those studies, the lower bound estimate for reliability has been reported to be consistently in the mid-.80s(see Miller-Smith, 2001;Young & Fox, 2002).

Data Analysis
The number of participants (n=120) falls within the established range as determined by the power analysis with an alpha level of .05, a medium effect size, and a power of .80 (see Young & Oto, 2004;Young & Fox, 2002;and Miller-Smith, 2001).Table 1 reveals the descriptive and experiential data for the participating high school administrators.As was expected given past studies in this area (e.g., Young & Oto, 2004), 97 percent of the administrators reported that they use paper credentials to screen applicants during the employment selection process.This adds a substantial degree of authenticity to the experimental design of this study.
Null hypotheses were first tested by performing an analysis of variance (ANOVA) independently for each of the dependent variables.An alpha level of 0.05 was established as the decisional criterion for rejecting the hypotheses.The participants' responses on the singleitem and multiple-item applicant assessment scales were submitted for analysis to determine possible main effects and interaction effects.Results of the ANOVA are reported in Table 3.
As evidenced by the results reported in Table 3, the gender and religion of assistant principal applicants did not result in any significant differences between the rat-ings by the evaluators.Thus, further testing to establish whether or not gender or religion accounted for any amount of variance among the evaluators' ratings simply was not necessary.The failure to identify a significant difference between the applicants' evaluations on both of the dependent variables reveals that neither gender nor religion influenced the assessments of these hypothetical assistant principal applicants by male or female evaluators.
Table 3 reveals, however, a significant difference between the mean ratings of male and female principals who were asked to rate the hypothetical applicants.Specifically, female evaluators rated all applicants significantly lower than male evaluators.Possible explanations for the significant difference between male and female ratings of the hypothetical assistant principal applicants are provided in the following section.

Discussion
The Condition of Education report reveals that in 2000-01, approximately 21 percent of teachers were male and 79 percent were female (NCES, 2006).According to the public school principal survey, the percentages of all male and female principals are reported for 2003-04 as 50.(NCES, 1994b) females held 35 percent of public school principal positions.The trend appears to be an increasing presence of female principals in public schools.Nonetheless, given that individuals typically are teachers before becoming principals and 79 percent of teachers are females, the disproportionate representation of female principals still does not make sense.Riehl and Byrd (1997), however, have offered an alternative explanation for the disproportionate number of females entering administrative positions.Rather than examining the potential impact of gender bias, they identified several variables that significantly affected the likelihood that female teachers would enter school leadership positions.They explored the extent to which gender and related factors influenced teachers' decisions to seek school leadership positions rather than examining the potential impact of gender bias by an employer.The identified variables included the family context (being married and having small children), educational qualifications (having an administration degree), and "socialization factors-having aspirations, qualifications, and experience" (p.60).Thus, the disproportionate representation of female principals may also be explained by factors influencing female teachers' decisions to avoid entering administration rather than solely by the influence of bias operating to exclude female applicants from administrative positions.
The variables identified by Riehl and Byrd are insightful but still neglect the issue of gender equity.If, in fact, public schools are seeking to promote gender equity-that is, to hire a percentage of female administrators proportionate to the percentage of female teachers-one may have expected to see higher evaluations for females, or at least an increased likelihood of offering an interview to female applicants for the assistant principal position.Such a result would have been consistent with the study by Reis et al. (1999), in which female applicants for the assistant principal position did receive higher evaluations than male applicants during the selection process.It is startling that less than 10 years ago female administrators received significantly higher ratings than males during this stage of the selection process, yet the present study reveals that the differences between the evaluations of male and female applicants are insignificant.
Although religious discrimination continues to be a factor affecting employment conditions, such as leaves of absence, free speech issues, and work attire (Beezer, 1982), the present study indicated that religious affiliation did not appear to influence decisions made during the pre-interview stage of the selection process.Despite the relatively few religious discrimination charges reported to the EEOC (U.S. Equal Employment Opportunity Commission, 2007) and the limited number of hate crimes reported by the FBI (2004), there has been limited empirical research on religious discrimination (Sheridan, 2006) This study hoped to begin a dialogue to address this void in the research literature.
Examining the potential for religious bias to negatively affect employment decisions is particularly impor-tant in light of concerns raised by Blumenfeld (2007) about the marginalization of non-Christians, nonbelievers, and minority Christian denominations in public schools.Although religious bias did not influence the pre-interview evaluations of hypothetical applicants, school leaders must continue to be vigilant.Finally, the percentages of participants (N=113) who reported their religion as Jewish (2 percent) and Catholic (31 percent) were far below the percentage of participants who reported their religion as Protestant (56 percent).This may indicate lack of religious diversity in public educational organizations, consistent with concerns raised by Blumenfeld.If promoting diversity among teaches and leaders is a goal in public school settings, one might also have expected to see a higher rating for Jewish or Catholic individuals seeking to attain an assistant principal position.
Guided by theoretical discussions about implicit stereotyping by Banaji and Greenwald (1995), this study used procedures that involved public school administrators who had to make evaluative judgments "that do not explicitly draw attention to their own conscious beliefs about gender [or religion]" (p.195) during the screening process of hypothetical applicants.The gender and religious affiliation of each applicant were noticeable in the materials (confirmed by pilot-testing of the instruments and materials) but were not explicitly disclosed to the administrators.If applicant evaluations had differed significantly according to gender and religious affiliation, such results may have been attributed to the implicit (unconscious) rather than explicit (conscious) prejudices of the evaluators.As indicated earlier, however, neither gender nor religious bias was detected.
There are several possible explanations for the failure to detect implicit prejudices related to religious affiliation and gender.One explanation relates specifically to limitations of the research methods and instruments used in the study.Instead of using the Implicit Association Test (IAT), the study relied on methods and materials that have been consistently used in past investigations of discrimination during the employment selection process (see Young & Oto, 2004;Young & Fox, 2002;Reis et al., 1999;and Young & McMurry, 1986).Although the reliability of this study' s applicant evaluation materials has been assessed in several studies (Young & Oto, 2004), the materials may not be appropriate for detecting implicit measures of attitude.Since implicit bias exists on an unconscious level that is more likely to be "captured by indirect measures" (Dasgupta et al, 2000), the procedures used in the present study may have been insuffi-cient to capture implicit or hidden biases.
Another possible explanation may be that gender and religious biases are waning in present society.Rudman (2004) and Greenwald et al. (1998) suggest that explicit bias has become increasingly unpopular in modern society to the extent that individuals consciously suppress their biases and prejudices.Individuals may also engage in a self-monitoring process so that they can overcome their biases or at least avoid revealing socially unacceptable beliefs, particularly when they are reporting their attitudes about ethnicity (Devine & Elliot, 1995).
In light of this trend in society, much attention has been focused on identifying whether implicit bias, in fact, exists separately from explicit bias.As indicated earlier, numerous efforts have been made to create a measure to isolate implicit bias (e.g., using specific measures, such as the IAT (Greenwald, McGhee, & Schwartz, 1998).According to Banaji, Bazerman, and Chugh (2003), implicit prejudice occurs because people "judge according to unconscious stereotypes and attitudes" (p.2).Yet several researchers in this area have reported that even when an individual maintains such prejudicial or biased beliefs, this does not automatically lead to discriminatory conduct; which may provide further explanation for the insignificant results obtained in the present study.
In the study, the only significant difference in evaluation occurred between female and male administrators; the female administrators rated all applicants significantly lower than did the male administrators.This phenomenon may not be surprising given that females who occupy traditionally male-dominated leadership positions often face gender role conflicts.According to Eagley, Makhijani, and Klonsky (1992), females in male-dominated roles are somewhat restricted as leaders.One such restriction may occur because females need to overcome gender stereotypes, such as beliefs that good leaders have masculine traits (Heilman, Martell, & Simon, 1988).
Female leaders may strive to overcome gender stereotypes by distinctly defining themselves in a culture that continues to adhere to "male-defined conceptions of leadership" (Mertz, 2006).For example, female leaders may seek to define their unique leadership abilities by responding conscientiously and discriminately to the numerous tasks that come across their desks, including the review of materials for an assistant principal applicant.The female principals may simply have approached the hypothetical applicant evaluation task of this study with careful attention to details, which then may have led to the lower evaluation of all assistant principal applicants.

Conclusion
The study' s results reveal no significant differences between the evaluations of applicants on the basis of gender or religious affiliation.In fact, the evaluations of all applicants by the participating school principals were surprisingly similar on both of the dependent variables.Given the limitations of the selected research design and the use of relatively brief hypothetical applicant materials, it may simply be that evaluators failed to notice the gender or the religious affiliation of the applicant.Alternatively, it may be that the evaluators failed to perceive gender and religion as sources of influence on future job performance.That is, the future performance of these applicants may have been judged based on their individual merits listed in the application materials and not on the influence of gender or religious biases of the evaluators.
If bias and stereotypes are not operating to exclude female applicants from leadership positions in schools, there may be other factors that negatively affect the transition from teaching to administration for females.Loder and Spillane (2005) address the impact of role conflict and role discontinuity in their qualitative study of female administrators.Based on interviews with 16 female administrators, Loder and Spillane report that the transition from teaching to leadership led to role conflict, in part, because the relationships with students and other teachers changed dramatically.It may also be that some teachers are reluctant to leave the classroom, so they fail to consider the principalship.The risk with relying solely on the role conflict theory, though, is that it places the burden for explaining the disproportionate number of female administrators compared to female teachers primarily on the female teachers.Thus, there must be ongoing dialogue from multiple perspectives and with a variety of theoretical models to reveal the internal and external barriers regarding females entering school leadership positions.
The examination and theoretical discussions about measures of explicit and implicit prejudices have raised numerous questions about the manifestation of prejudice in discriminatory behavior.Specifically, Brendl, Markman, and Messner (2001) suggest that prejudice and bias do not automatically lead to discrimination, despite existing assumptions that prejudice causes dis-crimination.These assumptions merit further examination, according to Brendl et al., because the link between implicit prejudice and discriminatory behavior has simply not been established.
Although prospective job applicants should still proceed with caution when they list certain information on their resumes, they also should enjoy some degree of relief based on the results of the present study.Unfortunately, the study' s results also fail to reveal that gender or religion positively affects employment opportunities.As suggested earlier, one may have expected to see higher evaluations for female or for Jewish or Catholic applicants, and maybe even an increased likelihood by principals to consider offering an interview to these applicants for the assistant principal position if there is a commitment to increase diversity in leadership positions.
The commitment to diversity in U.S. public schools, whether in schoolwide leadership positions or in classrooms, is less clear today than ever.Although data for this study were collected prior to the decision in Parents Involved in Community Schools v. Seattle School District (2007), the potential impact of this U.S. Supreme Court case on diversity as a compelling interest in public schools has yet to be realized.In this five-to-four decision, the U.S. Supreme Court declared that race-based student assignment plans used by two urban school districts are unconstitutional.The Court had recognized diversity as a compelling interest (Grutter v. Bollinger, 2003), but only in a higher-education context and only when diversity is broadly assessed.
Given the legal uncertainty fueled by the decision in Parents Involved in Community Schools v. Seattle School District (Parents), school leaders should proceed cautiously with goals to promote diversity in leadership positions.Although the Court specifically ruled against the use of student assignment plans that relied on race to promote diverse school settings, the legitimacy of school district plans to promote diversity may be suspect, in general, following this decision in Parents.Even though schools are faced with an increasingly diverse student body with complex learning and social needs, the Supreme Court has sent a chilling message to schools across the nation regarding the promotion of diversity.Thus, public schools that desire to increase diversity among teachers and leaders are encouraged to seek sound legal guidance and to consider the parameters established in Grutter v. Bollinger, particularly relating to the broad assessment of diversity.
Future research is also necessary in order to assess the potential for implicit bias to appear in discriminatory treatment of applicants during various stages of the selection process.Researchers could administer the IAT or a similar instrument to employment decision makers and then examine whether hiring decisions appear to be related to the IAT results (bias or nonbias) on a variety of demographic factors.Bias, whether implicit or explicit, could negatively influence decisions made during the selection process and have negative consequences on schools' efforts to recruit and hire educational leaders from a variety of diverse backgrounds.Thus, further examination of the differences between implicit and explicit bias should be conducted.

C CE EN NT TR RA AL L H HI IG GH H S SC CH HO OO OL L MIDDLEVILLE, STATE January 27, 2005
To Whom It May Concern: Please accept this letter in support of Robert Gold for an assistant principal position.Robert has been an excellent teacher in this school system for the past several years and he is well prepared for an assistant principal position.In this school system, Robert has been responsible for teaching three physical education classes and two health education classes.
Robert has consistently demonstrated good discipline and classroom management skills while providing numerous activities to enrich the instructional program in his classes.He has a solid rapport with students and clearly demonstrates sound educational practices.Robert' s interactions with staff members and with parents also demonstrate his genuine warmth and respect for others.
The varied interesting and creative activities that Robert utilizes in his classes are evidence of his excellent preparation at State University.Also, Robert has contributed to the community through his city recreation work and his various leadership roles at the Hillel Jewish Community Center.Robert Gold will be particularly successful in an assistant principal position because of his effective skills in planning and organizing.Robert has contributed greatly in his role as chairperson of the principal' s advisory council for the past two academic years at Central High School.I recommend him without hesitation.Please accept this letter in support of Rebecca Gold for an assistant principal position.Rebecca has been an excellent teacher in this school system for the past several years and she is well prepared for an assistant principal position.In this school system, Rebecca has been responsible for teaching three physical education classes and two health education classes.
Rebecca has consistently demonstrated good discipline and classroom management skills while providing numerous activities to enrich the instructional program in her classes.She has a solid rapport with students and clearly demonstrates sound educational practices.Rebecca' s interactions with staff members and with parents also demonstrate her genuine warmth and respect for others.
The varied interesting and creative activities that Rebecca utilizes in her classes are evidence of her excellent preparation at State University.Also, Rebecca has contributed to the community through her city recreation work and her various leadership roles at the Newman Catholic Community Center.Rebecca Gold will be particularly successful in an assistant principal position because of her effective skills in planning and organizing.Rebecca has contributed greatly in her role as chairperson of the principal' s advisory council for the past two academic years at Central High School.I recommend her without hesitation.Assist principal in overall administration of the school.

2.
Serve as principal in the absence of the regular principal.

3.
Propose schedules of classes and extracurricular activities.

4.
Work with department heads and faculty in compiling annual budget requests.

5.
Supervise the reporting and monitoring of student attendance.

6.
Assist in maintaining discipline throughout the student body.

7.
Perform such record-keeping functions as principal may direct.

8.
Perform other tasks and assume other responsibilities as the principal may assign.
UC CA AT TI IO ON NA AL L B BA AC CK KG GR RO OU UN ND D Master' s Degree: M.A. -State University Educational Administration Bachelor' s Degree: B.S. -State University Education P PR RO OF FE ES SS SI IO ON NA AL L E EM MP PL LO OY YM ME EN NT T Physical Education and Health Teacher Central High School Plain City, State C CE ER RT TI IF FI IC CA AT TI IO ON NS S Secondary Principal, State K-12 Physical Education Teacher, State Health Education Teacher, State E EX XT TR RA AC CU UR RR RI IC CU UL LA AR R E EX XP PE ER RI IE EN NC CE E Chairperson, Principal' s Advisory Council Member, Principal' s Advisory Committee to study grading system Advisor, Junior class Varsity Track Coach Junior Varsity Track Coach C CO OM MM MU UN NI IT TY Y A AC CT TI IV VI IT TI IE ES S Local Education Association Local Parks & Recreation Association Newman Catholic Community Center /fc C CE EN NT TR RA AL L H HI IG GH H S SC CH HO OO OL L B D DE ES SC CR RI IP PT TI IO ON N T TI IT TL LE E Assistant Principal Q QU UA AL LI IF FI IC CA AT TI IO ON NS S Secondary Principal Certificate, State Teacher Certificate, State R RE EP PO OR RT TS S T TO O Building Principal P PE ER RF FO OR RM MA AN NC CE E R RE ES SP PO ON NS SI IB BI IL LI IT TI IE ES S 1.

Table 1 .
Characteristics of Participating Public School Administrators

Table 2 .
Gender and Religion of Participating Public School Administrators a Some participants failed to indicate religious affiliation.

Table 3 .
Analysis of Variance Table T TE ER RM MS S O OF F E EM MP PL LO OY YM ME EN NT TSalary and work year will be established by the School Board.E EV VA AL LU UA AT TI IO ON NPerformance of this job will be evaluated in accordance with the School Board' s policy on Evaluation of Professional Personnel.