ACTIVATION, AUTOMATICITY, AND MENTAL HEALTH IMPLICATIONS OF COLONIAL MENTALITY

ACTIVATION, AUTOMATICITY, AND MENTAL HEALTH IMPLICATIONS OF COLONIAL MENTALITY

BY ERIC JOHN DAVID B.A., University of Alaska Anchorage, 2002 M.A., University of Illinois at Urbana-Champaign, 2004

DISSERTATION Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Psychology in the Graduate College of the University of Illinois at Urbana-Champaign, 2007

Urbana, Illinois

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

UMI N um ber: 3269875

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

®

UMI UMI Microform 3269875 Copyright 2007 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

C ertificate

of

C om m ittee

A pproval

University o f Illinois at Urbana-Champaign Graduate College April 12, 2007 We hereby recommend that the thesis by: ERIC JOHN DAVID Entitled: ACTIVATION, AUTOMATICITY, AND MENTAL HEALTH IMPLICATIONS OF COLONIAL MENTALITY

Be accepted in partial fulfillm ent o f the requirements fo r the degree of: Doctor of Philosophy

Signatures: 1-----------

»

\S

J

------------T --------------------

Director o f Research

Head o f Department

Committee on Final Examination1

Chairperson

■ommittee Memqer

Committee Member

Committee Member

Committee Member

Committee Member

* Required for doctoral degree but not for master’s degree

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

ABSTRACT Among Filipino Americans, Colonial Mentality (CM) is a specific form of internalized oppression that is characterized by an automatic preference for anything American and automatic rejection of anything Filipino that may be manifested overtly and covertly. Given that CM is theorized to be composed of a covert aspect and that it may operate automatically, introspectiondependent and explicit methods of investigating the construct may be limited. Thus, using indirect and implicit methodologies, three studies were conducted to test CM theory: Study 1 attempted to activate and capture the existence of a CM-consistent cultural knowledge system using explicit primes; Study 2 investigated whether CM may be activated outside of awareness, intention, or control using a more indirect and implicit priming methodology - a lexical decision priming task; and Study 3 tested the theory that Filipino Americans have automatically associated pleasantness with anything American and unpleasantness with anything Filipino using the Implicit Association Test. The results suggest that many Filipino Americans may hold a CMconsistent cultural knowledge system, that CM may be automatically activated, and that CM may operate outside of awareness, intention, or control. Mental health implications of CM among Filipino Americans are also explored.

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

To my Nanay, Tatay, Ate Ellen, Bonz, Margaret, Jass, and Alex

iv

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

ACKNOWLEDGEMENTS Thanks to my nanay for continuing to inspire me with the words “edukasyon lamang ang mai-papamana ko sa iyo.” Appreciation also goes to my tatay for supporting me and loving me in his own unique way. Salamat to Bunsoy, Ate Ellen, Jass, and Alex for always being in my heart and for always keeping me in theirs. Special appreciation and love are extended to Margaret Hoffman for all her love, support, and affection. Extreme gratitude goes to the Vinas family for being so generous and welcoming; I don't know how I would have done it without your family. I would also like to thank the entire Rochon, Ebue, Danner, and Hoffman families for all the help, assistance, support, and encouragements. Special thanks to my mentor, Sumie Okazaki, who gambled with a naive Filipino kid from Alaska and to whom I am forever indebted. Others who I would like to thank are: my lab mates in the Culture & Emotions Lab at the University of Illinois for tolerating my thoughts; my homeboys in Alaska and the Philippines for stimulating the researcher in me; the Clinical-Community Psychology Division for allowing me to explore issues that others do not think are important; Howard Berenbaum, Ying-yi Hong, CY Chiu, and Martin Manalansan for their suggestions; Thara Gagni for her assistance in data collection; the Philippine Student Association at the University of Illinois for their help in advertising the project; the National Institute of Mental Health through the American Psychological Association Minority Fellowship Program for providing me with financial support to conduct these studies (MH15742-26); and last but definitely not the least, the hundreds of Filipino Americans who participated in this project for sharing their experiences and feelings with me. “Maraming-maraming salamat po sa inyong lahat.

v

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

TABLE OF CONTENTS INTRODUCTION........................................................................................................................... 1 STUDY 1A METHOD..................................................................................................................30 STUDY 1A RESULTS................................................................................................................. 36 STUDY IB METHOD................................................................................................................. 37 STUDY IB RESULTS................................................................................................................. 39 STUDY 1 DISCUSSION............................................................................................................. 40 STUDY 2...................................................................................................................................... 42 STUDY 2 METHOD.................................................................................................................... 44 STUDY 2 RESULTS.................................................................................................................... 49 STUDY 2 DISCUSSION.............................................................................................................. 54 STUDY 3...................................................................................................................................... 56 STUDY 3 METHOD.....................................................................................................................57 STUDY 3 RESULTS AND DISCUSSION.................................................................................. 64 POST-HOC TEST OF STIMULI................................................................................................. 69 GENERAL DISCUSSION......................................................................................................

.73

TABLES....................................................................................................................................... 86 FIGURES..................................................................................................................................... 94 REFERENCES............................................................................................................................. 97 APPENDICES.............................................................................................................................112 CURRICULUM VITAE.............................................................................................................115

vi

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

INTRODUCTION Although Asian American psychology has experienced tremendous growth over the past three decades (Leong, Okazaki, & David, 2006; Okazaki, 2002), this burgeoning field continues to be plagued with several limitations. First, most studies in Asian American psychology use East Asian or multi-ethnic samples that usually fail to capture the cultural and psychological heterogeneity among Asian Americans. Second, there persists a lack of situating Asian Americans’ psychological experiences within historical and environmental contexts (e.g., Root, 2002). Third, most of these studies focus on Asian cultural variables (e.g., collectivism) and cultural adaptation processes (e.g., acculturation), and less attention is paid to Asian Americans’ experiences of historical and contemporary forms of oppression (David & Okazaki, 2006a). Lastly, there has been a call for a methodological move “beyond questionnaires” (Okazaki, 2002). The current set of studies attempts to address these limitations by examining, through the use of indirect and implicit methodologies, how internalized historical and contemporary oppression operates within and consequently affects Filipino Americans. More specifically, the studies below investigate: (1) whether centuries of colonialism has been internalized by some Filipino Americans so that such a condition may now be activated automatically, unintentionally, and without control; (2) whether some Filipino Americans have automatically associated inferiority to anything Filipino and superiority to anything American; and (3) whether internalized colonialism - a specific form of internalized oppression - has adverse consequences on the mental health and psychological well-being of Filipino Americans. The cognitive operation and psychological implications of internalized colonialism among Filipino Americans cannot be satisfactorily understood without proper historical and contemporary context. Therefore, I first provide a brief discussion of the historical and current

1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

experiences of Filipinos with and within the U.S., followed by a discussion of Filipinos’ colonial history. Next, a review of internalized oppression and implicit social cognition literature will be provided to situate the current investigation within the larger scholarship of ethnic minority and cultural psychology. The Filipino American Population Filipinos were one of the first Asians on U.S. soil, with documentation of shipwrecked Filipinos landing in California as early as the mid-1500s (Gomez Borah, 1995) and of Spanish­ speaking Filipinos settling in the bayous of Louisiana as early as the mid-1700s (Cordova, 1983). The Philippines was a U.S. territory between 1898-1946 and Filipinos were regarded as U.S. nationals until 1938, making Filipino Americans the only Asian American ethnic group to have a history of being directly colonized by the U.S. Because of their status as U.S. nationals, which allowed them to easily enter and work in the U.S., the colonial period was also the time during which large-scale migration of Filipinos to the U.S. began. Filipinos in America during this period were mostly plantation and cannery workers in the western states of California, Hawaii, Oregon, Washington, and Alaska. However, because they were not citizens, American laws did not protect Filipinos and their rights and they became the subject of brutal discrimination and maltreatments (Cordova, 1983; Bulosan, 2002). For example, blatantly racist signs such as “Positively No Filipinos Allowed” were common sights in storefronts as an effort to keep Filipinos from using hotels, restaurants, swimming pools, apartments, barber shops, and other public facilities (Cordova, 1983). Filipinos were exploited by their employers, were derogatorily regarded as “little brown monkeys,” and were hunted and murdered by civilians and police officers (Bulosan, 2002; Cordova, 1983). Carlos Bulosan, whose writings are regarded as “the first and best account.. .o f .. .what it was like to be a Filipino in California and its sister

2

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

states.. .between.. .1930 to 1941” (McWilliams, 2002, p. vii), stated that he felt “like a criminal running away from a crime (he) did not commit. And this crime is that (he is) a Filipino in America” (as quoted by McWilliams, 2002, p. vii). Bulosan further explained that a Filipino in America during the first half of the 20th century “is the loneliest thing on earth.. .He is enchained, damnably, to his race, his heritage” (as quoted by McWilliams, 2002, p. xiii). Although the Philippines was given its independence in 1946, American influence continues to persist in contemporary Philippines as evidenced by the maintenance of U.S. military bases in the Philippines until 1992, the continued presence of American soldiers in the Philippines to “train” Filipino soldiers, and the continued use of English as the primary language in Philippine education, law, government, business, and science. Indeed, as Kamow (1989) suggests, “.. .in no place is the (American) imperial legacy more alive than in Manila, where America’s presence is almost as dynamic now as it was during the days of U.S. rule” (p. 16). Today, Filipino immigration rate into the U.S. (60,000 per year) is second only to that of Mexicans, attesting to this group’s rapid population growth and making the majority of the Filipino American population post-1965 immigrants. According to the 2005 U.S. Census, Filipinos are the second largest Asian ethnic group in the U.S. today (3 million total) and the largest Asian subgroup in the most populated state of California (United States Department of State, 2006). However, despite their large representation and their unique historical and contemporary relationship with and within the U.S., Filipino Americans continue to be regarded as the “forgotten Asian Americans” (Cordova, 1983) or the “invisible minorities” (Cimmarusti, 1996). The invisibility of Filipino Americans is mirrored within psychology, as research focused on the Filipino American population is relatively fewer than available research on other Asian

3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

American groups (David & Okazaki, 2006a). Findings from the relatively sparse body of Filipino American-focused research suggest that they are facing plenty of alarming issues that are in need of attention. For example, Filipino American adolescents have one of the highest rates of suicide ideations and attempts in the U.S. (President’s advisory commission on Asian Americans and Pacific Islanders, 2001) and Filipino American adults have higher depression rates than that of the U.S. population (Tompar-Tiu & Sustento-Seneriches, 1995). In a Center for Disease Control and Prevention study conducted in 1993, it was reported that 45.6% of Filipina American adolescents thought about suicide - the highest suicide ideation rate among all ethnic groups (as cited by Agbayani-Siewart & Enrile, 2003). High rates of HIV/AIDS, unintended pregnancy, eating disorders, STDs, and alcohol and drug use have also been reported for Filipino American communities across the U.S. (Nadal, 2000). In terms of educational attainment, Filipino American high school students have one of the highest school dropout rates in the country (President’s advisory commission on Asian Americans and Pacific Islanders, 2001). The 1994 U.S. Census reported that U.S.-born Filipinos have substantially lower college graduation rates compared to U.S.-born Chinese, Japanese, and Koreans. Furthermore, at the University of California Los Angeles and the University of California Berkeley, institutions in a state where the largest Asian American ethnic group is Filipinos, U.S.-bom Filipinos have the lowest college admission rate among all racial or ethnic groups (Okamura, 1998). In terms of socioeconomic status, Filipino Americans generally have lower status than other Asian Americans and their levels of educational attainment yield the lowest socioeconomic returns with respect to jobs and salary levels among all racial or ethnic groups (Okamura & Agbayani, 1997). These alarming health, educational, and socioeconomic statistics demonstrate that Filipino Americans do not fit

4

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

the Asian American model minority stereotype and suggest that their experiences may not be comparable to those of other Asian Americans. Culturally, two salient characteristics distinguish Filipino Americans from other Asian American groups: (1) Catholicism is the predominant religion among Filipino Americans; and (2) Filipino Americans have high English proficiency and familiarity with the American culture. These cultural characteristics that make Filipino Americans different from other Asian Americans are directly related to their colonized history. Therefore, the centuries of colonialism they experienced suggest that the cultural and psychological experiences of Filipino Americans may also differ from those of other Asian Americans. Unfortunately, this unique colonial past combined with the current health, education, and socioeconomic statistics discussed above are often overlooked in studies that aggregate multiple Asian ethnic groups into one sample, which is what is often done in Asian American psychology research. Filipino Colonial History Filipino history is often summarized by Filipinos and Filipino Americans in one sentence - 300 hundred years in the convent and 50 years in Hollywood. A more telling version of the same summary is “.. .we lived for 300 years in the convent under Spanish colonial rule.. .we partied for 50 years in Hollywood under American commonwealth rule.” This statement, written by a Philippines-based journalist (Mangahas, 2005), touches on the persistent legacies of both colonial periods on modem day Filipinos. It expresses the continued Catholic influence on most Filipinos and suggests that Filipinos view American colonialism as a liberating, pleasurable, and enlightening experience. Certainly, this simplified but popular single sentence rendition of Filipino history reveals that Filipino culture, and how culture influences their psychological experiences, cannot be completely understood outside the context of Spanish and American

5

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

colonialism (David & Okazaki, 2006a). Indeed, even the culture of the field of psychology in the Philippines was influenced by Western cultural imposition, as the indigenous psychology movements in the Philippines arose in response to the perceived connections between colonialism and Western psychology (e.g., Enriquez, 1993). Thus, I provide a brief discussion of Filipinos’ colonial experiences below. Colonialism under Spain In 1521, Portuguese explorer Ferdinand Magellan claimed what is now known as the Philippine islands for Spain. However, it was not until 1571 when Spain’s Miguel Lopez de Legaspi established the Spanish city of Manila that Spain finally subdued the natives’ resistance and secured Spanish rule over the islands (Agoncillo, 1974). Spanish colonized Filipinos are believed to have experienced exploitation, abuse, brutality, corruption, injustice, and tyranny as their “dignity and honor.. .were (continuously) taken for granted” by the Spanish (Agoncillo, 1974, p. 124). As part of Spain’s civilization efforts, the natives’ indigenous culture and spiritual beliefs were replaced by Spanish culture and the Catholic religion. Jose Rizal, the Philippines’ national hero who was executed by Spain because of his criticism of Spanish rule, wrote: “.. .little by little they lost their old traditions, the mementos of their past; they gave up their writing, their songs, their poems, their laws in order to learn by rote other doctrines which they did not understand, another morality, another aesthetics different from those inspired by their climate and their manner of thinking.. .degrading themselves in their own eyes; they became ashamed of what was their own; they began to admire and praise whatever was foreign and incomprehensible; their spirit was dismayed and it surrendered to.. .this disgust of themselves” (as cited by Rimonte, 1997, p. 58).

6

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

During Spanish rule, Rimonte (1997) argued that Filipinos developed a sense of colonial debt - a condition that is characterized by a deferential attitude toward Spanish culture and Spanish people and the tendency to accept maltreatments by the Spanish rulers as the natural cost for progress or civilization. She further added that such a feeling of indebtedness to their past colonizers may still be widely held among modem day Filipinos and Filipino Americans as maintained by the Golden Legend. The Golden Legend is a popular historical belief that preHispanic Filipinos were uncivilized savages who were nobly civilized by Spain through the gifts of Spanish culture and Catholicism. During Spanish rule, Rimonte asserted that the Catholic Church and its friars were instrumental in endorsing the Golden Legend by promoting the idea that those who do not “Hispanicize” have “strayed from the prescribed Catholic path of righteousness” (p. 59). The Golden Legend has been maintained throughout the years in more contemporary presentations of Filipino history, as even noted Filipino historian Teodoro Agoncillo (1974) was known to praise the “beautiful Spanish language” (p. 85) and the Catholic religion, which he regarded as “the most poetic of all religions” (p. 89). After being subjected to more than 300 years of ethnic and cultural subjugation, a series of U.S.-encouraged Filipino revolutions seriously dismpted Spain’s control over the Philippines tK

toward the end of the 19 century (Agoncillo, 1974). However, before U.S. forces even landed in the Philippines, the Filipino revolution against Spain was already successful as the Filipino revolutionaries were able to gain control of the entire island of Luzon except for the walled Spanish city of Manila. The Filipino revolutionaries declared their independence and established the first Asian government based on a democratically developed constitution. However, the Filipinos’ independence was short-lived as Spain sold the Philippines to the U.S. for $20 million

7

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

during the Treaty of Paris in 1898, which signaled the end of the Spanish-American war and the beginning of another long period of colonization for the Filipinos. Colonialism under the United States Before the Treaty of Paris was formalized by American and Spanish peace commissioners, Filipino representatives in Paris attempted to argue that because Spain had already been ousted from the Philippines and because Filipinos already had an independent government before the treaty was signed, Spain had no right to transfer ownership of the archipelago to the U.S. (Agoncillo, 1974). Filipino representatives in the U.S. used the same argument before the treaty was approved and, thus, finalized, by the U.S. Senate. However, both efforts to keep Filipino sovereignty failed and the U.S. proceeded to take control of the islands. From 1899-1902, Filipinos desperately attempted to gain independence from the U.S. in what is known as the Philippine-American war, a war that cost the U.S. $600 million and approximately 10,000 soldiers. Overall, about 16,000 soldiers and 200,000 civilians were killed in three years. Consequently, led by Mark Twain’s Anti-Imperialist League, many Americans began questioning America’s presence in the Philippines. As a response to such anti-imperialist criticisms, President William McKinley used the idea of Benevolent Assimilation (Blount, 1913; Rusling, 1903) to defend his motivation and intentions for colonizing the Philippines. For example, in a speech delivered to a delegation of Methodist church leaders in 1903, McKinley explained: I walked the floor of the White House night after night until midnight; and I am not ashamed to tell you, gentlemen, that I went down on my knees and prayed Almighty God for light and guidance more than one night. And one night late it came to me this way: (1) That we could not give them back to Spain - that would be cowardly and dishonorable;

8

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

(2) that we could not turn them over to France and Germany - our commercial rivals in the Orient - that would be bad business and discreditable; (3) that we could not leave them to themselves - they were unfit for self-government - and they would soon have anarchy and misrule over there worse than Spain’s was; and (4) that there was nothing left for us to do but to take them all, and to educate the Filipinos, and uplift and civilize and Christianize them, and by God’s grace do the very best we could by them, as our fellow-men for whom Christ also died. And then I went to bed.. .and the next morning I sent for the chief engineer of the War Department, and I told him to put the Philippines on the map of the United States (pointing to a large map on the wall of his office), and there they are, and there they will stay while I am President! (Rusling, 1987, p. 23). As American imperialist efforts proceeded, the U.S. established a nationwide public school system in which most of the educators were Thomasites - over 500 American teachers who came to the Philippines in 1901 through the St. Thomas transport. In 1902, the number of Thomasites sent to the Philippines more than doubled as the U.S. became convinced that “education, instead of outright military suppression, was the more effective means to pacify the Filipinos” (Espiritu, 2003, p. 26) and “win the confidence, respect, and affection of the inhabitants of the Philippines” (Blount, 1913, p. 649). As part of America’s effort to “educate.. .uplift.. .civilize and Christianize” the Filipinos (Rusling, 1987, p. 23) and uphold the U.S.’s tutelary regime (Go, 2003), the Americanized educational system taught Filipinos the English language, inculcated Filipinos with American culture and values, and replaced Filipino worldview with American political ideals (Pido, 1997). Thus, the Thomasites’ information about the U.S. may have distorted the Filipinos’ view of American culture. Consequently, Filipinos may have developed a grandiose picture of the American life that contributed to their large-scale

9

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

migration into the western U.S. where they experienced blatant racism. Simultaneous to the development of their superior perceptions of American culture, Filipinos may have begun to devalue Filipino culture (e.g., Espiritu, 2003; Strobel, 2001), similar to how Filipinos are argued to have developed the belief that the Spanish and their ways of life were naturally superior and that the indigenous Filipino culture was that of uncivilized savages (Rimonte, 1997). For contemporary Filipino American immigrants and their children, the psychological legacy of colonialism may continue to exist through inter-generational family socialization, continuous endorsement of the American version of the Golden Legend (i.e., perceiving Americans as freedom fighters, liberators who saved Filipinos from Spain, the masters of democracy, and enlightening heroes), continued oppression in the U.S., and continued Americanization of the Philippines that further endorses the notion of American superiority over the Filipino ethnicity and culture (David & Okazaki, 2006a; Espiritu, 2003). To this day, the large annual influx of Filipinos into the U.S. is argued to be an effect of colonialism and continued Americanization of the Philippines. Indeed, as political scientist Rodriguez (1997) stated, “Colonialism has fostered a perception that.. .the standard of living.. .in the United States is the mark of a highly sophisticated society (and culture).. .adults dream of going to the United States as if longing to be reunited with a long-lost parent.. .children dream of becoming Americans in the hope that they will finally be able to live in Disney’s Kingdom.. .For many Filipinos, coming to America means the fulfillment of a lifelong dream...” (p. 317-318). In terms of their contemporary experiences in the U.S., a recent study examining Chinese and Filipino Americans’ experiences of racism showed that 98% experienced racism in the past year and 99% reported witnessing other Asian Americans being subjected to racism. Within this sample, Filipino Americans reported a higher frequency of vicarious racism and direct daily life racism

10

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

than Chinese Americans (Alvarez, Huang, & Liang, 2006). Thus, Filipinos and Filipino Americans have experienced centuries of ethnic and cultural subjugation in the Philippines and in the U.S. (Lott, 1976). Given that Filipinos and Filipino Americans have faced centuries of both historical and contemporary forms of oppression, it is possible that oppression has influenced their psychological experiences. Psychological Consequences of Oppression and Colonialism Internalized Oppression The psychological consequences of oppression is a central theme in ethnic minority psychology. Various scholars have described internalized oppression - a condition in which an oppressed individual or group come to believe that they are inferior to those in power - as a salient consequence of systematic and persistent oppression. For instance, Thomas (1971) theorized that internalized racism among African Americans leads to identity confusion and to the development of a White-dependent and inferiorizing identity (“negromachy”). The Black Identity Development Model (Nigrescence Models) proposed by Cross, Parham, and Helms (1991) also argued that internalized racial oppression may lead African Americans to highly value the dominant culture and simultaneously devalue their own, which could lead many to hold anti-Black sentiments or hate their blackness. Furthermore, it has also been demonstrated that racial oppression is negatively related to African Americans’ physical and mental health (Landrine & Klonoff, 1996; Kolonoff, Landrine, & Ullman, 1999). Internalized oppression has also been argued to be common among members of sexual minority populations. For example, in his minority stress model for LGB individuals, Meyer (2003) argued that both distal (e.g., discrimination) and proximal stress processes (e.g., concealing one’s sexual orientation) affect LGB individuals’ mental health. Meyer further

11

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

proposed that the most proximal of the stressors (on the continuum from the environment to the self) is internalized homophobia - a specific form of internalized oppression in which LGB individuals eventually re-direct negative homophobic societal attitudes toward themselves. Even in the absence of direct and explicit discrimination, he argued that internalized homophobia could harm LGB individuals’ self-regard. Empirically, it has been shown that internalized homophobia is related to negative psychological outcomes (Williamson, 2000). Internalized Colonialism: A Consequence o f Classical and Internal Colonialism Regarding the consequences of oppression on cultural groups around the world, the colonial model initially described by Frantz Fanon (1965) provides a theoretical framework for understanding the psychological effects of oppression on the oppressed. The classical colonial model is composed of four phases, with the first phase being the forced entry of a foreign group into a geographic territory with the intention of exploiting the new territory’s natural resources, including its inhabitants. The second phase involves the establishment of a colonial society that is characterized by the colonizer imposing its culture on the colonized, disintegrating the indigenous culture of the colonized, and recreating the culture of the colonized as defined by the colonizer. Such a cultural transformation of the colonized people’s indigenous culture is intended to more clearly differentiate between the colonizer’s superior or more civilized ways of life and the colonized people’s inferior or savage ways. Once the colonial society has clearly contrasted the colonizer and the colonized other, the third phase begins as the colonized are portrayed as wild, savage, and uncivilized peoples whom the colonizer have to nobly monitor, tame, and civilize. Thus, the third phase essentially conveys that tyranny and domination is necessary. The implementation of the first three phases eventually leads to the fourth phase, which involves the establishment of a society where the political, social, and economic institutions are designed to

12

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

benefit and maintain the superiority of the colonizer while simultaneously and persistently subjugating the colonized. Thus, colonialism is another specific form of oppression. According to postcolonial scholars, the major psychological effect of colonialism is internalized oppression, or more specifically, internalized colonialism. Based on his work in French colonized Algeria, psychiatrist Fanon (1965) argued that the sustained denigration and inhumane treatments that the colonized are subjected to under colonialism often lead to self­ doubt, identity confusion, and feelings of inferiority among the colonized. Based on his work in French colonized Tunisia and Algeria, Albert Memmi (1965) added that the creation of a colonizer-defined cultural identity for the colonized often leads the colonized to eventually believe such an inferiorizing identity. Paolo Freire (1970) further contended that because of the inferior connotations the colonial society has attached to their cultural and ethnic identities, the colonized might develop an intense desire to rid oneself of such identities and try to emulate the colonizer as much as possible. Thus, the colonized may begin to talk, act, and dress like the colonizer because the colonizer’s ways are seen as superior. Simultaneously, the colonized may also begin to shed oneself of anything from his/her heritage culture and ethnicity because such a heritage is seen as inferior. Furthermore, the colonized may eventually feel a sense of gratitude and indebtedness (i.e., colonial debt) toward the colonizer for civilizing and enlightening the colonized (Rimonte, 1997). Within the U.S., the historical and contemporary oppression that ethnic minority groups have experienced may be seen as one of internal colonialism. Although there is no recent forceful entry by a foreign group, internal colonialism is analogous to classical colonialism in that the established society is characterized by racial inequalities, cultural imposition of the dominant group on the minority groups, cultural disintegration of the oppressed groups’

13

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

indigenous culture, and cultural recreation of the oppressed groups’ ethnic and cultural identities as defined by the dominant group. Harrell’s (1999) theoretical discussion of the psychological consequences of oppression on African Americans is an excellent example of how applicable the classical colonial model is in describing the experiences of American ethnic minority groups. In his analyses, Harrell (1999) used Fanon’s term to describe an oppressive society - Manichean. A Manichean society is one that is essentially based on incompatible opposites such as good versus evil, light versus dark, white versus black (or brown), or more generally, the superior colonizer versus the inferior colonized other. In a Manichean society, anything that is of the dominated group, including language, physical traits, and cultural values and traditions, is ascribed with inferior, undesirable, or negative characteristics. Concurrently, anything that is of the dominant group is attached with superiority and desirability. Furthermore, a Manichean society also involves the destruction and reinterpretation of the history and culture of the oppressed through the eyes of the dominant group. Consequently, Harell argues that a Manichean society creates conditions that lead African Americans to develop self-hatred and encourage them to behave in self-destructive ways. Consistent with Fanon’s and Harrell’s theories, Tatum (1994) also proposed that colonialism, or more specifically, internal colonialism, is another possible explanation for the high rates of crime and delinquency among African Americans. She argued that crime and delinquency may be seen as the self-destructive behavioral responses to a society wherein opportunities for social mobility are limited because of one’s race. Among Native Americans, their experiences may be described as involving both classical and internal colonialism. The colonization that Native Americans experienced and the colonizer’s efforts to destroy and recreate indigenous Native American culture, as exemplified by the boarding school era, attempted genocide, and geographic displacement, is argued to have

14

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

led many Native Americans to lose their cultural identity and spirituality (McBride, 2002). The internalization of such historical and contemporary forms of oppression is also argued to contribute to cultural isolation, vocational stresses, and dysfunctional behaviors such as substance abuse and domestic violence among Native Americans (McBride, 2002). Duran and Duran (1995) and Brave Heart (1998) also argued that internalized oppression is promoted and passed on intergenerationally by continued oppression, lack of opportunities to critically and accurately understand history, forced Americanization, and socialization - contemporary forms of oppression that may be seen as internal colonialism. Furthermore, using the colonial model, these authors argue that colonialism and contemporary forms of oppression may contribute to the high rates of suicide, alcoholism, and domestic violence among Native Americans. The intergenerational transmission of the psychological consequences of oppression has also been observed among Jewish holocaust survivors and their children (e.g., Major, 1996; Sorscher & Cohen, 1997) and Japanese American internment camps survivors and their children (e.g., Nagata & Cheng, 2003; Nagata, Trierweiler, & Talbot, 1999). Among Hispanic Americans (e.g., Cubans, Dominicans, Mexicans, Puerto Ricans, etc.), the implicit effect of skin color on determining important life factors such as employment, income, and self-concept that exists in American society has been argued to be especially problematic for this ethnic minority group - a group that is “characterized by a rainbow of skin colors and diverse physical attributes” (Hall, 1994, p. 307). Hall argued that colonization and its accompanying “domination model of assimilation” (pp. 309-310) leads many Hispanic Americans to believe that light skin is the most advantageous, attractive, and desirable skin color. The internalization of such a skin-color ideal that is perpetuated in American society consequently results into a perceived necessity to become as white as possible in order for social

15

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

mobility. This desire to shed one’s natural, but societally undesired, physical traits has been argued to lead many Hispanic Americans to use “beauty” creams and other products such as bleach in order to whiten their skin, a phenomenon known as the “bleaching syndrome” (Hall, 1994). Indeed, according to Hall, many Hispanic Americans “will value and internalize all aspects of the mainstream culture - including the idealizations of light skin color - at the expense of their culture” (p. 310). Empirically, using a Mexican American sample, Codina and Montalvo (1994) found that darker skin color and loss of Spanish culture (i.e., language) was associated with higher levels of depression. A specific Hispanic American ethnic group that has received some psychological research attention in terms of their experiences of internalized colonialism is the Puerto Rican population. Strikingly comparable to the colonial experiences of Filipinos, Puerto Ricans were also colonized by Spain in 1493 and were sold to the U.S. during the same Treaty of Paris in 1898. Puerto Rico remains an American territory and Puerto Ricans continue to be regarded as U.S. nationals to this day. The effects of centuries of Spanish and American colonialism on the psychological experiences of modem day Puerto Ricans are remarkably similar to David and Okazaki’s (2006a; 2006b) description of the psychological consequences of Spanish and American colonialism among Filipino Americans. For instance, Varas-Diaz and Serrano-Garcia (2003) found that it is common for Puerto Ricans to experience identity confusion, feel ashamed of their ethnic and cultural identity, feel inferior about being Puerto Rican, and not have national pride. Furthermore, these researchers also reported that Puerto Ricans despise the stereotypical and mythical perceptions often imposed on them by mainland Americans, including Puerto Ricans in the mainland U.S. Finally, they also found that negative emotions such as shame, anger, desperation, and disillusion were associated with what the researchers called “the Puerto

16

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Rican experience” (p. 112). A similar psychological phenomenon relating to loss of identity has also been observed among the Chamorro group in the currently U.S. colonized island of Guam (Perez, 2005). Colonial Mentality Among Filipino Americans In her description of how centuries of colonialism may have affected the psychological experiences of Filipinos and Filipino Americans, noted psychologist Maria Root (1997) stated: Four hundred years of combined colonization.. .widened the Filipino gene pool with the possibilities of lighter skin, hair, and eyes.. .Spain introduced colorism; preferential treatment was clearly associated with lighter skin color. Centuries of this education primed the Filipino for vulnerability to internalize American rules of race. Colorism and then racism inculcated the notions “White is beautiful,” “White is intelligent,” and “White is powerful” in the psyches of many brown-hued Filipinos, thus inferiorizing the Filipino (p. 81). It has been argued that, among Filipinos and Filipino Americans, ideas of superiority, pleasantness, or desirability have been associated with, not just physical characteristics (Whiteness), but also anything American or Western - a condition of internalized colonialism popularly referred to as Colonial Mentality (David & Okazaki, 2006a; 2006b; Root, 1997; Strobel, 2001). David and Okazaki (2006b) conceptualized Colonial Mentality (CM) as being “.. .characterized by a perception of ethnic or cultural inferiority that is.. .a specific consequence of centuries of colonization under Spain and the U.S.” and that it “.. .involves an automatic and uncritical rejection of anything Filipino and an automatic and uncritical preference for anything American” (p. 241). It is important to note, however, that although some Filipinos and Filipino Americans are argued to have internalized the historical and contemporary oppression they have

17

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

faced, not all Filipinos may hold CM. This may be especially true for the Muslim Filipinos in the southern Philippine island group of Mindanao who has been resisting colonization and westernization ever since the beginning of Spanish occupation. Furthermore, CM is conceptualized as an individual differences variable whose existence and strength within Filipinos and Filipino Americans may also greatly vary (David & Okazaki, 2006a). Although the existence and strength of CM among Filipinos and Filipino Americans may vary, Filipino American scholars and community leaders have nonetheless speculated that CM is prevalent among members of this ethnic group. Fred Cordova (1973), a distinguished Filipino American historian, argued that colonialism and the accompanying cultural imposition, cultural disintegration, and cultural recreation have contributed to the ever-present Filipino ethnic/cultural identity crisis - confusion as to what constitutes an authentic Filipino culture and identity. Because the dominant presentations of Filipino culture and history is often presented in a mythical, distorted, and inferiorizing manner, such an identity crisis is believed to lead Filipino Americans toward the conclusion that there is no authentic Filipino culture and identity that one can be proud of, and thus, may lead to the perception of inferiority toward anything Filipino (David & Okazaki, 2006a). Lott (1976) also discussed the common presence of such a condition among contemporary Filipino Americans, which she attributed to the continued subjugation of this group within the U.S (i.e., internal colonialism) and the likelihood that immigrant Filipinos may have brought with them such a condition from the persistently Americanized postcolonial Philippines. An analysis of Filipino American student essays and community newsletters suggest that CM is indeed a common phenomenon within this ethnic group (Revilla, 1997). Bergano and Bergano-Kinney (1997) also estimated that at least half the Filipino American population has some form of CM. In terms of its impact on the Filipino American population, editorials in

18

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Filipino American community publications cite CM as a major reason for the lack of societal presence and social unity of the Filipino American community, and the lack of cultural pride, historical knowledge, and cultural appreciation among Filipino American individuals (e.g., Gaston, 2003). Psychiatrists Tompar-Tiu and Sustento-Seneriches (1995) also hypothesized that CM may possibly contribute to the high depression rates (27.3%) they found for this group. In their effort to assess, quantify, and more systematically study the CM construct and its impact on Filipino Americans, David and Okazaki (2006b) developed the Colonial Mentality Scale (CMS) for Filipino Americans. From this study, they found that CM may be manifested covertly (Covert Manifestations of CM [CMCM]; e.g., feelings of shame, embarrassment, and inferiority) and overtly (Overt Manifestations of CM [OMCM]; e.g., desiring to look and behave more White, discriminating against less Americanized Filipinos). Thus, a person may endorse OMCM but not CMCM, endorse CMCM but not OMCM, endorse both types, or endorse neither. Furthermore, using the CMS, David and Okazaki found that approximately 30% of their sample endorsed at least one type of CM manifestation, suggesting that many Filipino Americans hold CM. Interestingly, they also found that 30% of their sample reported depression symptoms that may be considered as clinically significant, similar to the findings by Tompar-Tiu and SustentoSeneriches (1995). Finally, David and Okazaki also found that CM is negatively related to a person’s personal and collective self-esteem, and positively related with depression symptoms. Although the CMS allows for a comprehensive and systematic assessment of CM and how it affects the psychological experiences of Filipino Americans, it is not without limitations. For instance, some Filipino Americans may not endorse the covert manifestations of CM for a variety of reasons (e.g., they discriminate against less Americanized Filipinos but do not necessarily feel ashamed of the Filipino culture; unaware of presence of CMCM; denial; etc.).

19

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Even though the CMS attempts to tap into both the overt and covert aspects of CM, it is still only a direct, explicit, and introspection-dependent self-report measure that is more oriented toward capturing the overt manifestations of CM. Several of David and Okazaki’s (2006b) findings are suggestive of such a limitation. For example, they found that fewer individuals endorsed CMCM (n=64) than OMCM (n=84) and only about 30% of the sample endorsed at least one type of CM manifestation, which is a lower frequency estimate of CM than what some scholars have suggested (e.g., Bergano & Bergano-Kinney, 1997; Pido, 1997; Revilla, 1997). Furthermore, David and Okazaki (2006b) also argued that CM “involves an automatic and uncritical rejection o f anything Filipino and an automatic and uncritical preference for anything American ” (p. 241). Thus, in theory, there are two critical characteristics of CM: (1) it involves associating Filipino-related objects with inferiority, unpleasantness, or undesirability and American-related objects with superiority, pleasantness, or desirability; and (2) it may operate automatically without intention or control. Given that some aspect of CM is theorized to operate automatically, introspection-dependent and explicit self-report measures of the construct such as the CMS may be insufficient to assess its implicit aspect. Indeed, a more implicit or indirect method of measuring CM may provide a more accurate estimate of CM frequency and how CM influences the psychological experiences of modem day Filipino Americans. Fortunately, there is an emerging body of research in implicit social cognition that may provide Asian American psychology with conceptual frameworks and methodological paradigms to go “beyond questionnaires” (Okazaki, 2002) in order to better capture and understand the covert and possibly less consciously accessible aspect of CM.

20

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Implicit Social Cognition Activating Colonial Mentality There are two types of cognitions - conscious and unconscious (Greenwald & Banaji, 1995). This widely accepted notion has its roots from implicit memory research, which has empirically demonstrated that past experiences (i.e., learning) unconsciously or implicitly influence later performance (e.g., recall). The acknowledgment that unconscious memory (or learning) exists has been documented as early as the late 1800s (Ebbinghaus, 1885/1964; Freud & Breuer, 1895/1960) and has been more recently supplemented by studies investigating implicit memory among amnesics (Baddeley & Warrington, 1970; Milner, Corkin, & Teuber, 1968) and implicit retention among normal participants (e.g., Jacoby, 1988). In contemporary implicit memory research, participants are exposed to, or primed with, a word such as “DOG” in the beginning of a study and then later asked to complete several tests, including word fragment completion (e.g., D G and C _ T), word stem completion (e.g., D __ and C

), word identification (presenting the word “DOG” and “CAT” rapidly on a computer

screen and asking participants to report what they saw), anagram solution (e.g., GDO and TCA), and picture fragment identification (e.g., showing an incomplete picture of a dog and a cat), among others (Roediger, 1990; Roediger, et al., 1992). Priming is said to occur when performance is better on studied (or previously exposed) words (i.e., “DOG”) than non-studied words (i.e., “CAT”) in these tests. Also, notice that most of the dependent tasks given above may be completed in multiple ways, such as “DIG” for word fragment completion, “DOT” in word stem completion, and “GOD” for anagram solution. Priming is also said to occur if participants who were primed with the word “DOG” in the beginning of the study consequently respond with the word “DOG” instead of the other possible responses in the subsequent dependent task.

21

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Finally, implicit memory research have also shown that prior exposure (e.g., learning) to an object or stimulus may later influence a person’s ability to remember or identify the object or stimulus even if the person cannot explicitly remember being exposed to (or learning about) the stimulus; hence the term implicit learning. The priming paradigm has since been extended to test the spreading activation theory of cognition, which states that the human mind is composed of a network of nodes (representing concepts) and links (representing associations or relations between concepts) and that exposure to a particular stimulus (e.g., the word “MOTHER”) will activate the corresponding node (i.e., the “mother node”) in one’s cognitive network which, in turn, activates other linked nodes (e.g., woman, father, wife, etc.) that a person may have developed through learning and experience (e.g., Anderson & Pirolli, 1984; Collins & Loftus, 1975; Fazio, et al., 1986). Furthermore, exposure to a particular stimulus and the resulting chain of activation may also function outside of a person’s explicit awareness or control. Using the popular semantic priming paradigm (e.g., Neely, 1977), numerous studies have found evidence supporting the notion that judgments or actions may be influenced by unperceived stimuli that automatically activates the cognitive network which, in turn, guides consequent judgment or action (e.g., Balota, 1983; Bomstein, 1992; Fazio, et al., 1986; Fazio, et al., 1995; Greenwald, Klinger, & Liu, 1989; Perdue, et al., 1990). The priming paradigm has also been used to demonstrate that mere exposure to highly valenced objects or words (e.g., “DEATH”) automatically activates and increases the accessibility of the corresponding affective or attitudinal evaluation (e.g., UNPLEASANT), suggesting that evaluations of or attitudes toward objects are automatically activated and may operate outside of one’s awareness, intention, or control (e.g., Draine & Greenwald, 1998; Greenwald, Draine, & Abrams, 1996). Thus, in this area of research, priming no longer simply

22

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

refers to implicit learning as in implicit memory research, but also implicit activation of particular attitudes, judgments, and behaviors (Greenwald, Klinger, & Schuh, 1995). More recently, the priming technique has been applied to the study of culture, demonstrating that cultural influences on psychological processes may be investigated using indirect methodologies. For example, the cultural priming technique has been used to study cultural cognition (e.g., Benet-Martinez, & Haritatos, 2005; Benet-Martinez, Leu, Lee, & Morris, 2002; Hong, Chiu, & Kong, 1997; Hong, Morris, Chiu, & Benet-Martinez, 2000), cultural differences in self-construals (e.g., Gardner, Gabriel, & Lee, 1999; Trafimow, Silverman, Fan, & Law, 1997; Trafimow, Triandis, & Goto, 1991), and in activating the salience of one’s racial, ethnic, or gender identity (e.g., Ambady, Shih, Kim, & Pittinsky, 2001; Devos, 2006; Steele, 1997; Steele & Aronson, 1995). Of particular relevance to the current investigation is the study by Hong, Chiu, and Kung (1997). Hong and her colleagues primed Hong Kong biculturals, who are believed to have internalized two separate cultural knowledge systems (Western and Chinese), with pictures of either Western cultural icons or Chinese cultural icons to activate either their Western cultural knowledge system or their Chinese cultural knowledge system, respectively. Next, all participants were asked to interpret an ambiguous stimulus (i.e., a picture of a school of fish, wherein one fish was swimming separately and ahead of the other fish). They found that those primed with Western cultural pictures interpreted the ambiguous stimulus in a more individualistic manner (e.g., the lone fish is the leader), whereas those primed with Chinese cultural pictures interpreted the fish picture in a more collectivistic manner (e.g., the group of fish is chasing the lone fish). Such results suggest that Hong Kong biculturals’ Chinese cultural knowledge system involves the concept of collectivism whereas their Western cultural knowledge system involves the concept of individualism. Hong and her colleagues concluded

23

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

that activating a particular cultural knowledge system may influence later behavior or judgment so that it is consistent with the activated cultural system. Based on their findings, Hong and colleagues (2000) formally proposed the Dynamic Constructivist Approach to the study of culture and cognition. They argued that many bicultural individuals possess two internalized cultures that take turns in guiding their thoughts, feelings, and behaviors. Furthermore, they stated that

. .internalized culture (is) a network of discrete,

specific constructs that guide cognition only when they come to the fore in an individual’s mind” (p. 709) and that “.. .activation of (a particular) cultural knowledge (system or network) may have important influences on emotions and motives as well as judgments and decisions...” (p. 718). For many ethnic and cultural minority groups, their experiences of their heritage ethnicity or culture may involve constant denigration and oppression, which they might internalize. For some Filipino Americans, it is possible that the Filipino cultural knowledge system involves ideas of inferiority, undesirability, or unpleasantness, whereas their American cultural knowledge system involves ideas of superiority, desirability, or pleasantness. Indeed, the existence of such a cultural knowledge construction will be indicative of the extent to which CM has been internalized. Thus, can the priming techniques described above be used to activate and observe such a CM-consistent cultural knowledge system among Filipino Americans? If so, can we activate it implicitly or indirectly and test whether CM may be activated automatically? Colonial Mentality as Implicit Associations Following the growth of research utilizing priming techniques, Greenwald, McGhee, and Schwartz (1998) introduced the Implicit Association Test (IAT) as an alternative technique for measuring individual differences in evaluative associations that underlie implicit attitudes. In contrast to the priming technique, the IAT does not (subliminally or supraliminally) prime,

24

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

present, or expose individuals to a word or object. Instead, the IAT directly tests a hypothesized learned and possibly implicit association (or link) between objects (or nodes) by asking individuals to categorize a set of attribute terms into their hypothesized corresponding categories. For example, individuals are initially (Block 1) asked to press “A” on a computer keyboard if they see a word referring to a FLOWER on the computer screen or press “5” if the word they see refers to an INSECT. Block 2 asks individuals to press “A” if the word they see is PLEASANT or “5” if the word they see is UNPLEASANT. Next, the participants are asked to press “A” if they see either a FLOWER or PLEASANT word and press “5” if they see either an INSECT or UNPLEASANT word for Block 3. Block 4 simply asks the individuals to switch the keys (i.e., press “A” for INSECT or “5” for FLOWER) and Block 5 asks them to press “A” for either an INSECT or PLEASANT word and press “5” for either a FLOWER or UNPLEASANT word. In this example, one would expect participants to find Block 3 easier (e.g., quicker reaction times, fewer mistakes) than Block 5 because of the existence of strong and automatic learned associations that people have of insects to unpleasant feelings and flowers to pleasant feelings. Consistent with the major premise of implicit memory, Greenwald and Banaji (1995) stated that the “signature of implicit cognition is that traces of past experience affect some performance, even though earlier experience is not remembered in the usual sense - that is, it is unavailable to self-report or introspection” (pp. 4-5). Many Filipino Americans may not readily identify colonialism and contemporary oppression (past experiences) as being influential in their current attitudes, feelings, and behaviors toward the Filipino culture and ethnicity. This may partly explain the lower than expected overall CM frequency estimates obtained by the CMS because this measure relies heavily on accurate self-report and introspection. Moreover, Greenwald and his colleagues further extended the reach of implicit cognition by stating that

25

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

“Implicit attitudes are manifest as actions or judgments that are under the control of automatically activated evaluation, without the performer’s awareness of that causation” (1998, p. 1464) and that implicit attitudes are “introspectively unidentified (or inaccurately identified) traces of past experience that mediate favorable or unfavorable feeling, thought, or action toward social objects” (1995; p. 8). Some Filipino Americans may not be able to accurately introspect and identify attitudes, thoughts, or feelings they hold toward the Filipino culture and ethnicity. Furthermore, some Filipino Americans may also be better at accurately identifying, recognizing, and reporting OMCM because such attitudinal and behavioral manifestations are more explicit or overt and, thus, more obvious. This may partly contribute to the lower frequency estimates of CMCM than OMCM as obtained using the CMS. Thus, is CM internalized enough by some Filipino Americans so that it may now operate automatically and unintentionally? Do some Filipino Americans automatically associate anything Filipino with inferiority and anything American with superiority, as proposed by CM theory (David & Okazaki, 2006a; 2006b)? Colonial Mentality and Mental Health Numerous studies have been conducted using the IAT procedure since its introduction in 1998, including investigations of attitudes toward African Americans, LGBT individuals, smoking, obesity, presidential candidates, and commercial products, among many others. One application of the IAT that is particularly relevant to psychological well-being and mental health is its application to the study of self-esteem and self-concept (Greenwald & Famham, 2000). In this application, one’s self-esteem is assessed by comparing one’s performance during the SELFPLEASANT block to one’s performance during the SELF-UNPLEASANT block. Better performance on the SELF-PLEASANT block is believed to indicate higher self-esteem.

26

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

However, one’s self-concept is not just composed of the individual or personal self. Instead, self-concept is composed of both a personal and a collective component and each component can be associated with either positive (or pleasant) or negative (or unpleasant) attributes (Tajfel & Turner, 1986). If personal self-esteem is the extent to which individuals evaluate their personal selves positively, collective self-esteem is the extent to which individuals evaluate the social groups they belong to positively (Crocker & Luhtanen, 1990, Luhtanen & Crocker, 1992). Furthermore, research shows that obtaining and maintaining positive personal and collective selves are essential for psychological health (Crocker, Luhtanen, Blaine, & Broadnax, 1994). Unfortunately, the existing application of the IAT to assessing one’s selfconcept has been limited to the personal component of the self alone. Thus, the implicit or automatic operation of one’s collective self and the associations or evaluations one may have about his/her social groups (an indicator of collective self-esteem) remain to be explored. One important social group for ethnic minority individuals is their ethnic or cultural group. The extent to which ethnic minority individuals evaluate their heritage ethnic or cultural group is one aspect of their collective self-esteem. For Filipino Americans, centuries of classical and internal colonialism may have resulted into CM, which in turn, may negatively influence their collective self-esteem. Such a negative regard toward one’s Filipino heritage may adversely affect one’s psychological well-being and mental health, which may partly explain the high rates of depression found for this population (Tompar-Tiu & Sustento-Seneriches, 1995). Indeed, Rimonte (1997) speculated that the “persistent self-hate of which many acts of anti-Filipinism are the chiefest (sic) manifestations.. .produces an acute, destabilizing, (and) discomfiting selfawareness... ” among Filipinos (pp. 41-42). Furthermore, Strobel (1997) identified feelings of anger, betrayal, confusion, doubt, and anxiety as some psychological effects of CM. Empirically,

27

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

using a direct and explicit self-report measure of CM (i.e., the CMS) on a national sample of over 600 Filipino Americans, David and Okazaki (2006b) found that CM is negatively associated with the psychological well-being and mental health of Filipino Americans. Furthermore, they found CM to be uniquely and significantly contributing to Filipino Americans’ depression symptoms above and beyond the contributions of self-esteem and status variables. More recently, also using the CMS to measure CM, David (under review) compared a conceptual model of depression that does not include CM to a model that includes CM using structural equation modeling and found that CM has a significant direct effect on depression symptoms among Filipino Americans. Furthermore, the CM model was able to explain approximately 63% of the variance in reported depression symptoms. Indeed, empirical evidence seems to suggest that CM is consistently related to Filipino American mental health. However, because CM is theorized to involve a covert component that the CMS may have failed to accurately capture, a combination of explicit and implicit measures of CM may be able to provide a more accurate understanding of how CM is related to the psychological well-being and mental health of Filipino Americans. Overview of Research In summary, the following studies attempt to address multiple questions. First, can priming techniques be used to activate and capture a CM-consistent cultural knowledge system among Filipino Americans (Study 1)? If so, can CM be indirectly primed or automatically activated by mere exposure to Filipino and American stimuli (Study 2)? Is CM internalized enough by some Filipino Americans so that it may now operate outside of awareness, intention, or control (Study 2 and Study 3)? Have Filipino Americans associated inferiority with the Filipino culture and superiority with the American culture (Study 3)? Finally, can we provide

28

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

further evidence to support previous findings that CM has adverse psychological effects on Filipino Americans (Study 2 and Study 3)? Study 1: Activating Colonial Mentality Due to centuries of classical and internal colonialism, many Filipinos’ Filipino cultural knowledge system may include ideas of inferiority, shame, and unpleasantness, whereas their American cultural knowledge system may include ideas of superiority, desirability, and pleasantness. Thus, when the Filipino cultural system is activated, the activation may influence or guide an individual to make an unpleasant, negative, or undesirable interpretation of an ambiguous stimulus. Similarly, it is possible that when the American cultural system is activated, such activation may influence or guide an individual to make a pleasant, positive, or desirable interpretation of the same ambiguous stimulus. Based on the literature on priming, spreading activation, the dynamic constructivist approach to culture and cognition, internalized oppression, and CM among Filipino Americans, it was hypothesized that such a CM-consistent cultural knowledge system construction, one that indicates inferior perceptions of the Filipino culture and superior perceptions of the American culture, will be observed. As an initial attempt at capturing such a hypothesized CM-consistent cultural knowledge system, two simple priming experiments were conducted. The sole purpose of Study 1 was to test whether a CM-consistent cultural knowledge system may be activated using priming techniques.

29

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

STUDY 1A METHOD Participants Using a snowball sampling procedure and the internet for study recruitment and administration, 172 self-identified Filipino Americans (59.90% female) from all over the United States participated in Study 1A. The average age of the sample was 28.31 years (SD -10.75), with a range of 18-66 years. Seventy-three percent were single, 22% were married, and the rest identified as “Other” (e.g., widower, divorced). Sixty-three percent had a college degree or higher and 42% had a yearly income of $55,000 or below. Sixty-two percent were second or later generation Filipino Americans, 52% were members of Filipino American organizations, and 65% were from West Coast states. Individuals who expressed interest in participating were randomly assigned to two conditions. However, although interested individuals were randomly directed to either the Filipino or the American condition, not all interested individuals actually completed the survey and became study participants. Thus, the number of participants for each condition slightly differs from one another. The Filipino condition contained 81 participants (60% female; average age of 26.53 years; 80% single; 60% with college degrees or higher; 40% with income of $55,000 or less; 69% 2nd or later generations; 59% were members of Filipino American organizations). The American condition contained 91 participants (59% female; average age of 29.93 years; 67% single; 76.7% with college degrees or higher; 44.2% with income of $55,000 or less; 56% 2nd or later generations; 48.4% were members of Filipino American organizations). A series of chi-square test of distributions found a statistically significant difference only on educational achievement between the participants in the Filipino and American conditions, %2(4, N = 172) = 12.75, p = .013. However, educational achievement did not seem to significantly influence the responses on the dependent variable, %2(8, N = 172) =

30

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

2.99, n.s. The mostly comparable demographic characteristics of participants between the two conditions suggest that random assignment was successful. Measures Word Fragment Completion (WFC) Tasks: For Study 1,1 borrowed the word fragment completion task often used in implicit memory research (e.g., Rajaram & Roediger, 1993; Roediger, et al., 1992) and used it as a way to activate either the American cultural knowledge system or the Filipino cultural knowledge system. Activating one of the cultural systems will make related constructs in the particular system more accessible and, “the more accessible a construct, the more likely it is to come to the fore in the individual’s mind and guide interpretation (of an ensuing ambiguous stimulus)” (Hong, et al., 2000, pp. 710-711). There were two separate WFC tasks used in this study, one containing Filipino cultural terms and the other containing American cultural terms. I generated 10 common terms that are easily and clearly identifiable as belonging to either the Filipino culture (five terms) or the American culture (five terms) in order to activate either the Filipino or the American cultural knowledge system. Each of the five cultural priming terms for one culture is conceptually the same as the five cultural priming terms for the other culture. For example, a cultural priming term referring to people (i.e., AMERICANS) from the United States in the American culture word list has a conceptual counterpart in the Filipino culture word list (i.e., FILIPINOS), a term referring to language used in the United States (i.e., ENGLISH) in the American culture word list has a conceptual counterpart in the Filipino culture word list (i.e., TAGALOG), and so on. Each cultural priming term represents a major cultural component: people (Americans Vs. Filipinos), language (English Vs. Tagalog), country (United States Vs. Philippines), capital city (Washington Vs. Manila), and the skin color commonly associated with people from each

31

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

country to represent phenotype (White Vs. Brown). Except for the terms “White” and “Brown,” where only one letter from each was randomly deleted, two letters from each of the other terms were randomly deleted. I also generated one word fragment that may be completed in two different and oppositely valenced manner -

ERIOR. This word fragment, which serves as

the dependent variable, may be completed either as “Superior” or “Inferior” and was presented as the last word fragment in both the Filipino and American WFC lists. The WFC tasks are presented in Appendix A. It is important to note that in implicit memory research, tasks such as WFC are traditionally used only as the dependent variable and not as the priming mechanism. The priming strategies often used in memory research are more indirect and less obvious such as having participants study a word list, showing participants a series of pictures or words, or rapidly (subliminally) flashing a picture or word on a computer screen. However, because this was my initial attempt at activating a CM-consistent cultural knowledge system, I reasoned that explicit, direct, and stronger primes might be needed. Thus, because the WFC task involves an effortful and deliberate retrieval of constructs (Graf & Mandler, 1984; Roediger, et al, 1992), I believed that such a task will likely activate either the Filipino or American cultural system depending on what terms are asked. In other words, I wanted to make sure that the priming strategy used will make participants think of either the Filipino or the American culture and the WFC task seems to accomplish this goal because the participants are actively retrieving (i.e., thinking) and producing (i.e., typing) culture-related information instead of passively receiving information (e.g., being shown a word or picture). Furthermore, because I used the WFC task as the priming mechanism and because I was not concerned with the participants’ performance on the prime words (e.g., length of time it took participants to complete the prime, errors, etc.), factors such as

32

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

length, readability, popularity (i.e., how common the word is), and overall difficulty (e.g., how many letters are missing) of the word fragment primes were not relevant for these purposes. That is, such factors are important to consider when the WFC task is used as the dependent variable (wherein the researchers are interested in participants’ performance on the WFC task), but not so much when it is used as the priming mechanism. The one WFC task in Study 1 in which participants’ performance was of concern, the dependent variable (i.e.,

ERIOR), takes such

difficulty factors into account and was made exactly the same for the two conditions. Finally, regardless of whether the terms “inferior” or “superior” may differ in terms of their frequency of use in daily language, their respective frequency should not be different between the two conditions if the priming task was not successful or if a CM-consistent cultural knowledge construction does not exist. Procedures The WFC tasks were administered in the beginning of another large-scale study about “Filipino American psychology.” Because the 2000 U.S. Census reported that Asian and Pacific Islanders have the highest rate of residential Internet access (Newburger, 2001) and because it is difficult to obtain large numbers of Filipino American participants, the study was made available on-line. The study was advertised through Filipino student and community organizations and “snowballing” was encouraged. Flyers and e-mails clearly stated that one has to be Filipino or Filipino American to participate in the study. Interested persons were screened for 2 criteria: (1) at least 18 years old and; (2) identify as Filipino. Qualified persons received the web page and password to access either the American condition survey or the Filipino condition survey. The survey condition to which a participant was directed was determined randomly and participants were not aware of the different survey conditions. Upon logging-in, they were screened again by

33

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

the survey, which was designed so that persons who did not satisfy the criteria were not able to continue. Those who were eligible were taken to the consent page, where they had to click on the “I Agree” button before completing the survey. Upon agreeing to participate, participants were asked to complete the WFC task, which was declared to be a test of whether the survey was functioning properly on their computers. They were reminded that their responses on the WFC tasks are unrelated to the survey they were about to complete. Each term on the WFC tasks appeared on separate pages. Participants were instructed to complete each term with the first word they could think of. Upon completion of the last WFC term, the rest of the web-based survey was presented and the participants were properly debriefed at the end. Recent investigations of web-based studies showed that this method generates samples and results that are of comparable quality to those obtained by traditional methods (Gosling, Vazire, Srivastava, & John, 2004; Kraut, et al., 2004). Study IA Hypotheses Under normal circumstances, or if the priming conditions did not have an effect, the number of participants who answer “Inferior” in the last word (i.e.,

ERIOR) of the Filipino

condition should not be significantly higher than the number of participants who answer “Inferior” in the American condition; and the number of “Superior” responses in the Filipino condition should not be lower than “Superior” responses in the American condition. That is, because participants were randomly assigned to the conditions, there should be no such differences in the distribution of responses between the two conditions if CM does not exist or if a CM-consistent cultural knowledge system was not activated within some Filipino Americans. Therefore, I hypothesized that the number of participants in the American condition who will fill-in the last word so that it is “Superior” will be significantly higher than participants in the

34

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Filipino condition who answers “Superior.” On the other hand, it was also predicted that the number of participants in the Filipino condition who will fill-in the same word so that it is “Inferior” will be significantly higher than the number of participants in the American condition who answers “Inferior.”

35

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

STUDY 1A RESULTS All participants correctly completed all the WFC primes, suggesting that the American or Filipino cultural words in the prime tasks were easily identifiable and commonly known to the participants. Regarding the results of the WFC dependent task (Table 1A), results showed that 27.2% of Filipino Americans primed to think of the Filipino culture interpreted the ambiguous stimulus (i.e.,

ERIOR) as “Inferior,” whereas only 11.0% of Filipino Americans primed to

think of the American culture interpreted the ambiguous stimulus as “Inferior.” Instead, 72.5% of Filipino Americans primed to think of the American culture interpreted the ambiguous stimulus as “Superior” compared to only 56.8% of Filipino Americans primed to think of the Filipino culture who interpreted the ambiguous stimulus as “Superior.” More Filipino Americans in the Filipino condition responded with “Inferior” (frequency=22) than what should be expected (15) and fewer responded with “Superior” (frequency=46) than what should be expected (53), suggesting that inferior perceptions were activated when they were thinking of the Filipino culture. The opposite pattern was observed in the American condition, where more Filipino Americans responded with “Superior” (frequency=66) than what should be expected (59) and fewer responded with “Inferior” (frequency=10) than what should be expected (17), suggesting that superior perceptions were activated when they were thinking of the American culture. Such a frequency distribution is significantly different from what is expected in the population, %2(2, N = 172) = 7.66, p < .025, and provides support to the hypotheses.

36

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

STUDY IB METHOD Participants To test the replicability of the results of Study 1A, the same experiment was conducted with a sample of Filipino American college students attending a regional conference for Filipino Americans. The participants were 78 (40 females) Filipino Americans with an average age of 20.02 years (SD=1.70). Forty-nine percent were bom in the United States. The reported median family income was $70,000. As in Study 1A, participants were randomly administered either the American or the Filipino WFC task. The Filipino condition consisted of 37 participants (43.2% female; average age of 20.14 years, *SZ>=1.89; 48.6% 2nd or later generations; 13.5% with college degree or higher; 78.4% had family incomes of $70,000 or above). The American condition consisted of 41 participants (58.5% female; average age of 19.93, 2500 ms) were observed from the current sample. As an indicator of the extent to which pleasant thoughts were activated by American-related terms (i.e., American+Pleasant facilitation score), mean latencies for the American prime+Pleasant target trials were subtracted from the mean latencies of the Neutral prime+Pleasant target trials (baseline pleasant activation). Similarly, mean latencies for the Filipino prime+Pleasant target trials were subtracted from the Neutral prime+Pleasant target trials to obtain a Filipino+Pleasant facilitation score. As an indicator of the Filipino rejection component of CM, a Filipino+Unpleasant facilitation score was computed by subtracting the mean latencies for the Filipino prime+Unpleasant target trials from the Neutral prime+Unpleasant target trials (baseline unpleasant activation). A similar formula was used to calculate an American+Unpleasant facilitation score (i.e., mean of American prime+Unpleasant target trials - mean of Neutral prime+Unpleasant target trials). The calculations of the facilitation scores are also consistent with conventional practice (e.g., Bargh, et al, 1992; Fazio, et al, 1986). Figure 1 pictorially presents the facilitation scores, with a negative score on the Y-axis indicating faster reaction times to target stimuli compared to baseline (i.e., response times to target stimuli after neutral primes) and a score in the positive direction indicating slower reaction times to target stimuli compared to baseline. As shown, participants’ reaction times were noticeably faster when responding to pleasant target terms after being exposed to American primes (American Preference Facilitation) than after being exposed to a neutral stimulus. Their

49

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

reaction times when responding to unpleasant targets after being exposed to American primes (American Rejection Facilitation) were not substantially slower than baseline (i.e., mean of Neutral+Unpleasant targets). The opposite pattern was observed when participants were exposed to Filipino primes in that their reaction times were noticeably faster when responding to unpleasant target terms (Filipino Rejection Facilitation) compared to their reaction times to unpleasant targets after being exposed to a neutral stimulus. Their reaction times to pleasant targets after being exposed to Filipino primes (Filipino Preference Facilitation) were not noticeably different from baseline (i.e., mean of Neutral prime+Pleasant targets). To test the reliability of such findings, a 2 (prime: American and Filipino) X 2 (target: Pleasant and Unpleasant) repeated measures analyses of variance (ANOVA) using the facilitation scores was conducted. Results revealed a statistically significant Prime X Target interaction, F (l, 25) = 6.03, p=.021, suggesting that the sample had reliably faster reaction times during Filipino prime+Unpleasant target trials than when unpleasant target terms followed American or Neutral primes, and that the sample had reliably faster reaction times during American prime+Pleasant target trials than when pleasant target terms followed Filipino or Neutral primes. None of the participants who completed the study in the laboratory correctly identified the hypotheses or purpose of the LDP task, suggesting that they were not able to use any deliberate cognitive strategy or expectancy that may have contaminated their response times. Furthermore, the sample’s error rate on the LDP task is low at 5.2%, suggesting that they were not responding randomly during the task and were putting forth good effort. Overall, these findings provide support to the notion that mere exposure to American-related stimuli automatically activates pleasant thoughts whereas mere exposure to Filipino-related stimuli automatically activates unpleasant thoughts, consistent with CM theory. Furthermore, such findings provide initial

50

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

evidence to the notion that CM involves an automatic association of pleasantness, superiority, and desirability to American culture and an automatic association of unpleasantness, inferiority, and undesirability to Filipino culture. Mental Health Correlates o f Colonial Mentality Based on previous theoretical works from Memmi (1965), Fanon (1965), Freire (1970), Strobel (2001), David and Okazaki (2006a), and others, as well as David and Okazaki’s (2006b) empirical findings, it was hypothesized that stronger American+Pleasant associations (faster reaction times when presented with American primes and pleasant target terms than when presented with Filipino primes and pleasant target terms), and stronger Filipino+Unpleasant associations (faster reaction times when presented with Filipino primes and unpleasant target terms than when presented with American primes and unpleasant target terms), will be related with higher CES-D scores and higher MASQ scores. The American Preference Index (API) was calculated by subtracting each participant’s American Preference Facilitation Score from their Filipino Preference Facilitation score. The Filipino Rejection Index (FRI) was calculated by subtracting each participant’s Filipino Rejection Facilitation Score from their American Rejection Facilitation score. Thus, positive scores on the API and FRI indicate either stronger American+Pleasant associations or stronger Filipino+Unpleasant associations, respectively, relative to the extent to which each individual may also have strong associations between Filipino+Pleasant and between American+Unpleasant. Although the facilitation scores are good indicators of the extent to which pleasant thoughts are activated by American-related stimuli and of the extent to which unpleasant thoughts are activated by Filipino-related stimuli, they are limited in some ways. The difference between the facilitation scores and the API/FRI is that facilitation scores are subtracted from

51

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

baseline/neutral reaction times and that they indicate activation of either pleasant or unpleasant thoughts, whereas APL/FRI indicate a stronger association of Filipino+Unpleasant than American+Unpleasant (FRI) and of American+Pleasant than Filipino+Pleasant (API). A facilitation score is an indicator of the activation of associated thoughts and, for some individuals, both American and Filipino stimuli may activate pleasant or unpleasant thoughts compared to baseline. In other words, an individual may have faster reaction times when presented with both American+Pleasant and Filipino+Pleasant trials than baseline, and an individual may also have faster reaction times when presented with both American+Unpleasant and Filipino+Unpleasant trials than baseline. However, although pleasant thoughts may be activated by both Filipino and American stimuli and unpleasant thoughts may be activated by both Filipino and American stimuli, the strength of the activation or association among individuals may still vary. For example, an individual may still have stronger associations (faster reaction times) between American+Pleasant than between Filipino+Pleasant even though both American and Filipino stimuli may activate pleasant thoughts within this person. Similarly, an individual may still have stronger associations between Filipino+Unpleasant than between American+Unpleasant even though both Filipino and American stimuli may activate unpleasant thoughts within this person. Thus, the facilitation scores are limited in their ability to account for the extent to which Filipinos may have stronger associations between American+Pleasant thoughts than between Filipino+Pleasant thoughts and between Filipino+Unpleasant thoughts than between American+Unpleasant thoughts. For these reasons, the API and FRI seems to be better measures of the relative strength of associations between American+Pleasant thoughts and between Filipino+Unpleasant thoughts.

52

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

The resulting intercorrelations between the variables are presented in Table 2. As shown, the API positively correlated with the FRI and both the API and FRI had positive correlations with the CMS subscales, CES-D, and the MASQ subscales, with most of the correlations reaching statistical significance. Such a pattern of results are consistent with theory and previous findings about CM (David, under review; David & Okazaki, 2006b), which provides support to the validity of the CMLDP task and the use of the API/FRI as indicators of CM. More importantly, the overall pattern of results are consistent with the hypothesis that stronger American+Pleasant associations and stronger Filipino+Unpleasant associations are negatively related with various mental health (e.g., CES-D, MASQ) variables.

53

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

STUDY 2 DISCUSSION Both Study 1 and Study 2 found that Filipino culture-related stimulus activates unpleasant, negative, undesirable, or inferior perceptions and that American culture-related stimulus activates pleasant, positive, desirable, and superior perceptions within some Filipino Americans. Such results suggest that unpleasant or inferior constructs are contained in, or closely linked with, the Filipino cultural knowledge system and that pleasant or superior constructs are contained in, or closely linked with, the American cultural knowledge system. The results are consistent with the Spreading Activation Theory and provide evidence to the notion that the centuries of oppression experienced by Filipino Americans has been internalized enough by some so that negative thoughts are automatically activated by mere exposure to Filipino-related stimuli and positive thoughts are automatically activated by mere exposure to American-related stimuli. Such results also suggest that there might be automatic links or associations between Filipino and unpleasant concepts and between American and pleasant concepts among some Filipino Americans. The results of Study 2 also provide further evidence that CM, as approximated by stronger American+Pleasant associations than Filipino+Pleasant associations and by stronger Filipino+Unpleasant associations than American+Unpleasant associations, is related with mental health and mental health-related variables with CM possibly contributing to higher levels of distress and lower levels of psychological well-being. However, given the small sample of Study 2 and the several non-significant relationships between the CM measures and mental health variables, the reliability of such findings remains unclear. Thus, Study 3 attempted to further test CM theory’s contention: (1) that some Filipino Americans have associated Filipino culture with inferiority, undesirability, and unpleasantness and American culture with superiority, desirability, and pleasantness; and (2) that CM is related to lower levels of

54

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

psychological well-being and mental health. Study 3 tested CM theory by using a method specifically developed to measure automatic associations - the Implicit Association Test (IAT).

55

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

STUDY 3

In addition to supplementing the findings from Study 1 and 2 that (1) CM may indeed be assessed using indirect methodologies, that (2) CM may operate automatically without awareness, intention, or control, and that (3) CM is negatively related to psychological well­ being and mental health, Study 3 also attempted to more directly test the theory that CM involves an “automatic and uncritical rejection of anything Filipino and an automatic and uncritical preference for anything American” (David & Okazaki, 2006b, p. 241) by testing if Filipino Americans have developed strong associations between anything Filipino and unpleasantness and between anything American and pleasantness using the IAT paradigm. Furthermore, Study 3 also attempts to investigate the relationships between an indirect measure of CM (i.e., the colonial mentality IAT) and theoretically-related constructs such as collective self-esteem, personal self-esteem, ethnic identity, acculturation, and life satisfaction. Based on previous studies (David, under preparation; David & Okazaki, 2006b), it was hypothesized that CM as measured by the IAT will be positively correlated with increased acculturation and negatively correlated with personal self-esteem, collective self-esteem, ethnic identity, and life satisfaction. Such findings will support the validity of using the IAT to measure the CM construct and provide additional evidence to the notion that CM may be related to mental health and mental health related variables.

56

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

STUDY 3 METHOD

Participants In order to obtain enough Filipino American participants and obtain implicit data from non-college students, data was collected over the Internet and from the Psychology-100 Subject Pool. This also allowed for a comparison between the two sampling methodologies for possible systematic differences. The sample composed of 44 Filipino Americans (24 females), of whom 32 were obtained from the Internet and 12 were from an introductory psychology course. The average age of the sample was 29.78 years-old (SD=6.9) and 22.7% reported being 1st generation Filipino Americans (the rest were 2nd or later generations). In terms of socioeconomic status, 63.6% of the sample had a high school degree and an additional 31.8% reported having at least a college degree. The median personal income was $12,000. Beside average yearly income, there were no statistically significant differences between the two sources of participants in terms of their demographic characteristics. The income differences between the two samples, however, were expected given that the introductory psychology sample were composed of college students of whom many reported no yearly income. A series of zero-order correlations between income and the other measures revealed no statistically significant relationships, suggesting that the income differences between the two sources of participants were not consequential in this case. In terms of the other measured variables, there were no statistically significant differences between the two samples. Among the entire sample, there were no statistically significant relationships between age and all the measured variables, as well as between income and all the measured variables. Finally, generational status and sex did not have significant effects on any of the measured variables.

57

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Measures The Implicit Association Test (IAT): As the implicit method of assessing CM, Study 3 used the IAT as administered through the Inquisit 2.0 Web Edition Software. The IAT allows for a direct test of hypothesized implicit attitudes or associations that individuals may have. The Colonial Mentality IAT (CMIAT) was composed of five blocks, with the first block involving PLEASANT-UNPLEASANT discrimination, the second block being the FILIPINOAMERICAN discrimination, the third block containing FILIPINO+PLEASANT or AMERICAN+UNPLEASANT discrimination, the fourth block being the reverse key stage (switching the keys to press for PLEASANT and for UNPLEASANT), and the fifth block being the FILIPINO+UNPLEASANT or AMERICAN+PLEASANT discrimination. According to Nosek, Greenwald, and Banaji (2005), the most effective IATs have at least four stimulus items for each category and each stimulus item must be clearly identifiable as belonging to one of the superordinate categories. Therefore, I chose a total of 40 terms (10 for each category) that are easily and clearly identifiable as belonging to either the Filipino or American culture and to either Pleasant or Unpleasant categories. For the Filipino and American cultural categories, I chose terms that represent major cultural components such as language, name of country, capital city, popular food, name of current president, popular historical figure, popular landmark, popular geographic locations, and skin color commonly associated with people from each culture. Four of the five terms used for each cultural category in Study 1 and Study 2 (i.e., language, country, capital city, and skin color) were used again for Study 3. The terms “American” and “Filipino” were not re-used for Study 3 because these were the terms used as superordinate cultural category labels for the CMIAT. I used the same pleasant and unpleasant terms as Study 2, with four of the unpleasant terms (i.e., embarrassing, shame, inferiority, and

58

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

sadness) and four of the pleasant terms (i.e., beautiful, attractive, intelligent, superiority) referring to attributes that are believed to be particularly indicative of CM. The rest of the pleasant and unpleasant terms were selected from the norms presented by Bellezza, Greenwald, and Banaji (1986). For a list of the IAT stimulus items used in Study 3, see Appendix C. The Rosenberg Self-Esteem Scale (RSES): One of the outcome variables for Study 2 was the RSES (Rosenberg, 1965) as the measure of personal self-esteem. It is composed of 10 items with possible responses for each item ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores indicating more positive evaluations of one’s personal characteristics. The RSES has produced test-retest reliability coefficients ranging from .81-.88, and Cronbach’s alphas ranging between .77-.88 (Blascovich & Tomaka, 1993). Various studies have also supported the construct and convergent validities of the RSES (e.g., Hagborg, 1993). An alpha of .84 was observed from the current sample. The Collective Self-Esteem Scale (CSES): Another outcome variable is the CSES (Luhtanen & Crocker, 1992). The CSES is a measure of the extent to which individuals generally evaluate their social groups positively. It is composed of 4 subscales consisting of 4 items each. The 4 subscales are Private CSE (e.g., “In general, I am glad to be a member of my racial/ethnic group”), Public CSE (e.g., “Overall, my racial/ethnic group is considered good by others”), Importance to Identity (e.g., “Overall, my race/ethnicity has very little to do with how I feel about myself’), and Membership Esteem (e.g., “I am a cooperative participant in the activities of my racial/ethnic group”). Participants are asked to rate their level of agreement for each item using a 7-point Likert-type scale (1 = strongly disagree to 7 = strongly agree). For the current study, the race specific version of the CSES will be used. Coefficient alphas for this version of the CSES have been reported to be .72 (Private), .88 (Public), .84 (Importance to Identity), and

59

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

.75 (Membership Esteem) (Crocker, Luhtanen, Blaine, & Broadnax, 1994). For the current sample, alphas of .94 (Private), .88 (Public), .82 (Importance to Identity), and .86 (Membership Esteem) were observed. The Vancouver Index o f Acculturation (VIA): To test concurrent validity of the LDP task, the VIA (Ryder, Alden, & Paulhus, 2000) was used. The VIA is a 20-item self-report measure developed to independently assess individuals’ levels of identification to the mainstream and their heritage cultures. The VIA includes items such as “I often participate in mainstream American cultural traditions” (mainstream subscale) and “I would be willing to date a person from my heritage culture” (heritage Subscale). Participants were asked to rate their level of agreement to each item on a 9-point Likert-type scale (1 = strongly disagree to 9 = strongly agree). Internal consistency coefficients of .91 (Chinese Americans) and .92 (East Asians) for the heritage subscale were reported in the initial validation of the VIA. The Heritage subscale was also reported to have mean inter-item correlations of .52 (Chinese Americans) and .53 (East Asians). The mainstream subscale produced internal consistency coefficients of .89 (Chinese Americans) and .85 (East Asians), and mean inter-item correlations of .45 (Chinese Americans) and .38 (East Asians). Concurrent validity of the VIA was supported by its theoretically consistent correlations with generational status, percentage of time lived in a Western/English speaking country, and percentage of time educated in a Western/English speaking country, among others (Ryder, Alden, & Paulhus, 2000). For the current sample, alphas of .93 (Heritage) and .87 (Mainstream) were observed. The Multidimensional Ethnic Identity Measure (MEIM): To further test construct validity of the LDP task, as well as to simultaneously explore the implications of CM on ethnic identity development, the MEIM (Phinney, 1992) was used. The MEIM is a popular measure of ethnic

60

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

identity development that has shown good reliability and validity (Roberts, Phinney, Masse, Chen, Roberts, & Romero, 1999). Higher scores on the MEIM indicate higher levels of ethnic identity development. The current sample had an alpha of .95. The Satisfaction With Life Scale (SWLS): The SWLS (Diener, Emmons, Larsen, & Griffin, 1985) was used as the measure of life satisfaction and indicator of psychological well­ being. It is a widely used five-item measure that has been shown to have good concurrent and discriminant validity, as well as excellent reliability (Pavot & Diener, 1993). Each item is scored on a 7-point scale with higher scores indicating higher levels of life satisfaction. An alpha of .82 was observed from the current sample. Finally, the self-report CM and depression measures administered during Study 2 were also used for Study 3. More specifically, the CMS (sample alphas: Intlnferior = .85; CulturalShame = .86; PhysChar = .89; WGDiscrim = .88; CD = .82) was used as the direct method of assessing CM, the MASQ (sample alphas: Anhedonia = .91; GD: Depression = .96) was used as the measure of depression, and the CES-D (sample alpha = .94) was used to assess general depression symptoms. Procedures The same Internet recruiting procedure described for Study 1 and the Psychology-100 subject pool procedure described in Study 2 were used for this study. Internet recruiting clearly stated that one has to have Filipino lineage to participate and that individuals who have participated in previous internet studies about Filipino American psychology cannot participate. In both participant sources, the study was administered through a computer. The only difference between the two samples is that the Internet participants knew beforehand that the study was about Filipino culture whereas subject pool participants did not. The order of measurement was

61

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

not counterbalanced. Instead, all participants completed the CMIAT before completing the explicit self-report measures. In addition to recent findings that implicit-explicit measurement order has no effect on participants’ performance (Nosek, 2005; Nosek, Greenwald, & Banaji, 2005), it was preferred to present the CMIAT before the other measures in order to prevent the participants from becoming aware of the purposes and intentions of the study if they completed explicit self-report measures first (e.g., the CMS). However, because recent findings continue to find block order effects in IAT performance (i.e., compatible first then incompatible second, or vice versa), the presentation of the compatible and incompatible blocks in the CMIAT was counterbalanced. That is, some participants were presented the compatible block first before being presented with the incompatible block, whereas some participants completed the blocks in the opposite order. Study 3 Hypotheses Based on previous theoretical works from Memmi (1965), Fanon (1965), Freire (1970), Strobel (2001), David and Okazaki (2006a), and others, as well as David and Okazaki’s (2006b) empirical findings, I hypothesized that, on average, reaction times during the FILIPINOUNPLEASANT and AMERICAN-PLEASANT block (compatible) will be faster (lower latencies) than the FILIPINO-PLEASANT and AMERICAN-UNPLEASANT block (incompatible). Such results will support David and Okazaki’s (2006a; 2006b) contention that some Filipino Americans may hold CM and that CM involves an automatic association of unpleasantness to anything Filipino and an automatic association of pleasantness to anything American. It was also hypothesized that Filipino Americans with faster reaction (lower latencies) times during the compatible block (i.e., AMERICAN-PLEASANT and FILIPINOUNPLEASANT) and slower reaction (higher latencies) times during the incompatible block (i.e.,

62

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

FILIPINO-PLEASANT and AMERICAN-UNPLEASANT), will have lower levels of personal self-esteem, collective self-esteem, and satisfaction with life. These participants were also predicted to have higher scores on the CES-D (depression) and the MASQ subscales. It was also predicted that such individuals will have lower levels of ethnic identity development using Phinney’s (1992) MEIM. Such results will provide further evidence that CM may lead to lower levels of psychological well-being, mental health, and ethnic identity development. It was also predicted that both the implicit (CMIAT) and explicit (CMS) methods of assessing CM will correlate consistently with theory. Specifically, I predicted that theoretically consistent correlations between the CM measures and VIA, CSES, and MEIM will be observed. Such results will provide evidence that both the CMIAT and CMS are valid measures of CM. Furthermore, it was predicted that the correlation between CMS and CMIAT will be moderate to low, not only because explicit and implicit measures of the same construct have consistently been shown to be minimally correlated, but because even though they are measures of the same construct, one method may be better at assessing different aspects of CM than the other. More specifically, the CMIAT may be better at tapping into the more covert aspects of CM than the covert subscales of the CMS. On the other hand, the CMS has the ability to assess the more overt aspects of CM (e.g., within-group discrimination, changing physical characteristics, colonial debt) that the CMIAT cannot and does not attempt to directly assess. Relatedly, it was predicted that a combination of both implicit (CMIAT) and explicit (CMS) measures of CM will best predict mental health outcome variables.

63

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

STUDY 3 RESULTS AND DISCUSSION Colonial Mentality as Implicit Associations Computation of the IAT scores followed the improved scoring algorithm proposed by Greenwald, Nosek, and Banaji (2003). These researchers evaluated multiple alternative scoring algorithms for the IAT in terms of their ability to: (1) resist confounds associated with response speed; (2) correlate with self-report measures; (3) obtain internal consistency; and (4) resist procedural confounds. They found that the best performing algorithm involves eliminating responses > 10,000 ms, eliminating participants who have more than 10% of their responses below 300 ms, computing the mean of correct latencies for each block, replacing error latencies with the block mean plus 600 ms, and averaging the resulting values for each block. To compute the IAT effect size for the entire sample - or D600 score - the differences between the mean latencies of the incompatible and compatible blocks were divided by the pooled standard deviation of both blocks (Greenwald, et al, 2003). In the present study, only two trial responses had latencies of > 10,000 ms and these responses were eliminated. Furthermore, no participant had more than 10% of their responses being below 300 ms. The sample’s error rate was low at 10.2%, suggesting that participants were not randomly responding during the IAT task and were putting forth good effort. In addition to the sample IAT effect (i.e., D600), individual IAT effect scores were also calculated for each participant (i.e., d600). A participant with a positive d600 score indicates stronger AMERICAN-PLEASANT and FILIPINO-UNPLEASANT associations than AMERICAN-UNPLEASANT and FILIPINO-PLEASANT associations, a pattern that is consistent with David and Okazaki’s (2006a; 2006b) conceptualization of CM. Thus, for the current study, the individual d600 score will be referred to as the CMIAT score. Finally, because the mean latencies for both the incompatible and compatible blocks were skewed, these latencies

64

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

were log transformed (Greenwald, McGhee, & Schwartz, 1998) to satisfy the normal distribution assumptions of consequent statistical analyses. The mean reaction times for the CMIAT are pictorially presented in Figure 2. Twentyfive participants had faster reaction times (i.e., lower latencies) during the compatible blocks (i.e., AMERICAN-PLEASANT and FILIPINO-UNPLEASANT) than during the incompatible blocks (i.e., AMERICAN-UNPLEASANT and FILIPINO-PLEASANT), producing positive CMIAT scores for these participants and suggesting that these individuals have associated pleasantness, desirability, or superiority with American culture and unpleasantness, undesirability, or inferiority with the Filipino culture. Also, the average latency (i.e., reaction time) of the entire sample for the compatible block (M = 1273.93 ms, SD = 486.00) is significantly faster than their average latency for the incompatible block (M= 1456.34 ms, SD = 447.25), IAT Difference = 182.40 ms (log = 0.07), t{43) = 2 3 1 ,p = .02, D600 = 0.39, suggesting that, on average, Filipino Americans have strongly associated PLEASANT with American and UNPLEASANT with Filipino. Colonial Mentality and Mental Health Variables A series of bivariate correlations were conducted to explore the relationships between the CMIAT scores (d60Q) and other psychological variables, including mental health. The results show that participants’ individual CMIAT scores had theoretically consistent relationships with other constructs, as presented in Table 3. More specifically, the CMIAT score had significant positive correlations with all the CMS subscales, providing support for the concurrent validity of the CMIAT. Moreover, the correlations between the CMIAT score and the CMS subscales ranged from .31 (CMIAT and CD) to .49 (CMIAT and WGDiscrim), suggesting that at the most, the CMIAT and the CMS share only about 24% of their variance. This provides support to the

65

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

notion that the CMS and the CMIAT may be tapping into different aspects of CM. The CMIAT score also had significant negative correlations with the CSES subscales and the MEIM, indicating that higher levels of CM is related to more negative evaluations of one’s ethnic group and lower levels of ethnic identity development, respectively. Finally, the CMIAT score positively correlated with mainstream acculturation and negatively correlated with heritage acculturation, suggesting the higher levels of CM is related to assimilation. These correlational patterns, which are consistent with theory and previous findings concerning CM (David & Okazaki, 2006b), provide further support for the construct validity of the CMIAT and suggest that CM is associated with mental health-related constructs. Table 3 also presents the interrelationships between the CMIAT score and various indices of mental health and psychological well-being. More specifically, the CMIAT score positively and significantly correlated with CES-D, GD: Depression, and Anhedonia. Such results suggest that higher levels of CM as measured by the CMIAT score is related to more experiences of general distress and depression symptoms. Furthermore, the CMIAT score also had significant negative correlations with personal self-esteem and the SWLS, suggesting that higher levels of CM is related to more negative evaluations of one’s personal characteristics and lower levels of life satisfaction, respectively. As also presented in Table 3, similar patterns of correlations were observed between the CMS subscales and the mental health and psychological well-being variables, with most relationships reaching statistical significance. Overall, the patterns of observed correlations between the CM measures (CMIAT and CMS) and the various indices of mental health and psychological well-being suggest that CM is negatively related to Filipino Americans’ mental health and psychological well-being.

66

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

It was also hypothesized that a combination of the CMIAT score and the CMS scores would be a better predictor of depression than either measure by itself. To test this prediction, a hierarchical multiple regression was conducted for the depression index (a composite score comprised of the sum of CES-D, Anhedonia, and GD: Depression scores). A total CMS score was calculated for each participant by summing the five CMS subscale scores. The total CMS score was entered on the first step, the CMIAT score was entered on the second step, and the interaction term of the CMIAT score and the total CMS score (i.e., CMIAT*CM interaction) was entered on the last step. All predictors were centered for the regression analysis and prior to the computation of the interaction term. The results of the final step of the regression are presented in Table 4, showing that significant R2 statistics were obtained for each predictor. The significant R2change statistic observed for the CMIAT in the second step suggests that this measure of CM predicts unique variance in depression symptoms that are not accounted for by the CMS. Also as hypothesized, results showed that the CMIAT*CM interaction term significantly predicted depression symptoms above and beyond the separate effects of the CMS and the CMIAT scores on depression symptoms. Such results suggest that an implicit or indirect measure of CM (i.e., the CMIAT), combined with a direct measure of CM (i.e., the CMS), predicts variance (approximately 11.4%) in depression symptoms that is unique from the variance predicted by the CMS and the CMIAT separately. In terms of the interaction between the CMIAT and the CMS, results suggest that the overtly reporting CM manifestations as measured by the CMS are positively correlated with depression symptoms only for participants with high CMIAT scores but not for participants with low CMIAT scores (Figure 3). Tests of simple slopes of the lines showed that increased CMS scores was significantly associated with higher depression symptoms as a function of higher

67

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

CMIAT scores, /3 = .679, t (25) = 4.44,/? < .001. Depression symptoms did not change significantly as a function of increased CMS scores alone, /3 = -.327, t{ 19) = -1.42,/? = .17, ns. Thus, it seems like overt reports of CM by themselves are not sufficient to predict depression symptoms. Instead, a combination of overtly reporting CM manifestations and implicitly associating inferiority, shame, and embarrassment with the Filipino culture is necessary to predict higher levels of distress and depression.

68

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

POST-HOC TEST OF STIMULI Because the WFC, LDP, and LAT studies that were previously presented utilized stimuli that had not been validated for addressing the current research questions with a Filipino American sample, the validity of the results remain in question. More specifically, it is unclear if Filipino Americans clearly and easily identify the stimuli as belonging to the Filipino, American, Pleasant, or Unpleasant categories. Thus, a post-hoc test of the stimuli used for the studies were conducted to systematically determine (1) what Filipino and American cultural terms are commonly thought of by Filipino Americans and if they: (2) equally regard the Filipino and American terms as pleasant or unpleasant; (3) clearly identify the Pleasant and Unpleasant terms as pleasant or unpleasant, respectively; (4) clearly identify the Filipino and American terms as Filipino or American, respectively; and (5) find all the terms equally easy and understandable. Methods Participants and Procedures Twenty Filipino American college students (15 female; 17 2nd or later generations) participated in the study during a regional Filipino American conference. The sample’s average age was 19.65 years {SD-1.04). These individuals reported that they have not participated in a study about “Filipino American Psychology” before or over the past year. Interested individuals were told that the study involved completing a short questionnaire about Filipino American psychology. Upon agreeing to participate, participants were given a short paper and pencil questionnaire that is divided into four parts. Part 1 involved an open-ended instruction that asked participants to think of Filipino or American cultural terms. The instructions were:

69

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

When people think of a category, they usually also think about other things that belong in that category. For example, when people think of the category “FURNITURE,” words such as chairs, tables, lamps, television, and others also come to their minds. When people think of the category “ANIMALS,” terms such as dogs, lions, cats, and other animals also come to their minds. Relatedly, when people think of the category “Mexican,” some people might also think about Mexico (country), Spanish (language), tacos (food/dishes), Mexico City (capital city), Guadalajara (another major city), Cesar Chavez (famous Mexican American leader), and other things or people that may be associated with the Mexican culture. We would like to ask you to think of the two categories below and try to list up to 20 words or terms that come to your mind when you think of each category. The first category is Filipino and the second category is American. Part 2 of the questionnaire asked participants to rate each of the stimuli used in Studies 1-3 in terms of the extent to which they elicit pleasant or unpleasant feelings or thoughts (i.e., PLEASANTNESS) using a 7-point scale (l=Very Unpleasant to 7=Very Pleasant). Part 3 asked participants to rate each of the stimuli in terms of how much they fit in the category of Filipino culture or American culture (i.e., FILDPINO-NESS) using a 7-point scale (l^Very American to 7=Very Filipino). Finally, Part 4 asked participants to rate each of the stimuli in terms of how easy it is for them to understand its meaning (i.e., EASINESS) using a 7-point scale (l=Very Difficult to 7=Very Easy). Post-hoc Test of Stimuli Results and Discussion Part 1 results revealed that that the participants identified 173 American-related terms. The 30 most named American terms are presented in Table 5, showing that 10 of the 11 American terms used in the WFC, LDP, and IAT studies were among the top 12 most frequently named terms. Only the term “Hollywood” from the LDP and IAT studies were not in the Top 30 most frequently named American terms. In terms of Filipino-related terms, the participants named 163 different terms. As shown in Table 5,10 of the 11 Filipino terms used in the WFC, LDP, and IAT studies were in the top 13. Only the term “Asia” (16th) did not make it on top of the word frequency tabulations. Overall, such results suggest that most of the American and

70

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Filipino terms that were used in the WFC, LDP, and IAT studies were commonly known to Filipino Americans and that many Filipino Americans easily think of such terms when they think of either the Filipino or American culture. The PLEASANTNESS, FILIPINO-NESS, and EASINESS rating means and standard deviations of the stimuli used in studies 1-3 are presented in Table 6. A series of within-subjects T-tests revealed that there was a statistically significant difference between the pleasantness ratings of American and Filipino terms, /(19) = -2.68, p = .01, with the Filipino terms being rated as more pleasant on average than the American terms. Such results are surprising given that CM theory and the WFC, LDP, and IAT results suggest that Filipino Americans should rate Filipinorelated stimuli as being less pleasant than American-related stimuli, if there are any differences at all. More importantly, however, such findings suggest that even though Filipino Americans may explicitly rate Filipino-related stimuli as more pleasant than American-related stimuli, more indirect and implicit methods of assessment may suggest otherwise. These results further highlight the benefits of using implicit methods in assessing CM and other attitudinal evaluations that may be susceptible to social desirability or other cognitive strategies. As expected, the pleasant and unpleasant terms used in studies 1-3 statistically differed in terms of PLEASANTNESS ratings, f(19) = 18.77,/? < .001, suggesting that Filipino Americans clearly perceive the two sets of stimuli as either pleasant or unpleasant. A statistically significant difference was also observed in the FILIPINO-NESS ratings of the American and Filipino stimuli, t(19) = -36.98,p < .001, suggesting that the sample clearly identified the two sets of stimuli as either Filipino or American. Such a statistically significant difference on the FILIPINO-NESS ratings was not observed between pleasant and unpleasant sets of stimuli, t( 19) = -1.27,/? = .22, suggesting that Filipino Americans did not explicitly report associating pleasant

71

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

or unpleasant terms with either the Filipino or the American culture. Finally, no statistically significant differences were found in terms of the EASINESS ratings between the pleasant and unpleasant stimuli, t( 19) = -0.42, p = .68, and between the Filipino and American stimuli, t(19) = 1.83,p = .83, suggesting that the sample found all the stimuli as equally easy to understand. Overall, the results indicate that Filipino Americans: (1) easily and commonly think of the Filipino and American cultural terms used in the WFC, LDP, and IAT tasks; (2) do not explicitly rate the Filipino-related stimuli used in the WFC, LDP, and IAT tasks as more unpleasant than American-related stimuli; (3) clearly identify the pleasant and unpleasant stimuli used in the WFC, LDP, and IAT tasks as being pleasant or unpleasant; (4) clearly identify the Filipino and American stimuli used in the WFC, LDP, and IAT tasks as belonging to either the Filipino or American culture; and (5) find all the pleasant, unpleasant, Filipino, and American stimuli used in the WFC, LDP, and IAT tasks as equally easy to understand. Such findings provide support to the validity of the WFC, LDP, and IAT tasks and further strengthen the arguments of CM theory.

72

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

GENERAL DISCUSSION Cognitive Operation o f CM Among Filipino Americans The current set of studies had two main goals: (1) explore the cognitive operation of CM among Filipino Americans by using implicit social cognition methodologies; and (2) investigate the mental health implications of CM among this ethnic group. Based on the findings, four main conclusions may be made concerning the cognitive operation of CM among Filipino Americans. First, colonialism or oppression has been deeply internalized by some Filipino Americans such that they now have a CM-consistent cultural knowledge system. Second, the findings suggest that CM among Filipino Americans may be activated using priming techniques (Study 1 and Study 2). Third, CM among Filipino Americans may be automatically activated by mere exposure to either Filipino- or American-related stimuli (Study 2). Lastly, CM among Filipino Americans has been deeply internalized such that Filipino-related stimuli are automatically associated with ideas of unpleasantness or inferiority and American-related stimuli are automatically associated with ideas of pleasantness or superiority (Study 2 and Study 3). Taken together, these conclusions are consistent with David and Okazaki’s (2006b) contention that CM among Filipino Americans involves a covert component, that CM is characterized by an inferior perception of anything Filipino and superior perception of anything American, and that CM may operate automatically without awareness, intention, or control. The findings regarding the cognitive operation of CM highlights the potential contextual influences on how some Filipino Americans may behave in various contexts. For example, if a Filipino American individual who may have CM is in a context wherein he/she is with several other Filipinos who speak English with strong Filipino accents or are speaking a Filipino dialect, his/her CM-consistent cognitions may be activated. Such cognitive activation, in turn, may lead

73

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

the individual to behave in ways that are consistent with CM such as discriminating or distancing him/herself from the other Filipinos, especially if the activation of the CM-consistent cognition is not further scrutinized (i.e., cognitive elaboration) and the person is acting spontaneously. Indeed, a review by Gawronksi and colleagues (in press) of the literature on implicit social cognition suggest that activated associations predict spontaneous behaviors. Also, in a context wherein a Filipino person who may have CM is faced to make a decision between a Filipino product (or person) or an American product (or person), the Filipino person’s CM-consistent cognitions may become activated and lead the person to choose or prefer the American product (or person). This latter example is especially salient in mere observations of modem day Philippines, where Western companies and products continue to promulgate and flourish over local ones as Filipinos continue to prefer and support anything Western over anything Filipino. A note should be made regarding the conscious accessibility (i.e., awareness) of such automatic cognitions. Although such an interpretation is consistent with the conventional conceptualization of cognitions or attitudes (e.g., Greenwald & Banaji, 1995) that are reflected by implicit methods such as Word Fragment Completion (WFC), Lexical Decision Priming (LDP), and Implicit Association Tests (IAT), recent developments in the social cognition literature have challenged the common assumption that implicit cognitions or attitudes are unconscious or introspectively inaccessible. Instead, new empirical evidence (e.g., Gawronski, Hofmann, & Wilbur, 2006) suggests that “most associations are consciously accessible” and “are used for explicit judgments” (Gawronski, LeBel, & Peters, in press, p. 21). Consequently, an alternative conceptualization proposed by Gawronski and his colleagues (e.g., Gawronski & Bodenhausen, 2006; Gawronski, LeBel, & Peters, in press) argue for a model that does not depend on such an unconscious assumption and one that instead highlights and distinguishes

74

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

between activation and validation cognitive processes. These researchers argue that indirect or implicit measures of attitudes, such as the WFC, LDP, and IAT tasks, reflect the activation of learned associations in a person’s memory. Validation, on the other hand, is reflected by selfreport measures such as the CMS and is the process of assessing the subjective truth or falsity of the activated associations. They further explained that activation of associations in a particular person’s cognitive system may take place independent of the result of the person’s validation process; that is, regardless of whether the person considers such activation as accurate (true) or inaccurate (false). Using the activation-validation model, some Filipino Americans may disagree with the cognitive associations that are activated within them when they are faced with Filipinoor American-related stimuli and, thus, do not endorse items on the CMS. However, as Gawronski and colleagues (in press) have argued, rejecting a particular activation or association as false does not necessarily erase the activation or mean that the activation did not take place. Thus, although some Filipino Americans may not endorse CMS items as a result of their cognitive validation, a CM-consistent cultural knowledge system may still have been activated and still exist. The opposite pattern, though less likely, may also take place in that some Filipino Americans may endorse CM items as a result of their cognitive validation even when they do not hold a CM-consistent cultural knowledge system. Nonetheless, regardless of whether one subscribes to the conventional conceptualization of unconscious (or introspectively inaccessible) cognitions or to the alternative activation-validation model, the current set of studies suggest that a CM-consistent cultural knowledge system exists and may be activated within some Filipino Americans, and that CM may operate automatically without intention or control.

75

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Given that CM or a CM-consistent cultural knowledge system may exist or be activated even when some Filipino Americans may not openly acknowledge it, and that activation of learned associations may influence consequent spontaneous behaviors, introspection- or validation-dependent (Gawronski, LeBel, & Peters, in press) explicit methods (i.e., self-reports) of assessing this construct may be limited. Indeed, using self-report methods (i.e., the CMS), previous research have found that Filipino Americans less frequently endorse feeling inferior and embarrassed for being Filipino (covert manifestations of CM) than the other more overt and obvious CM manifestations (e.g., desiring to look more white, discriminating against FOBs, etc.). It is possible that many Filipino Americans may not be able to accurately identify (or may not be able to explicitly admit) feelings of inferiority or embarrassment in relation to their heritage ethnicity and culture. Thus, more implicit or indirect methods of assessing internalized oppression (e.g., CM) may be useful in more accurately identifying the extent to which one may have internalized the oppression he/she have experienced. Using the IAT as a measure of the extent to which the concepts of “inferior” and “embarrassing” are strongly associated with the Filipino culture, the reported studies demonstrated that indirect methods of assessing the covert aspects of CM (i.e., CMIAT) had satisfactory validity as evidenced by the strength (i.e., moderate relationship with the CMS) and directionality of their relationships (e.g., ethnic identity, collective self-esteem, etc.) with other constructs. Furthermore, indirect measures of CM (i.e., the CMIAT), both by itself and in combination with more direct measures of CM (i.e., the CMS), were found to predict unique variance of general distress and depression symptoms separate from the variance predicted by the CMS alone, providing support to the research and clinical utility of using indirect assessment methods to approximate the CM construct.

76

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Additionally, preliminary evidence suggest that the Filipino, American, Pleasant, and Unpleasant stimuli used in the WFC, LDP, and IAT studies are valid in that Filipino Americans are able to easily and clearly identify the stimuli as belonging to either the Filipino, American, Pleasant, or Unpleasant categories. Taken together, the results of the current set of studies provide support to the utility and validity of using indirect and implicit methods such as WFC, LDP, and IAT in studying CM among Filipino Americans. Future studies may explore the utility and validity of such methodologies in studying the frequency, cognitive operation, and mental health implications of internalized oppression among other historically oppressed groups such as African Americans, Native Americans, and Hispanic Americans. Another interesting finding is the significant interaction effect of the CMIAT and the CMS in predicting depression symptoms. The results suggest that Filipino Americans who overtly display manifestations of CM such as discriminating against less-Americanized Filipinos, desiring to look and act more white, and feeling indebted to their past colonizers and tolerating the oppression they face only experience increased depression symptoms if they also feel ashamed, embarrassed, or inferior for being Filipino. It is possible that those who discriminate against less-Americanized Filipinos or desire to look more white may indeed be more Americanized or more “white looking,” which might allow them to better “fit-in” and function in the general society, which in turn, may contribute to better well-being. In other words, those who discriminate against less-Americanized Filipinos, desire to look more white, or feel indebted to their past colonizers might have such attitudes and behaviors function as protective factors against the stressors that they face by protecting their self-esteems and minimizing the stress appraisals of the oppression they experience. Furthermore, the finding that CM does not seem to be associated with increased depression symptoms without the accompanying feelings of shame,

77

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

embarrassment, and inferiority also speaks to the experiences of some Filipino Americans who report that although they discriminate against less-Americanized Filipinos, desire to be more White, or feel fortunate to be in the U.S., they are still proud to be Filipino and do not experience depression symptoms. Nevertheless, such overt displays of CM may still be detrimental to the community as a whole as it creates division within the community and prevents Filipino Americans from speaking out against oppression. Furthermore, such overt acts of CM may also be detrimental to future generations of Filipino Americans who witness such behaviors, which in turn, may contribute toward the development of feelings of shame, embarrassment, and inferiority for being Filipino. Mental Health Implications o f CM Among Filipino Americans The current studies’ findings provide support for CM theory and its ability to explain Filipinos’ and Filipino Americans’ preference for and deference toward anything American, and their tendency to have inferior perceptions and attitudes toward the Filipino culture and ethnicity. More importantly, however, the current findings help illuminate the alarming mental health consequences of colonialism, ethnic/cultural subjugation, and more generally, oppression, in that individuals who are members of historically and contemporarily oppressed groups may eventually internalize the oppression they experience in such a deep way that it creates within them a cultural knowledge system that is characterized by automatic negative cognitions and perceptions of their heritage ethnicity and culture. Consistent with cognitive behavioral theories on psychopathology (e.g., Beck, Ruch, Rush, Emery, & Shaw, 1979), underlying such automatic thoughts, attitudes, (e.g., “Lighter skin is more attractive or desirable.”) or behaviors (e.g., making fun of Fresh-Of-the-Boats or less Americanized Filipinos) are maladaptive general beliefs (e.g., “Being White or American is better than being Filipino.”) that have been developed

78

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

from previous experiences (i.e., colonialism and contemporary oppression). A combination of such thoughts and beliefs contribute to the creation of dysfunctional self-schemas (e.g., “I’m Filipino, therefore I am not attractive and I am inferior to Whites.”) that may lead to symptoms of psychological distress and depression. For Filipino Americans, the centuries of ethnic and cultural subjugation (both historical and contemporary) they have experienced seem to have created a general assumption or belief that anything Filipino is inferior to anything American and that such a belief may underlie the automatic thoughts, attitudes, and behaviors that many Filipino Americans display today (e.g., Bergano & Bergano-Kinney, 1997; David & Okazaki, 2006a; Lott; 1976; Pido 1997). Such inferiorizing automatic cognitions, attitudes, and behaviors toward the Filipino ethnicity and culture, which is a part of Filipino Americans’ self-concept, may be damaging to their self-esteem and contribute to the development of various psychopathology such as depression. Another troubling implication is that centuries of oppression seem to have damaged the minds of the oppressed in such a deep way that the self-denigration may become so commonplace and be accepted as “normal,” deserved by the oppressed, part of the culture, and exist without impact awareness (Gawronski, Hofmann, & Wilbur, 2006) or acknowledgement by the oppressed. The high correlations between the indirect measures of CM (i.e., LDP and IAT) and direct measures of CM (i.e., the CMS) suggest that Filipino Americans may view CMconsistent behaviors as socially-acceptable both by the mainstream society and the Filipino community. Indeed, Gawronski and colleagues (in press) have provided a comprehensive review of the literature on explicit and implicit methods and concluded, among other things, that high concordance between explicit and implicit methods of assessing social attitudes partly depends on whether the attitudes are perceived as socially acceptable or desirable. Furthermore, the high

79

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

correlations observed between the indirect methods of assessing CM and the direct method of assessing CM suggest that Filipino Americans do not think too much of such CM-consistent attitudes and behaviors (i.e., cognitive elaboration), consistent with the findings that correlations between the two kinds of measures tend to decrease as a function of increased cognitive elaboration (Gawronski, et al., in press). Such an acceptance, tolerance, or lack of awareness regarding the impact of one’s experienced oppression is consistent with what Freire (1970) termed “culture of silence.” Such a silence or tendency to not acknowledge and be aware of the impact of oppression raises the troubling possibility that it may contribute to the maintenance of the status quo, which implies that the oppression one experiences may remain unchallenged and likely continue. Indeed, once oppression is internalized, the oppressed is faced with the difficulty of facing a “battle on two fronts: the oppressor within and the oppressor without” (Bulhan, 1985, p. 123). Given that many Filipino Americans have automatic negative perceptions of and attitudes toward their heritage culture and ethnicity, such deeply held denigration of their ethnic selves may have adverse effects on their collective self-concepts. In turn, a damaged collective or ethnic self-concept may negatively influence their psychological well-being and mental health. Indeed, as Crocker, Luhtanen, Blaine, and Broadnax (1994) and Taylor (1997) have argued, individuals’ evaluations of their collective or ethnic selves are as vital to their psychological well-being as their evaluations of their personal selves. The presented studies support this argument, as individuals with CM-consistent automatic cognitions (Study 3) reported significantly more distress, depression, and anxiety symptoms and significantly lower levels of psychological well­ being than individuals who did not have CM-consistent automatic cognitions. Also, both indirect (i.e., CMLDP and CMIAT) and direct measures (i.e., CMS) of CM were consistently found to be

80

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

negatively correlated with mental health-related constructs such as enculturation, personal self­ esteem, collective self-esteem, life satisfaction, and ethnic identity. Furthermore, CM as measured by indirect and direct methods, positively correlated with general distress, general anxiety, and depression symptoms. These findings are consistent with theory and previous empirical research (David, under review; David & Okazaki, 2006b) suggesting that CM is an important contributing factor to the mental health of Filipino Americans. Because previous studies’ and the current studies’ results suggest that a high proportion of Filipino Americans may hold CM-consistent cognitions and behaviors, and that CM is an important potential contributor to mental health, it is recommended that mental health researchers and service providers working with Filipino Americans should become familiar with this group’s colonial history, their post­ colonial experiences, the construct of CM, and the ways in which CM may contribute to their well-being and mental health. It should be emphasized, however, that it is important to view CM as an individual differences variable and to avoid the assumption that every Filipino or Filipino American has CM, experiences poor ethnic self-regard, and that CM accounts for every Filipino or Filipino American individual’s experiences of psychological distress. Another implication for mental health researchers and service providers working with Filipino Americans is to be cognizant of how Filipino Americans may perceive the researcher’s or clinician’s race or ethnicity and, consequently, their competence. Given that the current studies’ results suggest that superior or pleasant perceptions are activated by mere exposure to American-related stimulus and that inferior or unpleasant perceptions are activated by mere exposure to Filipino-related stimulus, initial contact by researchers or clinicians with their Filipino American participants or clients may already begin to shape their attitudes and behaviors. More specifically, Filipino Americans may initially have a more positive perception

81

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

of a White researcher or clinician, whereas Filipino Americans may initially doubt the competence of a non-White or Filipino researcher or clinician who is perceived to be not very Americanized or westernized. Thus, it seems that it is more imperative for non-White and Filipino professionals to establish their credibility and demonstrate their competence to their Filipino American participants or clients early than for White researchers and service providers. However, although researchers and service providers are recommended to be aware of how their race or ethnicity may influence their Filipino American participants and clients, researchers and service providers should again keep in mind that CM is an individual differences variable and not automatically assume that every Filipino American person holds CM. Finally, in addition to being open to incorporating the notion of CM in our conceptualizations of Filipino American mental health and psychological experiences, researchers and service providers should also collaborate in designing, implementing, and evaluating intervention programs that are specifically intended to decolonize Filipino American mentalities (e.g., Halagao, 2004; Strobel, 2001). Having a culturally- and sociopoliticallysensitive conceptualization of Filipino Americans’ psychological experiences and developing culture-specific interventions may contribute toward addressing the underutilization of mental health services by Filipino Americans. Indeed, studies have found that members of this group are less comfortable seeking mental health services even compared with other Asian American groups (Ying & Hu, 1994). Using data from 2230 Filipino Americans who participated in the Filipino American Community Epidemiological Study, the most comprehensive study of helpseeking among Filipino Americans to date, Gong, Gage, and Tacata (2003) found that 75% have not used any type of mental health care service. Given that previous studies have found alarmingly high rates of depression and distress among Filipino Americans (e.g., David, under

82

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

review; David & Okazaki, 2006b; Tompar-Tiu & Sustento-Seneriches, 1995) and that the evidence on help-seeking behaviors suggests that members of this group are not likely to seek mental health services, many Filipino Americans probably have mental health needs that are not appropriately addressed. Along with stigma, poor service quality and cultural mistrust may also contribute to Filipino Americans’ disinterest in seeking mental health services. Several efforts have since been implemented to improve services and eradicate cultural mistrust. In the process, culturally sensitive practices have gained wide popularity in research and service settings. As one way to be culturally competent and effective, it is recommended that researchers and service providers incorporate sociopolitical factors such as internalized oppression (i.e., CM) in conceptualizing Filipino American mental health and develop interventions that specifically target internalized oppression. Not only may a CM-informed conceptualization and CM-specific interventions lead to an improved understanding of Filipino Americans’ experiences, but they may also improve rapport, reduce cultural mistrust between service providers and clients, and lead to better therapeutic outcomes. Such added cultural sensitivity may contribute toward improving the effectiveness of mental health services for Filipino Americans, which in turn, may contribute toward reducing the disparity in service utilization. Limitations In all studies, the stimuli set used were small and may not be representative of the entire Filipino, American, Pleasant, and Unpleasant knowledge networks. Also, the stimuli were all verbal text, which also limits the generalizability of the findings. Future studies using a larger set of stimuli or other types of stimuli (e.g., images, sounds, etc.) should be conducted to test the generality of CM-consistent cognitions. Furthermore, the Filipino stimuli used in this study may be argued as primarily salient only to some Filipinos - namely, Filipinos who are Tagalogs, who

83

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

are Catholics or Christians, whose heritage is from the northern Philippine island of Luzon that includes Metro Manila, or whose heritage is not from the southern Philippine island group of Mindanao. Indeed, the concept of CM may not be salient or an issue at all for Filipinos from Mindanao, where a large Muslim community continues to exist and strongly resist the westernization of their culture as they have ever since the advent of Spanish occupation in the early 1500s. It is important to remember that although the majority of Filipinos in the Philippines and in the U.S. are argued to have experienced centuries of oppression, not all Filipinos in the Philippines and in the U.S. may have internalized such oppression and, thus, hold CM. Again, it should be emphasized that CM is an individual differences variable that may or may not be true for some Filipinos and whose existence among Filipinos may also vary. Furthermore, the current studies’ findings are limited in the sense that most of the participants were college students or had leisurely access to computers with internet. Future studies may explore the frequency, operation, and psychological implications of CM using a more representative sample of the Filipino American population. Finally, given the continued American influence in modem day Philippines, future studies may also explore such questions using a representative sample of Filipinos in the Philippines. It should also be noted that although the relationships presented in Studies 2 and 3 between the CM measures and various mental health outcome variables were mostly statistically significant, such findings were obtained using small samples. Although CM theory and previous findings on larger samples (e.g., David & Okazaki, 2006b; David, under review) found that CM significantly contributes to various mental health outcome variables and suggests that the current studies’ findings are reliable, previous studies did not utilize an indirect measure of CM. Thus, the current findings that the CMLDP and the CMIAT significantly predict mental health

84

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

outcome variables should be interpreted with caution and the utility of such indirect measures of assessing CM should be subjected to future studies using larger and more representative samples. Finally, the cross-sectional design of the studies does not support the directionality of the relationships between variables. Future studies using more sophisticated designs (e.g., longitudinal) is needed to determine causality. Conclusion Despite the limitations, the presented studies significantly contribute to our understanding of the CM construct as it operates within and consequently affects the Filipino American population. More specifically, the results provide further evidence that CM is a vital construct to consider in terms of Filipino American psychology. Such findings support the theory that some Filipino Americans may hold a deep, internalized, and automatic inferior, unpleasant, undesirable, or negative perceptions of anything Filipino and automatic superior, pleasant, desirable, or positive perceptions of anything American, which is consistent with our theoretical understanding of CM. Furthermore, the studies provide further support to the notion that CM continues to adversely influence the psychological well-being and mental health of modem day Filipino Americans. Therefore, the presented studies also contribute to our understanding of the psychological consequences of colonialism and oppression among Filipino Americans. Finally, the presented studies also demonstrate the utility of using experimental designs and implicit methodologies in studying CM and internalized oppression. It is hoped that such added theoretical, methodological, and scientific rigor to the study of CM or internalized oppression may assist toward: (1) regarding this construct as an important factor in ethnic minority psychology; (2) regarding the study of CM or internalized oppression as a legitimate scientific enterprise; and (3) moving the scholarship of cultural and ethnic minority psychology forward.

85

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

TABLES Table 1A

Contingency Table for Study 1A

Superior

Filipino Condition

Other (e.g., Exterior, Interior, Ulterior)

Inferior

Total

Freq.

%

Exp. Freq.

Freq

%

Exp. Freq.

Freq.

%

Exp. Freq.

Freq.

46

56.8

53

22

27.2

15

13

16.0

13

81

American Condition

66

72.5

59

10

11.0

17

15

16.5

15

91

Total

112

65.1

112

32

18.6

32

28

16.3

28

172

Table IB Contingency Table for Study IB

Superior

Inferior

Other (e.g., Exterior, Interior, Ulterior)

Total

Freq.

%

Exp. Freq.

Freq.

%

Exp. Freq.

Freq,

%

Exp. Freq.

Freq.

Filipino Condition

17

45.9

23

20

54.1

13

0

0

1

37

American Condition

31

75.6

25

8

19.5

15

2

4.9

1

Total

48

61.5

48

28

35.9

28

2

2.6

2

86

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

41

78

Table 2: Intercorrelations Between Variables for Study 2

Variables

1

3

2

4

5

6

7

8

9

10

CMLDP 1.

API

-

2.

FRI

.81**

-

CMS Subscales 3.

Intlnferior

.32*

4.

CulturalShame

5 3

5.

PhysChar

.34*

6

.

7.

WGDiscrim CD

5 9

5 7

**

** .76**

7 4

**

.52** .35*

.65**

-

** .67** .57** 70** .53**

.50** 51** .37*

.46** .41*

.61**

-

Menial Health .1 2

.34*

.53** .64** .30

.43*

.27

-

9. Anhedonia

.05

.23

.2 2

** .27

.37*

.05

-.05

10. GD: Dep.

.09

.26

.37*

.52** .29

.33*

.03

-.20 -.53** -

8

. CES-D

4 7

-

Note. N=26; CMLDP = Colonial Mentality Lexical Decision Priming; API = American Preference Index; FRI = Filipino Rejection Index; CMS = Colonial Mentality Scale; Intlnferior = Internalized Cultural and Ethnic Inferiority; CulturalShame = Cultural Shame and Embarrassment; PhysChar = Physical Characteristics; WGDiscrim = Within-Group Discrimination; CD = Colonial Debt; VIA = Vancouver Index of Acculturation; CES-D = Center of Epidemiological Studies - Depression Scale; CSES = Collective Self-Esteem Scale; Imp. To Identity = Importance to Identity; Private CSE = Private Collective Self-Esteem; Public CSE = Public Collective Self-Esteem; GD: Dep. = General Distress: Depression; MEIM = Multidimensional Ethnic Identity Measure; RSES = Rosenberg Self-Esteem Scale; * p < .05; ** p < .01

87

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Table 3: Intercorrelations Between Variables in Study 3

Variables

1

3

2

4

5

6

7

8

10

9

10

11

12

13

14

CMIAT 1.

CMIAT Score

-

CMS Subscales 2.

Intlnferior

3.

CulturalShame 41**

4.

PhysChar

5. 6

.

.38**

4 4

84**

-

** .70**

WGDiscrim

.50** .76**

CD

31**

4 7

64**

-

.82** 71**

-

** .36** .57** .5 4 **

-

CSES Subscales 7. Imp. to Identity _ 4 1 * * . -.30* 8

. Membership

4 7

**

3 9

.

-.36** -.17

--.32* -.14

* * . . 4 4 ** ,-.36** ■-.30* ■- . 0 1

.33*

9. Private CSE

-.52** -.70** _ 7 9 ** -50** _ 5 9 **

10. Public CSE

_42** -4 8 * * -.55** -.43** _ 4 7 ** -.13

.27*

-.54** -.50** -.64** - 4 5 ** -.55** -.19

- .2 1

-

4 q** .63**

-

.51**

7 7

**

37** 73**

7 9

** .65**

-

VIA Subscales 11. Heritage 12. Mainstream

.38** .24

.26*

.34*

.49** .36** -.19

.19

-.18

- .1 0

.00

M ental Health 13. CES-D

.52** .50** .45**

14. Anhedonia

.49**

15. GD: Dep.

4 7

4 3

**

3 9

**

**

4 4

3 7

** .23

42** .34*

- .2 2

- 72** -.6 6 ** -.53**

.23

-.14 --.72** -.62** -.57**

** .35**

.2 1

-.19 - .75** -.65** -.59**

** -.60** -.51** -.51**

- .2 2

** .47** .49**

4 4

Other Variables 16.MEIM

-.51**

17. RSES

-.28*

_

4 9

-.13

18. Life Satisfaction -.52** -.34*

- .2 0 _

3

^**

-.06 - .2 2

-.03

.0 1

-.29* -.05

.39** .69**

.80** .6 6 **

.24

.29*

.33**

.26* .53**

.06

.22

50** .35**

88

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

.22 -.43** -.25*

Table 3: (Continued) Intercorrelations Between Variables for Study 3

Variables

15

16

16. MEIM

-.58**

-

17. RSES

-.34*

17

Continued 15. GD: Dep. Other Variables

18. Life Satisfaction -.62**

.25* .42**

3 7

**

-

Note. N=44; CMIAT Score = Colonial Mentality Implicit Association Test Score; CMS = Colonial Mentality Scale; Intlnferior = Internalized Cultural and Ethnic Inferiority; CulturalShame - Cultural Shame and Embarrassment; PhysChar = Physical Characteristics; WGDiscrim = Within-Group Discrimination; CD = Colonial Debt; VIA = Vancouver Index of Acculturation; CES-D = Center of Epidemiological Studies - Depression Scale; CSES = Collective Self-Esteem Scale; Imp. To Identity = Importance to Identity; Private CSE = Private Collective Self-Esteem; Public CSE = Public Collective Self-Esteem; GD: Dep. = General Distress: Depression; MEIM = Multidimensional Ethnic Identity Measure; RSES = Rosenberg Self-Esteem Scale; * p < .05; ** p < .01

89

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Table 4:

Hierarchical Regressions of Colonial Mentality Measures on Depression Symptoms

Predictor Variables

0

R2

Adjusted R2

R Change

F Change Sig.

.231

.230

.212

.230

12.552 < ,001

.328* .339

.307

.109

6.739