Asian Americans as Model Minorities? A Look at Their Academic Performance

GRACE KAO

University of Chicago

The image of Asian-Americans as "model minorities" is driven, in part, by the high academic achievement of Asian-American children. To evaluate this characterization, I use the National Education Longitudinal Study of 1988 (NELS:88) to compare Asian and white eighth graders on reading and math test scores and grades. Results indicate that the difference between Asians overall and whites on reading and math test scores can be explained by differences in family background. However, analyses by Asian subgroups reveal that Chinese, Korean, and Southeast Asian youth earn higher math scores while Pacific Islanders earn considerably lower math and reading scores than their white counterparts. Whites and Asians differ substantially in their grade performance, which suggests that cultural and behavioral differences may be an important influence on grades. Analyses of Asian subgroups show no statistical difference between ethnic groups, although only Southeast Asians earn perceivably higher grades than whites after controlling for the effects of family characteristics.

Introduction

Recent years have witnessed increasing speculation about the academic success of Asian-American students. The relatively high socioeconomic standing of Asian-Americans (compared with blacks and Hispanics), low rates of marital disruption, and the relative success of Asian-American students suggests an almost problem-free home environment (Ta-kaki 1989; Hurh and Kim 1989). The phenomenon of academic suc-

American Journal of Education 103 (February 1995)

0195-6744/95/0302-0001 $01.00

February 1995 121

Asian Americans as Model Minorities?

cess is part of a broader discussion of Asian¹ success in the United States that has prompted the media to coin the term "model minority" for this population.

Support for the model minority image of Asians ranges from full acceptance to complete dismissal. In favor of the model minority image, Sowell writes, "Groups that arrived in America financially destitute have rapidly risen to affluence, when their cultures stressed the values and behavior required in an industrial and commercial economy. Even when color and racial prejudice confronted them--as in the case of the Chinese and Japanese--this proved to be an impediment but was ultimately unable to stop them" (Sowell 1981, p. 284). Other scholars have criticized the homogeneous characterizing of Asians, arguing that this image is just another phase in the evolution of stereotypes of Asian-Americans (Hurh and Kim 1989; Takaki 1989; Steinberg 1989). Yet another group has offered conditional support for Asian-American success in some arenas of life but notes that Asian-Americans are still not on a par with whites in other realms (Hirschman and Wong 1981; Barringer et al. 1990). Barringer et al. (1990), for instance, found that while Asians have higher levels of education than whites, economic returns to education are lower for Asians than whites.

Despite the attention Asian academic success and achievement has received from the media, there has been little research on the mechanisms responsible for this phenomenon. The relative neglect of Asians' educational experience can be explained in two ways. First, the success of Asians is not a problem in need of a solution. Second, until recently, a nationally representative survey of students with adequate samples of Asian-Americans has been lacking; hence researchers have been unable to examine on a national level whether and how academic achievement of Asians differs from that of the white population.

The motivation for the few studies of Asians' school performance stems from a desire to decode the "secrets of their success." Indeed, if Asian students are academically successful because of certain parental practices in the home environment, there may be lessons to policymak-

Grace Kao is a Ph.D. candidate in the Department of Sociology at the University of Chicago. She is currently working on a research project (with Marta Tienda) that examines racial and ethnic stratification in educational outcomes among youth, and her dissertation work focuses on the construction of ethnic identity and how it circumscribes conceptions of success and aspirations among adolescents.

122 American Journal of Education

Као

ers that can benefit others. In the words of Mike Wallace of CBS's 60 Minutes, "They must be doing something right. Let's bottle it" (from Takaki 1989, p. 474). Implicit in this view is that cultural differences between Asians and whites hold the secret to understanding their success. An alternative interpretation of the high educational performance of Asians students is that it merely reflects socioeconomic differences between Asians and whites. That is, because the average Asian parent has completed more schooling than the average white parent and because parental schooling is a critical determinant of their children's academic success, differences between Asian and white children simply mirror these differences. I hope to clarify the relative merits of both interpretations.

This article has several aims: (1) to examine whether Asian eighth graders overall and students from eight Asian ethnic groups (specifically, Chinese, Filipino, Japanese, Korean, Southeast Asian, Pacific Islander, South Asian, and West Asian)² perform better in school than their white counterparts as measured by their grades, math test scores, and reading test scores; (2) to characterize the features of students' home and school environments that affect their academic outcomes; and finally (3) to analyze the diversity of educational performance and the home environment among the eight Asian ethnic groups.

The analyses proceed as follows. First, I present a literature review to formulate the theoretical issues and substantive concerns in the area of educational performance among ethnic minorities. The empirical analysis begins with comparisons between Asians overall, Asian ethnic groups, and whites on a variety of background measures and outcome variables. Subsequently, I compare math and reading scores to consider whether Asians, whites, and Asian subgroups differ in their skill development, and I examine how background variables, family context, home educational resources, and student characteristics affect grades. The concluding section considers why Asians overall receive higher grades than white students and examines the pattern of differentiation within the Asian population, emphasizing the different parent-child relationship among Asians and the higher expectations some Asian parents have for their children, by using insight gained from focus-group discussions.

In most of the descriptive and analytic tables, I examine Asians as a group and Asian ethnic groups separately. Although the category "Asian" incorporates many heterogeneous groups, there is reason to evaluate how Asians fare relative to whites. That is, in order to properly evaluate the model minority thesis, which implicitly lumps all Asian groups together, it is critical that I measure how Asians as a group differ from whites. If Asians as a group perform at comparable or

February 1995 123

Asian Americans as Model Minorities?

lower levels than whites, then that alone provides evidence against the caricature of Asians as model minorities. However, as we will see, the heterogeneity of family educational resources within the Asian population makes any single characterization of this group inadequate.

Ethnicity and Academic Performance

There are several approaches to understanding ethnic differentials in educational performance. One approach links academic performance to the meanings attached to ethnic labels (Fordham and Ogbu 1986; Matute-Bianchi 1986; Ogbu 1991; Gibson and Ogbu 1991). Matute-Bianchi found that in a California school, Japanese-American identity was compatible with academic success, but that being a good student was less consistent with being Mexican (Matute-Bianchi 1986). Some minority students who succeed in school are criticized by their peers as "trying to act white," so that a positive identification with black peers, for instance, may include not doing well in school. Ogbu also finds that blacks identify largely in opposition to whites; so if success in school is labeled as a "white" activity, this may discourage blacks from pursuing academic goals (Clark 1983; Ogbu 1991).

Another type of explanation for differential academic performance focuses on culture. For instance, Schneider and Lee (1990) argue that the cultural components that benefit East Asian children's school performances include "(1) the East Asian cultural tradition which places a high value on education for self-improvement, self-esteem, and family honor, and (2) the determination by some East Asian families to overcome occupational discrimination by investing in education" (Schneider and Lee 1990). However, their research does not directly provide evidence that East Asians value education more than whites, blacks, Hispanics, or other Asian groups.

Schneider and Lee (1990) conducted in-depth interviews to illustrate how academic achievement is inextricably linked to children's perception of what makes their parents happy. Their insights suggest a working hypothesis, namely, that white parents, on average, express satisfaction with their children if they are successful in one of the many realms of youths' lives (school, sports, music, or other hobbies), but Asian parents express satisfaction only when their children have near perfect academic performance (i.e., straight A's). This difference comes not only from differential expectations, but also from differences in the expressiveness of Asian and white parents (Stevenson and Stigler 1992).

Another explanation comes from Ogbu's (1991) analysis of blacks. He argues that although black parents value schooling, they warn their children that American society does not reward blacks and whites

124 American Journal of Education

Као

equally for such credentials. On the basis of their concrete experiences, black parents offer their children ambivalent and perhaps contradictory values toward education (Ogbu 1991). Although Asian-Americans also receive returns for education unequal to those of whites (Bar-ringer et al. 1990), their higher levels of education partly compensate for this inequity. Asian parents may be promoting higher levels of educational attainment for their children to compensate for anticipated discrimination in the job market. Nonetheless, it is interesting that the experience of discrimination, according to different scholars, accounts for the relative failure of blacks and the success of Asians in educational achievement (Sue and Okazaki 1990; Schneider and Lee 1990; Ogbu 1991).

Finally, Sue and Okazaki (1990) attempt to identify the parental behaviors among Asian and white parents that can clarify the differences found in school achievement. They argue against cultural explanations of Asian achievement and instead propose that, aware of the obstacles (discrimination) in the labor market, Asian parents and children learn to value education, which is related to but not entirely synonymous with career success. Their review of prior research also revealed difficulties in linking the average behavior of Asian parents to their children's performance. As a group, Asian parents were more likely to insist on unquestioned obedience to parents and to believe in minimal parental involvement, and they were less likely to have expectations for mature behavior and to encourage two-way communication between their children and themselves compared to the average behavior of white parents. The average behavior of Asian parents would have predicted low academic achievement for the sample as a whole, yet the mean grade point average of Asian children was the highest of any group (Dornbusch 1989; Sue and Okazaki 1990).

These results suggest that systematic group differences in parental behavior exists between Asians and whites. It may be that, on average, Asian parents may not only have higher expectations of their children, but also are unwilling to negotiate these terms. Children understand this message and are obligated to their parents to do well in school. For the children who are able to attain academic success, their parents' high expectations may further their own educational aspirations, but for those who cannot, the effects of such pressure are potentially harmful.

Parental Resources and Academic Achievement

Much of the research about what promotes academic success has concentrated on three general domains of the students' social environment that influence academic achievement, namely, parental involvement,

February 1995 125

Asian Americans as Model Minorities?

home resources, and peer influences (Murnane et al. 1981; Baker and Stevenson 1986; Fehrman et al. 1987; Rumberger et al. 1990; Astone and McLanahan 1991).

First, some emphasize the importance of parental involvement (Baker and Stevenson 1986; Fehrman et al. 1987). Baker and Stevenson (1986) argue that to be effective managers of their children's educational careers, parents must know the demands of the school, their children's performance, and when to use their managerial skills. They found that mothers with higher socioeconomic standing had more accurate information about their children's performance and contact with the school. Mothers with at least a college education were more likely to choose college-preparatory classes for their children, regardless of their children's grade point average (Baker and Stevenson 1986). High levels of parental involvement, measured by knowledge of the child's activities in school and frequent contact with the school, have positive effects on children's academic achievement (Baker and Stevenson 1986; Fehrman et al. 1987; Hoover-Dempsey et al. 1987; Sue and Okazaki 1990; Astone and McLanahan 1991).

Some researchers have focused on home resources, which can include the presence of books, a home computer, or newspapers (Teachman 1987; Stevenson and Stigler 1992). Teachman (1987) argues that the availability of home resources creates an environment conducive to studying by displaying a positive orientation toward schooling. He also points out that parents with more education and income are probably more motivated to provide home resources for education.

Finally, others have examined the effect of peer groups on academic achievement. Although there is some disagreement about whether peers are generally a positive or negative influence on educational performance, researchers recognize the importance of peers in affecting educational achievement. Hallinan (1988) argues that much of the evidence suggests that peer group culture usually hinders learning. On the other hand, Corsaro and Eder (1990) find that older adolescents select friends who have similar attitudes toward school, college plans, and achievement outcomes. Their findings suggest that peers only reinforce the adolescent's own educational plans and performance. The National Education Longitudinal Study of 1988 (NELS:88) measures of peer evaluation allow us to examine the respondents' self-perceptions of how others think of him/herself, thus allowing us to examine group differences in self-perception.

126 American Journal of Education

Као

Conceptual Model

Sue and Okazaki (1990) claim that there are two general hypotheses to explain Asian-American academic achievement. One hypothesis invokes the idea that Asians are better equipped for schooling or more skilled than their white counterparts, while the second argues that there is something distinctive in the value system or the behavior of Asians that promotes successful outcomes in school.

The first set of analyses addresses whether Asians and whites differ in the assessment of their achievement. Here I use reading and math scores from the test developed for the NELS:88 respondents as a proxy for an objective evaluation of their achievement level.

Then I examine the second hypothesis using their self-reported grade point averages as a measure of academic performance that is more sensitive to behavioral differences. Fehrman et al. (1987) argue that "grades are given more frequently in schools than are achievement tests and can therefore give a more continuous indication of a student's academic performance. Grades are also more readily understood by parents and students due to their frequency and universal use. Furthermore, grades are likely to be more sensitive than achievement test scores to student effort" (p. 331).^s

Figure 1 presents the conceptual model. Our model includes measures of family background and structure, educational resources, and student characteristics. Parent's education⁴ and family income⁵ measure socioeconomic status. Parent's education has consistently been shown to be associated with higher achievement (grades, test scores), as well as greater levels of parental involvement (Murnane et al. 1981; Baker and Stevenson 1986; Fehrman et al. 1987; Rumberger et al. 1990; Astone and McLanahan 1991).

Measures of family structure include whether the respondent is the oldest child, the number of siblings of the student, and the headship structure of the household. Being the oldest child should have a positive effect on grades, because of the greater sense of responsibility placed on the oldest. We would expect the number of siblings to have a negative effect on grades, because of increased competition for material and psychological resources. However, Caplan et al. (1992) found that among Indochinese youth, siblings benefit from doing homework together. When older siblings tutor their younger siblings, younger siblings benefit from the help they get, while older siblings gain an opportunity to review educational materials.

Measures for household headship include variables for living with both parents and living with mother only. Preliminary analyses suggested that these two types of headship characterize the difference

February 1995 127

Asian Americans as Model Minorities?

Family Background Race

Urban/Suburban/Rural Parent's Education Family Income

Family Structure Oldest

Home Educational Resources

Home Environment (books, place to study, etc.)

Outside classes

Family Communication behaviors

Number of Siblings Headship structure Immigrant Mother

Student Characteristics Ever Repeat LEP

Number of hours spent on homework Activites in school Activities outside of school

^^ Intermediate causal mechanism not

analyzed in this paper.

^ Analysis in this paper.

"^ ^- Feedback loop from outcome measures

not analyzed in this paper.

Fig. 1.--Conceptual model

between Asians and whites in the relationship between headship and scholastic performance. Since these are compared with other headship structures, living with both parents is hypothesized to have positive influences on grades while living with mother only should have no effect (Astone and McLanahan 1991).

Prior studies indirectly suggest that having an immigrant mother may raise grade performance because first-generation parents may be more optimistic than those who have been subjected to disparate treatment for many generations (Ogbu 1991). Mother's immigrant status also proxies for the effect of not becoming "Americanized" (i.e., living and adopting the culture and values of the United States) and being able to exert more control over their adolescents since traditional (Asian) values promote more authoritarian behavior from parents (Sue and Okazaki 1990; Schneider and Lee 1990). Finally, if cultural influences benefit Asian students' performance, then the manifestation of

128 American Journal of Education

Као

cultural distinctiveness should be more apparent among those who are less acculturated.

Building on Teachman's (1987) idea of home resources, I expanded the definition to include not only items such as having more than 50 books, having regularly received a magazine or newspaper, or having a typewriter, but also whether or not respondents have a personal computer at home and a specific place to study. In addition, we examine the effects of family communication about academic concerns on grade performance. Recent studies suggest that the greater the interaction between parent and child, the greater the positive effects of communication (Baker and Stevenson 1986; Fehrman et al. 1987).⁶

The conceptual model also allows for individual differences that influence educational achievement. Having ever repeated a grade between fifth and eighth grades should lower current grades because these experiences may act as negative signals to teachers as well as to decrease attachment to school. A variable indexing "limited English proficiency" (LEP)⁷ is included as a control because lack of English fluency may lower grades. Limited English proficiency status also differentially affects the Asian population, with Southeast Asians suffering most from LEP status because of their more recent arrival, while LEP status does not significantly affect the lives of most South Asians since they tend to have excellent English skills.

Other behavioral indicators that influence academic achievement include the number of hours spent on homework per week,⁸ whether the child participates in any activities in school, and whether the child participates in any activities outside of school. Extracurricular activities signal an attachment to school and hence should be associated with higher grades. Furthermore, activities outside of school (such as Boy Scouts or Girl Scouts, religious clubs, summer institutes) may heighten interest in school by broadening the child's experiences.

Data and Methods

The primary data used for this study come from NELS:88. The NELS:88 used a two-stage stratified probability design to select a nationally representative sample of schools and students. The first stage resulted in 1,052 participating schools, and the second stage produced a random sample of 26,435 students, of which 24,599 participated. The NELS:88 includes oversamples of Asian-American and Hispanic students, resulting in 1,527 Asian students. The NELS:88 also surveyed the respondents' parents, two of each student's teachers (one from English or history and the other from mathematics or science),

February 1995 129

Asian Americans as Model Minorities?

and a school administrator. These analyses are restricted to base year survey data.

The NELS:88 is unique in other ways as well, namely, in that it provides information directly from the parents, teachers, and schools. It allows, for instance, a comparison of answers between the student's expected educational attainment and the parent's aspiration for their child's educational attainment. Student respondents also took a series of cognitive tests that provide a standardized assessment of skill development.⁹

The secondary source of data are three focus groups conducted during the Spring of 1992 (see app. С for a more detailed description). Each of these focus groups (one with Asian, one with black, and one with Hispanic students) consisted of open-ended responses from a group of university students on a series of questions regarding the salience of their ethnic identity and how their identificiation plays out in everyday life. These discussions will be used primarily to illuminate the statistical analysis and to further suggest the mechanisms in educational performance.

Population Comparisons

Since higher levels of parental education are associated with academic success (Baker and Stevenson 1986), table 1 presents the distribution of parent's education. The relative edge Asian parents have is particularly evident at the highest educational levels, as 49 percent of the Asian parents are college graduates compared with only 35 percent of the white parents. Nonetheless, the educational advantage of Asian over white parents must be qualified. More than 8 percent of the Asian parents had less than high school education compared to less than 6 percent of the white parents. In addition, there are striking differences within the Asian population. Overall, South Asian parents are overwhelmingly better educated than their other Asian counterparts. Over 80 percent of South Asian parents have at least a college degree as compared to 35 percent of white parents. On the other hand, Southeast Asian parents have much lower levels of education than white parents. Almost 20 percent of Southeast Asian parents have less than high school education while only about 6 percent of white parents have similarly low levels of education. Chinese parents are more heterogeneous than white parents, with a larger proportion falling in both the lowest and highest educational levels--specifically, about 13 percent have less than a high school education while about 43 percent have at least a college degree. When comparing the mean levels of

130 American Journal of Education

TABLE 1

Parent's Education

						ASIAN SUBGROUPS
		ASIANS					Southeast	Pacific	South	West
	WHITES	(All)	Chinese	Filipino	Japanese	Korean	Asian	Islander	Asian	Asian
Less than high
school (%)	5.5	8.4	12.6	1.1	3.3	2.2	19.6	12.4	1.6	7.1
High school
graduate (%)	19.5	11.4	15.3	8.7	4.4	10.4	14.7	12.4	2.5	16.1
Some college (%)	40.0	30.9	28.6	30.5	29.7	23.0	43.6	48.3	14.8	29.0
College graduate
(%)	17.4	24.0	15.6	41.8	40.7	32.2	11.1	15.7	17.2	19.4
Master's degree (%)	10.9	12.4	13.9	7.6	13.2	19.1	6.2	7.9	27.9	10.3
M.D. or Ph.D. (%)	6.7	12.9	13.9	10.2	8.8	13.1	4.9	3.4	36.1	18.1
Mean educational
level	14.6	15.2***	14.90*	15.53***	15.65***	15.90***	13.69***	14.09	17.51***	15.28**
	(2.49)	(2.83)	(3.12)	(2.18)	(2.22)	(2.50)	(2.62)	(2.37)	(2.50)	(3.02)
N	16,116	1,453	294	275	91	183	225	89	122	155

Source.--NELS:88.

Note. -- Numbers in parentheses are standard deviations. * Significantly different from whites at p < .05 level. ** Significantly different from whites at/) < .01 level. *** Significantly different from whites at p < .001 level.

Asian Americans as Model Minorities?

education, we see that only Southeast Asian parents have lower levels of education than white parents.

A mechanism through which parent's education leads to academic success is educational aspirations. Educational aspirations deserve special attention because they help to explain why some students are more motivated than others to achieve high grades. At the same time, the level of grades one receives shapes one's educational aspirations, so that educational goals and grades are mutually reinforcing. Table 2 presents two measures of educational aspirations: parents' expectation and child's own expectation. The means reveal that, overall, Asians rank higher than whites on both of these measures. Asian parents were almost two and a half times as likely as white parents to expect their children to complete postgraduate degrees--47 percent versus 20 percent. South Asian parents also have extraordinary expectations; about 92 percent of the South Asian parents expect their children to graduate from college while 72 percent expect their children to earn graduate degrees. Even though Southeast Asian parents have lower levels of education than white parents, they have higher expectations for their children's eventual educational attainment than white parents. However, Asians are not uniformly high in their aspirations. Note that almost 20 percent of Pacific Island students do not plan to go beyond high school, while 15 percent of their parents also have such low aspirations. Only 45 percent of Pacific Islander parents expect their children to graduate from college, compared to over 60 percent of white parents. Nonetheless, over three-quarters of parents and students from all other Asian groups aspire to graduate from college. In fact, with the exception of Pacific Islanders, both parents and children from each Asian ethnic group have higher educational aspirations than white parents and eighth graders. From these results, it would be reasonable to hypothesize that Asians achieve higher grades than whites simply because they have higher educational aspirations.

Table 3 examines whether having high aspirations is the key to success for Asian youth by comparing grades, math, and reading test scores of Asian and white students at each level of educational aspirations. Overall, we find that Asians have slightly higher math scores but comparable reading scores relative to whites. With respect to grades, Asians average 3.24 grade points compared to 2.96 for whites. If Asians earn higher grades simply because they have higher educational aspirations, then we should expect to find no difference in grades between whites and Asians with comparable educational aspirations. Instead, we find that at each level of aspirations, Asian students still perform better than white students in terms of grades. It appears that Asians overestimate the level of grades needed to realize their level of

132 American Journal of Education

TABLE 2

Parents' and Children's Educational Aspirations: Whites, Asians, and Asian Ethnic Groups

		Asians					Southeast	Pacific	South	West
	Whites	(All)	Chinese	Filipino	Japanese	Korean	Asian	Islander	Asian	Asian
Parents' aspirations:
High school or
less (%)	11.5	6.5	3.8	7.5	3.5	5.9	7.8	15.9	3.5	6.5
Some college (%)	26.7	13.2	12.0	11.5	10.6	5.3	16.1	39.0	4.3	13.8
College graduate
(%)	41.4	33.8	32.0	45.2	41.2	31.4	30.4	24.4	20.0	37.7
Beyond college (%)	20.4	46.5	52.3	35.7	44.7	57.4	45.6	20.7	72.2	42.0
iviedn level 01 aspirations	15.41	16.41***	16.65***	16.18***	16.54***	16.81***	16.28***	15.0*	17.22***	16.30***
	(1.84)	(1.81)	(1.66)	(1.75)	(1.59)	(1.68)	(1.91)	(1.99)	(1.47)	(1.79)
N	15,254	1,341	266	252	85	169	217	82	115	138
Eighth grader's
aspirations:
High school or
less (%)	10.5	5.1	3.6	4.9	6.7	1.6	5.9	18.6	0.8	4.9
Some college (%)	19.7	15.0	12.7	18.5	5.6	5.9	18.9	32.0	6.3	16.0
College graduate
(%)	45.1	37.1	37.7	43.0	45.6	33.7	39.1	28.9	30.2	33.3
Beyond college (%)	24.7	42.8	46.1	33.6	42.2	58.8	36.1	20.6	62.7	45.7
Mean level of
aspirations	15.68	16.35***	16.53***	16.10***	16.47***	16.99***	16.11***	15.03***	17.10***	16.40***
	(1.83)	(1.73)	(1.63)	(1.69)	(1.67)	(1.37)	(1.77)	(2.04)	(1.30)	(1.77)
N	16,236	1,515	308	286	90	187	238	97	126	162

Source. --NELS:88.

Note.--Numbers in parentheses are standard deviations. * Significantly different from whites at p < .05 level. *** Significantly different from whites atp < .001 level.

Asian Americans as Model Minorities?

TABLE 3

Comparison of Grades, Math Test Scores, and Reading Test Scores Overall and within Educational Aspirations Levels: Asians and Whites

		Math	Reading
	Grades	Scores	Scores	N
Overall mean achievement:
Whites	2.96	52.50	52.32	15,750
	(.75)	(9.90)	(9.78)
Asians	3.24***	54.91***	51.95	1,477
	(.72)	(10.46)	(9.99)
Within levels of educational
aspirations:
High school or less:
Whites	2.23	43.96	44.04	1,693
	(.75)	(7.55)	(8.66)
Asians	2.46*	44.58	43.85	72
	(.83)	(8.69)	(9.06)
Some college:
Whites	2.60	47.63	48.10	3,088
	(.69)	(8.53)	(9.11)
Asians	2.80***	48.61	45.91	219
	(.76)	(8.72)	(8.66)
College graduate:
Whites	3.07	53.68	53.37	7,082
	(.66)	(9.15)	(9.00)
Asians	3.20***	53.94	51.28***	544
	(.67)	(9.95)	(9.22)
Beyond college:
Whites	3.36	57.86	57.29	3,901
	(.62)	(8.96)	(8.41)
Asians	3.53***	59.20**	55.62***	635
	(.55)	(9.38)	(9.44)

Source. --NELS:88.

Note. -- Values in parentheses are standard deviations. * Significantly different from whites at/) < .01 level. ** Significantly different from whites at p < .05 level. *** Significantly different from whites at/" < .001 level.

educational aspirations, regardless of the amount of education they plan to accumulate. However, we do not find a clear "Asian advantage" in test scores. The general pattern regarding test scores, once educational aspirations are controlled for, reveals no differences in math scores except among those with the highest aspirations. With respect to reading scores, however, Asians score comparable to or lower than

134 American Journal of Education

Као

whites, with white students who aspire to graduate from college or attend graduate school earning higher scores than Asian students with comparable aspirations. Thus, Asians outperform whites in terms of grades rather than test scores.

Table 4 compares Asian and white youth on various background measures, including those used in the regression analysis. Financial resources increase the likelihood that students will be able to pursue college plans. Although the average incomes of white and Asian families do not significantly differ, more Asian parents reported having begun to save money for their children's college expenses. Moreover, of those who have set aside money for college, Asian parents had saved significantly more, and planned to save significantly more by the time their children graduated from high school, than white parents. We stress that this behavior does not reflect greater financial resources but rather a greater commitment of resources for education among Asian parents than among white parents.

Home resources are important in creating an environment conducive to good study habits and strong academic achivement. Again, differences between Asians and whites are apparent. While more Asian students have a place to study, more white students report having their own room. This reinforces the notion that even where differences in material resources are minimal, as between Asians and whites, the stronger commitment to education among Asian parents is evident in their emphasis on study over privacy for their children. Asians are also more likely than whites to have a personal computer in the home. Other differences reflecting academic priorities are discernible in NELS:88. For instance, Asian parents are more likely to enroll their children in art, music, ethnic history, and computer classes.

Asians also differ from whites in their parenting behavior. For instance, Asian parents are more likely to restrict the number of hours watching television, while white parents are more likely to restrict the types of programs and how late the child may stay up to watch television. Asian parents' interest in limiting the number of hours spent watching television is probably driven by the desire to protect the amount of time available for homework. Asian parents are also more likely to have rules about maintaining grades (74 percent vs. 68 percent) but are less likely to have rules about doing chores around the house (80 percent vs. 88 percent). This pattern of rules is also consistent with a slightly greater emphasis on the importance of schoolwork over all other responsibilities the child may have at home.

Asian parents are less likely than white parents to talk to their children about school experiences or their plans for high school and beyond, and they are less likely to help with homework. Overall, par-February 1995 135

TABLE 4

Background Characteristics of Students by Race

	Asians	Whites
Financial resource :^a
Mean family income ($)	48,676	46,964
	(43,937)	(39,579)
Have money for post-high-school education	.62***	.53
Of those who have money saved up, how
much now ($)	6,959***	5,845
	(5,739)	(5,303)
How much by time child finishes high
school ($)	10,532***	9,640
	(5,005)	(4,948)
Home resources:¹¹
Place to study	.59***	.40
Own room	.73***	.86
Personal computer	.52***	.48
Outside classes :^a
Art	.11*	.09
Music	.34***	.29
Ethnic history	.11***	.04
Computer	.16***	.12
Household rules :^a
Rules about TV programs	.61***	.70
Rules about how late can watch TV	.75***	.85
Rules about how many hours TV daily	.50***	.40
Rules about maintaining grades	74***	.68
Rules about doing homework	.90	.91
Rules about doing chores	.80***	.88
Parent-child interaction:³
How often talk to child about school
experiences⁰	3.49***	3.81
	(.73)	(.45)
How often talk about high school plans^c	3.21***	3.37
	(.79)	(.68)
How often talk about post-high-school plans⁰	3.10* (.83) 2.05*	3.20 (.72) 2.26
How often help child with homework^d	3.10* (.83) 2.05*	3.20 (.72) 2.26
	(1.02)	(.96)
Characteristics of student:^b
Ever repeat	.02***	.05
Limited English proficiency	.08***	.78
Number of hours worked for pay	3.03***	4.91
	(5.09)	(5.97)
Peer evaluation:^be
Popular	.91***	.99
	(.59)	(.57)
Athletic	.95***	1.03
	(.73)	(.71)
Good student	1.47***	1.26
	(.58)	(.61)

Као

table 4 (Continued)
	Asians	Whites
Peer evaluation (continued)'}'* Troublemaker	.26*** (.51)	.35 (.57)

Source.--NELS:88.

Note.--Values in parentheses are standard deviations.

^a The number of respondents varies from about 1,300 for Asians to 15,000 for whites (parent questionnaire items).

^ь The number of respondents varies from about 1,500 for Asians to 16,000 for whites (student questionnaire items).

^c Answer categories are 1 = not at all to 4 = regularly.

^d Answer categories are 1 = seldom or never to 4 = almost every day.

^e 0 = not at all (popular, athletic, a good student, etc.), 1 = somewhat, 2 = very.

* Significantly different from whites at p < .05 level.

*** Significantly different from whites at/" < .001 level.

ents who help with their children's homework have lower achieving children. Recall that Sue and Okazaki (1990) also found that Asian parents scored lowest on parental involvement measures yet their children had the highest grades. For many Asian parents, a sense of duty clearly marks the boundaries of responsibility. In an ideal world, parents are responsible for the economic support of their children and providing the material resources for their children's education. In turn, children are responsible for performing well in school, and, if they do so, they have fulfilled their most important duty to their parents.

Asians are less likely than whites to have ever repeated fifth, sixth, seventh, or eighth grade (2 percent vs. 5 percent) but are more likely to be labeled LEP. Grade repetition and LEP status are hypothesized to lower grades. Finally, students' impression of their peers' evaluations indicate that Asians are more often considered to be good students but whites are more likely to be considered popular, athletic, or troublemakers. Being considered a good student reinforces the identity of doing well in school.

Table 5 presents the background characteristics of the Asian ethnic groups. Although Asians and whites do not differ with regard to income, these tabulations show that there are substantial differences in material resources within the Asian population. South Asians are the most economically advantaged, while Southeast Asians are the least. Pacific Islanders also have lower family incomes, although their small number in the sample makes it more difficult for us to distinguish

February 1995 137

TABLE 5

Background Characteristics of Students: Whites and Asian Ethnic Groups

	White	Chinese	Filipino	Japanese
Financial resources:³
Mean family income ($)	46,964	45,714	47,169	60,104**
	(39,579)	(41,702)	(36,413)	(38,825)
Have money for post-high-
school education	.53	.73***	.63**	.62
Of those who have money
saved, how much now	5,845	7,443***	5,837	5,804
	(5,303)	(5,792)	(5,419)	(4,823)
How much by time child
finished high school	9,640	10,688*	9,790	9,723
	(4,948)	(5,066)	(4,929)	(4,872)
Home resources:¹"
Place to study	.40	.56***	.60***	.63***
Own room	.86	.63***	.76***	.88
Computer	.48	.54*	.50	.62**
Outside classes:³
Art	.09	.13	.07	.13
Music	.29	.33	.34	.42*
Ethnic history	.04	.09***	.03	.09*
Computer	.12	.15	.14	.18
Household rules:³
Rules about TV programs	.70	.50***	.64	.54**
Rules about how late child can
watch TV	.85	.68***	₇₈*,	.77*
Rules about how many hours
TV daily	.40	.44	.51***	.51*
Rules about maintaining
grades	.68	.71	.73	.72
Rules about doing homework	.91	.87*	.91	.86
Rules about doing chores	.88	.73***	.88	.80*
Parent-child interaction:³
How often talk to child about
school experiences^c	3.81		3.68	3.74
	(.45)	(.85)	(.50)	(.56)
How often talk about high
school plans'	3.37	2.99***	3.39	3.36
	(.68)	(.85)	(.68)	(.70)
How often talk about post-
high-school plans^c	3.20	2.90***	3.24	3.28
	(•72)	(.86)	(.70)	(.69)
How often help child with
homework^d	2.26	1.91***	2.11*	2.09
	("96)	(.98)	(1.00)	(.98)
Characteristics of student:^b
Ever repeat	.05	.02*	.02**	.02
LEP	.01	.11***	07,	.07***
Number of hours worked
for pay	4.91	2.78***	2.72***	2.91**
	(5.97)	(4.91)	(4.43)	(4.91)
Peer evaluations:¹""
Popular	.99	.78***	.96	.96
	(.57)	(.59)	(.60)	(.53)
Athletic	1.03	82***	.99	.96
	("71)	(.71)	(.74)	(.69)

138

Korean	Southeast Asian	Pacific Islander	South Asian	West Asian
56,401** (46,019)	28,789*** (32,744)	38,765 (39,751)	77,983*** (55,244)	52,920 (50,462)
.58	.52	.55	.72***	.53
7,106* (5,727)	4,417* (4,996)	5,324 (5,811)	10,412*** (5,252)	8,375*** (6,069)
11,094** (4,788)	8,329* (5,112)	9,242 (5,758)	12,911*** (3,976)	11,590** (4,975)
.62* .83 .61	.60* .60* .34***	.56 .71* .37*	.71* .82 .71*	.54** .82 .58*
.15 .52* .15*** .18*	.07 .14* .10* .12	.13 .35 .12*** .13	.17** .38* .20* .28*	.09 .37* .16*** .16
.60**	.60**	.73	.83***	.64
.69***	.71***	.83	.91	.81
.45	.43	.53*	.78***	.51**
.72 .88 .78***	.75* .88 .82**	.73 .96 .89	.84* .98* .83	.74 .91 .73***
3.42*** (.70)	3.23*** (•82)	3.61*** (•62)	3.68** (.55)	3.67*** (.68)
3.09*** (.77)	3.02*** (.83)	3.22 ("78)	3.49* (.68)	3.38 (.73)
3.15 (.77)	2.96*** (.93)	3.14 (.90)	3.27 (.73)	3.17 (.87)
1.92*** (.97)	1.89*** (1.06)	2.43 (1.08)	2.36 (1.00)	2.09* (1.01)
.02* .04***	.02 .17***	.09 .06***	.01* .04***	.05 .07***
2.94*** (4.32)	3.06 (5.32)	3.82 (5.63)	1.78*** (4.17)	4.47 (6.59)
.97 (-55) .90* (.69)	.85*** (.62) .96 (.74)	1.02 (.57) 1.15 (.74)	.88* (.55) .88* (.69)	1.02 (.61) 1.07 (.75)

(Table continues on p. 140.)

139

Asian Americans as Model Minorities?

table 5 (Continued)
	White	Chinese	Filipino	Japanese
Good student Troublemaker	1.26 (.61) .35 ("57)	1.52* (.54) 22* (.49)	1.43*** (.58) .26* (.51)	1.26 (.62) .34 (.56)

Source. --NELS:88.

Note. -- Values in parentheses are standard deviations.

^a The number of respondents varies: about 15,000 for whites, 227 for Chinese, 250 for Filipinos, 84 for Japanese, 165 for Koreans, 211 for Southeast Asians, 83 for Pacific Islanders, 114 for South Asians, and 134 for West Asians (parent questionnaire items).

^bThe number of respondents varies: about 16,000 for whites, 307 for Chinses, 280 for Filipinos, 93 for Japanese, 187 for Koreans, 238 for Southeast Asians, 96 for Pacific Islanders, 126 for South Asians, and 160 for West Asians (student questionnaire items).

^c Answer categories are 1 = not at all to 4 = regularly.

^d Answer categories are 1 = seldom or never to 4 = almost every day.

^e 0 = not at all (popular, athletic, a good student, etc.), 2 = somewhat, 3 = very.

* Significantly different from whites at p < .05 level.

** Significantly different from whites at/> < .01 level.

*** Significantly different from whites at/) < .001 level.

them from whites statistically. In addition to South Asians, Koreans and Japanese families also have higher incomes than whites, while Chinese, Filipino, and West Asian families have comparable family incomes. Nonetheless, despite the lower socioeconomic status of the Southeast Asians (and Pacific Islanders), the share of those who have begun to save money for postsecondary education is comparable to that of whites. In addition, Southeast Asians manage to provide a place to study in proportions comparable to those of their other Asian counterparts even though they have fewer economic resources.

A greater share of South Asian, Japanese, Korean, and West Asian students take outside classes than white youth. South Asian households have more rules about watching television, maintaining grades, and homework than all other groups. With the exception of South Asians, other Asian ethnic groups generally have fewer rules about permissible kinds of programs or how late they may watch television, but more rules about how many hours they can watch daily than white eighth graders.

With respect to parent-child interaction about school, Chinese and Southeast Asian parents seem to have the least amount of interaction. Koreans also do not interact with their children as much as white parents, except in terms of how often they talked about post-high-school plans. With the sole exception of South Asians, whose parents spoke about plans for high school more often than white parents, parents from all Asian ethnic groups generally interact with their

140 American Journal of Education

Као

Korean	Southeast Asian	Pacific Islander	South Asian	West Asian
1.51*** (.54) .29 (.53)	1.57* (.56) .18* (.43)	1.31 (.69) .42 (.63)	1.66* (.48) .15* (.36)	1.38** (.60) .31 (.35)

children less about school experiences and were less likely to help their children with their homework.

In terms of student characteristics, Southeast Asians are most likely to be classified as limited English proficient, and all Asian ethnic groups had greater proportions of LEP students than whites. Pacific Islanders appear to have higher rates of grade repetition than whites, but their small numbers in the sample make this observed difference not statistically significant. Chinese, Filipinos, Koreans, Southeast Asians, South Asians, and West Asians are all more likely to be considered a good student than their white counterparts.

To sum up, the Asian population is heterogeneous but some behavioral similarities remain. No matter what the level of income (Southeast Asians report an income about half of that of whites), greater shares of Asians than whites report that they have a place to study in their homes. Other material resources, however, are harder to obtain. South Asians outrank all other Asian groups in parental socioeconomic status, educational aspirations, home educational resources, outside classes, and in the number of rules about homework and so forth. Southeast Asians, despite their lack of education and income, perform well. Pacific Islanders are the least different from whites, although they also provide children with a specific place to study in proportions comparable to other Asian groups: yet, they have the lowest educational aspirations. The upcoming analysis reveals how groups with comparable social and economic resources perform academically.

February 1995 141

Asian Americans as Model Minorities'? Results

Appendix A describes the variables used in the multivariate analysis and their operational definitions.¹⁰ To evaluate the hypothesis that Asians and whites differ in their levels of skill development, I analyze the background characteristics that affect math and reading test scores. Subsequently, to evaluate the hypothesis that Asians and whites differ in behavior, I examine the determinants of grades among Asians and whites. I also analyze how the achievement of Chinese, Filipino, Japanese, Korean, Southeast Asian, Pacific Islander, South Asian, and West Asian students compares to that of white students.

Test Scores

Table 6 presents the estimates of multivariate regression analyses using math and reading scores separately as the outcome variables of interest. We present two versions of each model--the first includes a dummy variable for Asians, which allows us to examine the mean effect of "Asian" ancestry, while the second model incorporates each ethnic group separately in order to examine ethnic differences. Results show that even after holding the effects of sex and parental socioeco-nomic status constant, Asians as a group earn higher math test scores but comparable reading scores relative to whites. But, when we expand the definition of material resources to include household structure, having an immigrant mother, and other resources, Asians and whites no longer differ in either math or reading test scores. Among the Asian ethnic groups, Chinese, Koreans, and Southeast Asian students earn higher math scores than whites from similar family backgrounds. Filipinos, Japanese, South Asians, and West Asians earn comparable math and reading scores, while Pacific Islanders score much lower than whites from comparable socioeconomic backgrounds. With the exception of Pacific Islanders, all Asian ethnic groups earn comparable reading scores.

It is striking that Pacific Islanders are less successful than the remaining Asian subgroups. What sets this group apart from their Asian counterparts may be their greater acculturation to the American mainstream. Our finding is consistent with recent literature that suggests that immigrant minorities are best positioned to achieve academically (Rumbaut 1990; Gibson 1993) because they have not yet adopted American peer norms. During a recent presentation, Margaret Gibson (1993) stated that many teachers like recent immigrant youth best as

142 American Journal of Education

Као

students. One teacher stated, "They're great when they first arrive, but after a few years, they get corrupted" (Gibson 1993).

Results from the expanded model produce substantively interesting results. Despite taking into account socioeconomic status, home resources and outside classes increased test scores while patterns of family interaction have no effect. This is expected, as the former two factors have to do with fostering intellectual activities, while the latter has more to do with encouraging enthusiasm for academic performance. Also, experiences of grade repetition and limited proficiency in English lowers test scores by three to four points.

Overall, our analysis finds no difference between Asian and white youth from comparable family backgrounds and home resources in the assessment of their skill development. But, subgroup differences within the Asian population were present, with Chinese, Korean, and Southeast Asians earning higher math scores than whites from comparable family backgrounds and home resources while Pacific Islanders consistently earned lower math and reading scores than whites, even after controlling for the effects of differential home resources.

Grades

Table 7 presents the effects of background variables on grades with separate models for Asian (both as a single group and as eight subgroups) and whites. Model 1 only includes measures of urbanicity, parental education, and family income, while model 2 adds measures of family structure, home resources, and student characteristics. I present separate models, as preliminary analyses suggested significant interactions between "Asian" and several independent variables. Pooled models were also examined, and I will discuss these results in table 8 in order to compare Asians and whites from similar backgrounds. Please note that because these regression analyses are weighted to account for the sampling scheme and varying response rates, the effective sample of Asians (and whites) is greatly reduced. This makes it extremely difficult for the effects of independent variables in the Asian models to reach statistical significance.

Overall, we can see that Chinese, Koreans, Southeast Asians, and South Asians earn significantly higher grades than West Asians (the baseline group) from similar family socioeconomic backgrounds. Filipinos, Japanese, and Pacific Islanders, on the other hand, earn grades similar to those of their West Asian counterparts. However, when we add measures of family structure, home resources, and student

February 1995 143

TABLE 6

Effects of Background Characteristics on Standardized Math and Reading Test Scores: Asians and Whites

		MODEL 1			MODEL 2
		With i	\sian		With

Оставить комментарий © Copyright Kao Grace (kazgugnk@yahoo.com) Обновлено: 08/08/2014. 209k. Статистика. Статья: США	Ваша оценка:

Связаться с программистом сайта

"Заграница"

Путевые заметки

Это наша кнопка

Kao Grace: другие произведения.

Asian Americans as Model Minorities? A Look at Their Academic Performance