Integrative, Critical Review of the Literature


Schooling has historically not considered the possible educative value of including native culture and languages and has instead emphasized integration into the mainstream, with the result that many cultural minority students may not be achieving the success they might otherwise achieve.  In many cases, their families have come to mistrust public school. Given this situation, this Chapter will critically review research studies that help to identify practices to help cultural minority students achieve while utilizing the resources of their home culture and language.

The Effect of Tracking and Stereotyping by Teachers

A study by Elhoweris et al. (2005) sought to find out if a child’s ethnicity affected whether or not his/her teacher would refer them for gifted and talented programs. Two hundred and seven teachers in a large Midwest school district were given a short vignette about a student who possessed gifted characteristics. Ninety two percent of the teachers were female, and 83% were white. One third of the vignettes revealed the student in question was white, another one third revealed the student was black, and one third gave no racial information as a control group. The teachers were randomly assigned to one of the 3 groups. The teachers were asked to decide if the student should be referred to a gifted and talented program. Ethnicity had a significant effect, (p!.05), on the teacher’s decision. Even though all the information about the student was exactly the same except for the ethnicity, the African American student was rated the lowest of the three groups.

This study (Elhoweris et al., 2005) is not as generalizable as it could be because it was conducted only in the Midwest. Racial attitudes vary around the country, and may actually be more severe in some places, and less severe in some places. I think also this could be applied to other minorities. It would be interesting to compare the rates for different minorities’ referrals, especially Asian students, who are typically assumed to be model students, even when they do not display the characteristics of such a student.

Hosp and Reschly (2004) investigated the connection of race and ethnicity to placement in special education programs, rather than the gifted and talented programs Elhoweris et al. (2005) looked at. Hosp and Reschly looked to discover what the predictors for the overrepresentation of minority students might be besides race, and so factored academic achievement in to the equation as well as demographic information (race and ethnicity) and economic information.

The researchers (Hosp & Reschly, 2004) looked at the rate of five different ethnic groups of students being assigned to special education, African American, Native American, Asian Pacific Islander (APA), Hispanic, and Caucasian.  Programs for students with mental retardation, emotional disturbance, and learning disabilities were analyzed. African American students were shown to be overrepresented in classes for the mentally retarded as well as programs for students who are emotionally disturbed. Native American’s were overrepresented in classes for the learning disabled. Less APA students are identified in all three categories than would be expected considering their percentage of the general population.  As Ruan (2003) found, teachers often overestimate the abilities of APA students due to stereotyping, which can disadvantage Asian students who do need the extra help but are not identified and given intervention. Hosp and Reschly (2004) also reported that African American, Hispanic, and Native American students do not make up a proportion of the students in gifted and talented programs that would be expected given their proportion of the general population.

For this study (Hosp & Reschly, 2004), data on the demographics of students enrolled in special education programs was collected. The ratio of the percentage students in programs for each of the three disability categories to the percentage of each ethnicity in the general population was compared to the same ratio for white students to figure out the relative risk ratio. Achievement statistics were gathered from school districts and their websites. Data that was compatible with the design came from only 16 states, but the researchers stressed that these 16 states represented all the major regions of the United States. For the study, “due to the large number of comparisons, an alpha level of p= .005 was used” (Hosp & Reschly, 2004, p. 190).

For each of the three groups, mental retardation, learning disabilities, and emotional disturbance, the variance for all the racial groups was 32.8 %, 24.4%, and 30.1% respectively (Hosp & Reschly, 2004). Economic factors were stronger in determining special education membership than race was, but in some ways it was found to correlate with race, so it is difficult to differentiate the causality coming from one or the other. The academic consideration was a strong influence for only 2 of 12 categories (categories consisted of percentage of students in special education in relation to the number in the general population of the school for each racial group compared to the ratio for white students, for each of the three special education types). The other ten categories, it “accounted for a significant amount of variance for six of the models… [but] for the remaining four models, the academic block did not contribute a significant amount of unique variance” (Hosp & Reschly, 2004, p. 192).

The findings report that for mental retardation special education classes, economic factors were the strongest influence out of economic, demographic, and academic achievement categories (Hosp & Reschly, 2004). The variance was .27 for African American students, .21 for Hispanics, .246 for APA, and .162 for Native American, p <.005. For emotional disturbance, race (demographic) was the most statistically significant influence (variance =.193 for African American students, .313 for Hispanic, .140 for APA, and .259 for Native American students, p<.005).

Academic performance most affected referral and membership in programs for the learning disabled (Hosp & Reschly, 2004). Results were an independent variance of .228 for African American, .031 for Hispanic, .137 for APA students, and .078 for Native Americans.

APA students had the strongest predictor as race in all three special education categories, while the majority of cultural minority students seemed to have significant influence from all three student identifiers (race, income, and achievement) (Hosp & Reschly, 2004). For all the racial groups, academic performance seemed to affect membership in the special education groups less than economics or race, but was slightly stronger in affecting placement in classes for mental retardation.

Racial demographics were stronger for African American and APA students than for Hispanic or Native American students (Hosp & Reschly, 2004). While the academic predictor was the weakest overall, it did contribute significantly to the placement of students in 8 of the 12 groups.

The study (Hosp & Reschly, 2004) eliminated small districts from the sample that only had a few minority students. This was because in a district with under a certain number of students of a specific ethnicity, officials are not allowed to report test scores to the public, because the pool of students is so small the confidentiality of reporting it is does not meet privacy standards. It is much easier to figure out who is who out of a group of 5 students than it is to identify one student out of 50 or even 500. This was necessary because they could not obtain the information, so thus there was no way the researchers could include it in the results. The study is thorough for the information attained, but since small districts could not publish the information, it is does not paint a complete picture.  Patterns of enrollment may be different in different types of districts.  It could be that in these rural districts with few minority children that there is even less culturally relevant teaching that provides a chance at success for these students.

A critique of the study is also addressed by the researchers (Hosp & Reschly, 2004). The fact that the research was done far removed from any individual student’s achievement makes it hard to identify exactly where students who should not be in special education are being enrolled in these classes. The researchers suggest that “research needs to be extended to the individual level” (p. 196). The fact that the individual students were not assessed by researchers seems problematic, because they could not assess student’s in the same way teachers could, they only looked at a few factors that may well be, unfortunately, associated with race. By not assessing any students themselves, we as readers cannot tell whether race or class alone, and not academic performance, was what drove the teachers to assign a student to special education classes.

To summarize, ethnicity has a negative effect on whether a teacher will refer a student to gifted and talented program, even when all other factors are the same (Elhoweris et al, 2005). Conversely, it has been shown that African American students are referred to classes for the emotionally disturbed more than white students, and low socioeconomic status was a strong factor in recommendation to classes for the mentally retarded(Hosp and Reschly, 2004).

The Effect of Class and Socioeconomic Status on Teaching and Learning Craig, Connor, and Washington (2003) found that African American students

from low income families who attended state funded preschools performed better in their oral language and cognitive skills by the time they reached third grade than middle class African Americans who did not attend these preschools. In the Detroit school the study was conducted in, seventy five percent of the students were African American. All subjects spoke African American English (AAE). Fifty students were involved in the study; 30 boys and 20 girls. Half were in kindergarten and half were in the preschool class. The middle class students who started in kindergarten did not attend preschool.

The researchers (Craig et al., 2003) pre-assessed the children for oral language and cognitive skills when they first arrived at the school, and conducted formative assessments along the way. The assessments were conducted by African American female examiners. These examiners spoke African American English with the students during the tests, which were audio recorded to check for reliability. Assessments used were subtests of the Kaufman Assessment Battery for Children and samplings of students use of expressive language in describing pictures. Samples were scored by segmenting responses into communication units and analyzing them for complexity of syntax, diversity of vocabulary, and mean length of each unit. The computer program Computerized Language Analysis was used to evaluate the responses.

Students in preschool tested lower when they first started school, possibly because they were younger, and showed no significant improvements at the end of first grade (Craig et al., 2003). But by the end of third grade, the lower SES preschool students had surpassed the children who started in kindergarten and did not attend preschool in reading comprehension skills, the slope difference in the improvement between the groups was 6.68 (p<.001 ). The study found that coming from a family with low socioeconomic status affected a student’s reading acquisition less than an early diagnosis of reading difficulties. Students who had had their special needs addressed a year earlier were at a better place by the time they reached third grade regardless of their socioeconomic status than students who may have had these issues addressed a year later.

The study (Craig et al., 2003) concluded that with proper intervention, social class should not make a major difference in a student’s success. One critique of this study is the fact that students were tested upon their entry into school, the lower SES group a year earlier when they went into preschool. Comparing the students at different stages in their development may not make for accurate contrast of the two groups, since with age oral language develops, especially distinguishable in young people,. However, testing all students a year before kindergarten started, or testing all students at the inception of kindergarten and comparing them this way would be more accurate.