Improving Test Performance Among Culturally Diverse Gifted Students
Harmon, D.
Understanding Our Gifted
Open Space Communications
Summer 2004

Written by Deborah Harmon, this article provides information on how to improve test performance across different cultures.

What are some practical recommendations for improving the test scores of culturally diverse students?

When trying to explain the discrepancy between the numbers of white students in gifted education programs and the number of culturally diverse students in gifted education programs, it becomes apparent that there are multiple factors at play with the most prominent one being the identification process. Most districts begin the identification process with nominations from teachers. Teachers are given a checklist of gifted behavioral characteristics based upon the ways giftedness presents itself in the dominant culture. Parents are also requested to complete a checklist of the same gifted behaviors described with language that is often unfamiliar to them. Achievement tests may be administered and, in many cases, an intelligence test is given. The data is reviewed and many culturally diverse gifted students are eliminated because they do not score high enough on behavioral checklists or achievement and intelligence tests (Ford & Harris, 1992).

While identification methods are beginning to include other kinds of assessments, more emphasis is usually placed on achievement and intelligence tests. To understand why certain groups may not perform well on standardized tests, we need to understand the history of these tests.

History of Intelligence Testing
Sir Francis Galton launched the testing movement in the late 1800s when he published his theory about intelligence. Galton believed that heredity determined intellect, placing Ancient Greeks as most intelligent, followed by Anglo-Saxons next, and Africans on the bottom. Alfred Binet and Theodore Simon developed the first cognitively based intelligence test in 1895. Lewis Terman, who also believed that heredity determined intellect, made substantial revisions to this test, norming it on American society and renaming it the Stanford-Binet intelligence test. Terman's sample size included 1,000 middle-class individuals who were of Western European descent. The Stanford-Binet was revised again in 1937 and in 1960. In 1972, it was normed on a population that included people of color.

According to Terman, children who score between 70 and 80 (average IQ score is 100) on the intelligence test represent the level of intelligence that is common among Native Americans, Mexicans, and African Americans. Terman described this level of intelligence as dull and suggested that these children be segregated and given an education commensurate with their innate ability (Valencia & Suzuki, 2000).

In 1916, a "new psychology" was introduced based upon the construct of individual differences with a focus on heredity and behavior. Studies comparing the differences in intelligence between whites and African Americans were conducted leading to the development of the "mulatto hypothesis," which suggests that lighter skinned African American Children are more intelligent than dark-skinned African American children.

In 1920, the army developed the National Intelligence Tests (NIT) to assess the intellect of soldiers. This test was standardized on a white middle-class population. Terman encouraged the idea of using intelligence tests for educational tracking and in the work force. During the next 10 years, many school districts used intelligence tests to determine on which educational track a student should be placed (Valencia & Suzuki, 2000).

Otto Klineberg, a professor at Columbia University, investigated the notion that African Americans are less intelligent than whites and disproved the hereditarian theory. KIineberg noted that those who are familiar with the language of the intelligence test are privileged and those who are not are at a disadvantage (Valencia & Suzuki, 2000).

George Isidore Sanchez, a Mexican American scholar, raised seven issues about intelligence testing that hold true today (Valencia & Suzuki, 2000).

  1. Intelligence tests are not normed on Spanish speaking people in this country.
  2. Test items are not representative of the Spanish-speaking culture.
  3. The entire nature of intelligence still is a controversial issue.
  4. Test results from Spanish-speaking individuals continue to be accepted uncritically.
  5. Revisions or translated tests are not necessarily an improvement on test measures.
  6. Attitudes and prejudices often determine the use of test results.
  7. The influence of testing on the educational system is phenomenal. (p. 21).

In the 1970s, Native Americans, Mexican Americans, and African Americans fought against the use of intelligence tests for placing students in special education. This led to a mandate stating that

  • multiple data must be used for assessment
  • assessment must be in the student's dominant language
  • everyone should have due process

Throughout the 1960s and the 1980s, there was concern about cultural bias against minority students. Cultural bias refers to those items on the test that are based on white middle-class experiences. Students who do not come from white middle-class backgrounds will not be familiar with those items and are at a disadvantage.

Looking at the history of intelligence tests, some themes arise.

  • Intelligence tests claim to identify the intellectual ability of an individual.

  • Intelligence tests are comprised of various tasks, many of which are based upon the experiences of white middle-class society.

  • Variance on intelligence tests is believed by some to be due to inherent ability which is characterized by race. Others believe that the environment and experiences of individuals impact performance on intelligence tests.

  • Intelligence tests have been and continue to be used to include or exclude students from education programs.

  • The development of standardized achievement tests was part of the evolution of intelligence tests. Standardized achievement tests suffer from the same ailments as intelligence tests, are also used to determine educational programming, and more recently have been used to determine funding for schools.

  • Both intelligence tests and standardized achievement tests consistently present a disparity between white middle-class students and students of color.

Standardization of the Test
Throughout the history of intelligence tests, problems nave existed with the process of standardizing or norming intelligence and achievement tests. The average score on intelligence tests was originally the average for white middle-class individuals - not everyone in society. Even as the norming of tests began to include students of color, samples did not re1lect the diversity found in classrooms. When adjustments of the norm were made for diverse populations, it was often viewed as "dumbing down" the tests. Subsequently, in the case of gifted education, where giftedness is usually identified as two standard deviations (30 points) from the average (middle-class white students), students of color are often not identified. Two standard deviations from the average of 100 is a score of 130. For students of color, the average is around 85-87 which means that these students must perform three standard deviations (45 points) from their average to be identified as gifted. According to Berlak (2001)/ the difference in the average scores has persisted over time on intelligence tests, norm-referenced tests, and proficiency tests from all test publishers, regardless of the grade level from kindergarten through graduate school. Differences are also seen within dominant culture students from lower socioeconomic levels. A most accurate norming of a test occurs when a school district standardizes tests on its own students.

The Language of the Test
The language of tests has a great impact on the performance of students. Tests are written in the language of the dominant culture-Standard English. Many culturally diverse students do not speak Standard English at home or at school.

In a research study examining the difficulties culturally diverse students face in test taking, 165 inner-city African American and Latino 4tfi- and 5th-grade students, including 14 identified gifted students, were asked how they thought they performed after completing three different achievement tests in October. The overwhelming majority stated that they weren't sure because they didn’t understand what the test was asking them to do. When asked to identify what instructions were problematic, it became apparent that the vocabulary and phrases used in the instructions were confusing to them.

A vocabulary list containing language that appeared on the tests was constructed with activities that allowed students to learn and use the vocabulary. Activities included word of the week, creating a synonym and antonym list for words, and requiring students and teachers to use the new vocabu1ary throughout the school day. Test taking strategies involving selecting the best answer, multiple choice questions, and determining word meaning were taught to all students. Mnemonics and creative raps assisted students in remembering grammar rules. Students learned how to manage their time when taking tests.

Test taking strategies were effective in helping students learn some of the language of the tests. The students were questioned about their tests the following April. All of them were able to determine whether or not they did well. They all reported that they understood the directions on the test and much of the vocabulary used in test items. Teachers reported improvement in most students' test performances. Students reported that the test taking strategies helped them not only on the achievement tests but also in their daily classroom work.

The Context of the Tests
There are great differences in the culture, language, and experiences of white middle-class society and the rest of society. When standardized tests use items and responses from white middle-class experiences on tests, they give an advantage to students who have had those experiences and make it almost impossible for those who have not.

The more an individual is unlike the white middle-class student, the greater the disparity between IQ scores. In recent years, major test publishers have reviewed and eliminated test items with obvious cultural and racial bias, yet many biased items still remain. According to Berlak (2001), there is a continuous 8-10 percent point gap between students of color and white middle-class students which suggests that the bias is actually systemic and structural. Fair Test National Center for Fair and Open Testing (2004) is an organization investigating the misuse of standardized tests. According to this organization, even when students have been equally prepared educationally, there are still large gaps in SAT scores when comparing gender and race.

Student Attitudes about Test Taking
Given the history of intelligence and achievement tests, it is not surprising that African American students as well as other students of color hold negative attitudes toward tests. Such attitudes, in turn, impact students' performance on standardized tests. African American students' negative perception of achievement and intelligence tests contribute to lowering test-taking motivation, greater anxiety, and poor performance on cognitive ability tests (Berlak, 2001).

Test Anxiety in Students
McKay and Doverspike (2001) suggest that test performance of African Americans may be affected by stereotype threat, a form of anxiety that results when a person is concerned that her performance might confirm a negative stereotype about her affiliated group. African Americans become anxious and fear that performing poorly on a cognitive measure will confirm the negative stereotypes about their cultural group not 15eing very intelligent. This anxiety interferes with concentration and attention. Research examining test anxiety among African American elementary students in a low socioeconomic school district found that 41% of the students experienced test anxiety. Students who experienced test anxiety also had lower achievement levels and lower perceptions of their ability to take tests (Townsend, 2002).

Teacher Expectations
As with achievement in the classroom, test performance is influenced greatly by teacher attitudes and expectations of students. Teachers need to become culturally competent so they can understand the needs and challenges of students who are culturally different from them. To become culturally competent, teachers must first become aware of their own culture, attitudes, and beliefs and then learn the cultures of their students.

Many teachers determine what to teach based upon their perceptions of their students' background and may unconsciously subscribe to the belief that culturally diverse students are not as intelligent as white students. With the emphasis on tests, teachers are often forced to teach test taking strategies. Popular strategies for raising the test performance of culturally diverse populations usually involve teacher-centered remedial instruction which is not very effective (Townsend, 2002). According to Hilliard (2000), what is needed is student-centered learning involving critical thinking and higher level thinking skills, including metacognitive thinking.

Although many public schools are developing fairer and more accurate alternative measures, the identification of gifted and talented students continues to rely' upon standardized tests (Alliance for Childhood, 2001). Therefore, in order to identify more culturally diverse gifted students, teachers must try to level the playing field by teaching students the language of the test, providing them with an equitable education, holding high expectations of them, and encouraging them to do their best.

Recommendations

  1. The demographics of the test sample should be examined and compared to the demographics of the students being tested. Many standardized tests publish ways to make adjustments for differences in the demographics of student populations. Consideration should also be given to the frequency of standardization of the test.

  2. While the concern of African Americans around bias on test items has been heard and attempts have been made to eliminate obviously biased items, consideration for other culturally diverse populations is needed. Eliminating bias toward other ethnic groups must include sensitivity to the nuances of language, attitudes towards teachers and education, and the role of the family (Arnold & Lassman 2003).

  3. Research has found that simply discussing intelligence or achievement tests as tools that give information about how learning occurs and how information is processed have an effect on test performance. Giving students a pep talk to let them know that they are prepared and capable of performing well on the test has been shown to decrease anxiety and increase test motivation (McKay & Dovespike, 2001).

  4. While all students may have concerns and anxiety about test taking, culturally diverse students experience more stress. Relaxation techniques are very successful in helping students deal with high levels of anxiety, and teachers need to employ such techniques in their classrooms.

  5. Students must be taught how to think critically, problem solve, and engage in metacognitive thinking.

  6. The language of the test should be taught. Teaching vocabulary and word meaning can be incorporated into existing curriculum.

  7. If a student doesn't have a clear understanding of English, that student should receive accommodations for language.

  8. Test-taking skills, strategies, and mnemonics should be taught to all students.

  9. Students should be taught to manage their time when taking tests.

  10. Students should experience many practice tests using test taking strategies and relaxation techniques.

  11. Accommodations should be made for those students who need to self-talk or move by placing them in areas of the classroom where they won't disturb other students.

  12. An appropriate, comfortable environment should be provided for students to take the test.

  13. Make sure that all students understand the directions of the test. If students do not understand, you must find a way to help them.

Deborah Harmon is Assistant Professor in the Department of Teacher Education, College of Education, Eastern Michigan University. Her focus is on the identification of and curriculum for culturally diverse gifted students.


Permission Statement

The appearance of any information in the Davidson Institute's Database does not imply an endorsement by, or any affiliation with, the Davidson Institute. All information presented is for informational purposes only and is solely the opinion of and the responsibility of the author. Although reasonable effort is made to present accurate information, the Davidson Institute makes no guarantees of any kind, including as to accuracy or completeness. Use of such information is at the sole risk of the reader.

Close Window