Other Sources of Bias
While significant progress has been made in development of unbiased tests, other sources of bias may present an obstacle to educating culturally diverse stu- dents. Teachers’ different expectations for students of various racial or social class- es are correlated with less effective instructional practice (Leacock, 1969; Murray, 1996). Teachers lack the knowledge and skills to successfully interact with stu- dents who are different from themselves (Ladson-Billings, 1995). Teachers often fail to identify potentially qualified students for selective programs in schools, especially when students are culturally different from them (Peterson, 2000).
It is important for teachers to familiarize themselves with common phras- ings of test questions and to be more aware of the various linguistic patterns of the many ethnic/cultural groups represented by the students in their classrooms. Teachers forced to use standardized tests can help students practice their lan- guage development resulting in improved test taking skills (Shields, 1997). Native American students whose teachers integrated local standards with state standards and aligned curriculum with assessment through a portfolio process have made achievement gains (Koelsch & Trumbull, 1996). Demmert (2001) suggested that teachers maintain linguistic and cultural congruence between home and school, educate students in their native language, and use local knowledge and culture in the curriculum to improve the academic performance of Native American stu- dents. For English language learners, extra time and/or dictionaries should be provided or the language can be modified (Nelson-Barber & Trumbull, 2007). Teachers should pay attention to students’ test-taking needs and accommodate their cultural backgrounds in order to increase the validity of test scores.
Cultural Bias in Assessment | Kim, Zabelina | 135
Educators should be aware of all potential sources of cultural bias so that they can support students from different cultures academically and psychologically. If test results are interpreted without consideration of cultural and educational factors of certain groups’ the scores may inaccurately reflect the actual ability and knowledge of those students. The results may be biased even if the tests themselves appeared to be unbiased (Skiba, Knesting, & Bush, 2002). Therefore, teacher training in culturally responsive pedagogy is necessary (Klingner et al., 2005).
ALTERNATIVE ASSESSMENTS
Alternative assessments, in contrast to traditional or standardized tests, are considered more culturally fair than paper and pencil tests. Power tests that mea- sure how much the students know are considered more culturally fair than speed tests that measure how much the students answer correctly within a time limit. Oral instructions are more culturally fair than written instructions, non-verbal contents are more culturally fair than verbal contents, and familiar contents to the students are more culturally fair than unfamiliar contents. Therefore, among alternative assessments, the most culturally fair assessments would be non-timed, hands-on performance assessment on familiar contents with oral instructions, such as group projects, informal assessments, or interviews.
Group Projects Group projects are often used as one alternative assessment method. However,
the fairness of group assessment results may depend on students’ cultures: African American (Aronson & Bridgeman, 1979; Berry, 2003) and Mexican American (Aronson & Bridgeman, 1979) students tend to learn well through collabora- tive learning methods, whereas Euro-American students do not. A meta-analysis (Rohrbeck, Ginsburg-Block, Fantuzzo, & Miller, 2003) found that peer-assisted learning is most effective with young, urban, low-income, and minority students. However, female students, working-class students, students of color, and students who are considered low achievers by their teachers and peers tend to be marginal- ized and prevented from engaging in meaningful discussions with their groups (Esmonde, 2009). Further, Native American students’ willingness to participate is diminished by the competitive nature of group assessments (Nelson-Barber & Estrin, 1995; Swisher & Deyhle, 1992).