Individual differences in general and special education assessment

¶ … Reliability and Validity-- Responding to Diversity: Fairness in Assessment

When discussing the statistical reliability and validity of tests that attempt to identify students with special needs, it is critical that the tests also take into accounts students' diverse cultural as well as academic idiosyncrasies and needs, suggest the section "Responding to Diversity: Fairness in Assessment," of Libby G. Cohen and Lauren Spencier's text Assessment of Children With Special Needs. (Addison Wesley, 2004) The authors suggest that an assessment test may be considered valid by some educators simply because it meets norm-referenced criteria regarding standardized assessment of students, that is, because the test assesses the student according to test's validated and accepted criteria about relatively 'normal' the student is cognitively, compared to an group of his or her peers.

However, Cohen and Spencier remind the reader that merely because a test adheres to the concepts of standard error of measurement and confidence intervals in creating such a 'reliable' norm does not automatically make it a fair exam in cultural terms. For example, what if a child does not speak English as his or her primary language at home, yet the test of intelligence is primarily verbally based, and administered in English? Furthermore, who constitutes the evaluated student's peer group as the norm he or she is to be compared to, when evaluating the test results? Is it from the student's age group, nationally, or according to standardized psychological norms, or from the student's own cultural, social, and demographic background, given the differences that may affect how a student responds to the test? The ability to assess differences in a child's culture from the test designer and the administrator that are culturally determined, but do not critically affect the child's ability to learn is a crucial component of sensitive and accurate testing, according to the authors, and an important caveat for educators to remember.

Summary: Chapter 4. Reliability and Validity-- Responding to Diversity: Fairness in Assessment

The question of the usefulness of assessment tests, as well as test scoring procedures is called into question by the evident racial and differences of many test components, however unintentional these potential sources of cultural bias may be. For example, a test that requires students to make use of vocabulary words only pertinent to certain areas of the country, whether rural or urban (a city child may have never seen a cow, or know that a cow and a bull are the same animal) might result in poorer assessment of that child than is warranted. A Caucasian child might not be asked to describe common Vietnamese foods, but a recent Vietnamese immigrant might be called to do so on an intelligence test.

This is hardly culturally fair in assessing intelligence, even if children who are immigrants or bilingual might benefit from additional resource room help. Remember, even though a bilingual child may need English help, this is not a reflection of his or her general intelligence quotient, even if poor English ability may result in a lower test score on an English-administered test. After all, an affluent child would likely have a poorer score, if tested in Spanish -- why should a recent immigrant be subject to the same demands? The solution is to test the child for intelligence in the language he or she is most comfortable speaking, and supplement English classroom help with resource ESL aids, rather than more general educational resource room help.