This 3-year research project sought to determine whether language differences between Gullah-influenced and English speaking students in Charleston County (South Carolina) influenced test performance and subsequent interpretation of test results. A sample population of 503 educable mentally handicapped, learning disabled, and regular students (ages 7-14) were separated into two groups: those in each category who spoke standard English, and those who were Gullah-influenced. Correlations between Full Scale Intelligence Quotient (FSIQ) scores and mean language assessment scores were calculated for the two groups in each of the three categories. Results indicated that degree of control of standard English significantly affected student performance on the WISC-R (Wechsler Intelligence Scale for Children--Revised). Educational implications for possible misdiagnosis and placement are discussed in the context of an examination of the referral process in Charleston County. It is recommended that black students exhibiting Gullah influence in their speech patterns be assessed for special education placement in a manner which does not discriminate against their primary speech patterns. The degree of control of standard English must be determined. Rather than referral for special education, classes in standard English usage should be available to all students who need such instruction. (JW)