In this analysis of intelligence testing of minority group children, the implications of inadequate testing practices are discussed. Several aspects of test design are examined: deficiencies in intelligence testing, cultural bias, construct validity, and diagnostic utility. A sample set of results derived from a Stanford-Binet test administered to 257 respondents is examined: statistical data are included. The author concludes that "investigations of cultural biases in intelligence testing have established the fact that the most widely used test procedures are 'penalizing' for non-Anglo, lower socioeconomic groups." Teachers are cautioned of the dangers in using group-test results to plan academic programs geared to individual needs. (RL)