Construct validity concerns for the IPT I Oral Grades K-6 Spanish Second Edition (IPT-S) as a measure of native oral language proficiency are examined. The examination included describing a subset of items that contributes most to overall score and native-language proficiency designation. Correlations between this subset of items and the overall score are calculated. Empirical evidence suggests that if a student cannot or does not answer four questions from the second section in a complete sentence, then the student will either be labeled as a non-speaker or a limited speaker in the native language. The authors argue that the requirement that speakers respond in complete sentences reflects a naive view of language proficiency inconsistent with linguistic research, and characterize the requirement as "academic bias," that is, a prejudice resulting from a confusion of academic content knowledge related to language arts with actual linguistic ability, having the result of arbitrarily favoring members of the educated classes. The authors conclude that the IPT-S should not be used for Spanish language assessment with Spanish-speaking children. (Contains 1 figure and 5 tables.)