This paper reports on a study comparing oral English proficiency as conventionally measured by instruments currently used in many school districts throughout the United States, with Spanish language abilities of individuals with Spanish language background. Oral language proficiency is defined as the results of a quantitative measure applied to a speech sample, while language abilities are defined as what a speaker can actually do with the language. The first section of the paper deals with theoretical and practical concerns in language proficiency assessment (LPA), with emphasis on the content of language proficiency and the effect of situational context on language. The focus of the study reported in the second part of the papers is how language proficiency measures currently used to classify speakers relate to those speakers' language abilities. For this purpose, fifth and sixth graders of Hispanic background, generally from bilingual classrooms, were chosen as subjects of the research. The major concern of the research was to sample and describe the language abilities of these students in both Spanish and English. The final phase of the study was a test situation, that is, simulation of usual conditions of LPA administration. The results were compared with speech obtained in situations in the previous phase of the study. (AMH)