Since the introduction of the expression "communicative competence" by Hymes in 1972, programs for the teaching and testing of English as a second or foreign language have increasingly been evaluated in terms either of their ability to promote communicative competence or of their sensitivity to a communicative view of language competence and performance. Unfortunately, despite this evident trend, there is no consensus on the nature of the communicative competence construct(s), nor has sufficient empirical support been provided for the various explanatory models proposed. The present study was conducted to survey the theoretical literature related to communicative competence; to identify major variables said to comprise the construct(s); to test for the comparative presence and measurability of such variables in typical native/nonnative speaker university academic communication; to propose a tentative model of communicative competence as a synthesis of these variables; and to examine the relationship of the TOEFL® test (Test of English as a Foreign Language™), TSE® (Test of Spoken English™), and TWE® (Test of Written English™) scores with the various elements of the tentative model. Accordingly, a schema was devised for the systematic elicitation of academic communicative language performance, and this schema was applied in the assessment of the 79 adult, English-as-a-second-language students who were chosen for the study. In all, 158 hours of video-recorded language-interaction episodes and 12 communicative writing samples were gathered by trained interlocutors, and observations on 40 communicative competence variables (21 in an oral modality, and 19 in a written modality, of which 18 in each modality are reported here) were rated by trained evaluators. Variables were examined for reliability and correlation with subtest and total scores of the TOEFL, TSE, and TWE examinations. Results provide information about the comparative contributions of some theory-based communicative competence variables to domains of linguistic, discourse, sociolinguistic, and strategic competencies. In turn, these competency domains were investigated for their relation to components of language proficiency as assessed by the TOEFL, TSE, and TWE tests. Twelve oral and twelve written communication tasks were also analyzed and rank ordered for suitability in eliciting communicative language performance.