A Confirmatory Multivariate Study of the Nature of Second Language Proficiency and Its Relationship to Learner Variables (Acquisition, Covariance, Analysis, Structural Equation Models, Lisrel, Foreign, Teaching English as A Foreign Language, Tefl)

107 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1985.The purpose of this study was to investigate both the nature of second language proficiency and its interrelationships with learner variables using a wide range of language proficiency measures and powerful multivariate statistical techniques. First, competing hypotheses regarding the nature of second language proficiency were tested against each other using confirmatory factor analysis. Second, the relationships between different aspects of language proficiency and selected learner variables were investigated using causal modeling techniques. Regarding the nature of second language proficiency, a correlated traits hypothesis was tested against a second-order hypothesis. As for the relationships between proficiency and learner variables, four hypotheses were tested against each other: (a) background learner variables ('formal learning in home country,' 'home country and family use of English,' and 'formal learning in an English-speaking country') and motivation affect proficiency; (b) background learner variables affect proficiency while the latter affects motivation; (c) background learner variables affect proficiency while proficiency and motivation reciprocally affect each other; and, (d) background learner variables affect proficiency while motivation and proficiency are unrelated.A sample of 334 subjects were administered ten language tests in addition to a learner variables questionnaire. The results of the confirmatory factor analysis showed that both a correlated traits and second-order hypotheses represented the data to precisely the same degree. In addition, causal modeling analysis revealed that of the four models representing the interrelationships between proficiency and learner variables a model hypothesizing no relationships between the motivational and proficiency variables accounted for the data best. The results, their implication for language teaching and testing, and directions for future research were discussed.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD