A study examined the morphosyntactic mechanism of a common code-switching (CS) pattern, the use of an English adjective (content morpheme) + Korean "-ita" (a system morpheme meaning "be") in the speech of Korean-English bilinguals. Data were drawn from audiotaped conversations of three subjects with their family members or bilingual friends and from CS utterances selected by the researcher from daily conversations. The discussion of results addresses two issues. First, the data did not suggest a theory of bound morpheme and equivalence constraints to be universally applicable to explanations of CS phenomena, but rather demonstrated that a matrix language frame model was more explanatory for Korean and English morphosyntax. Second, two possible explanations for the production of this pattern emerged: (1) a tendency toward nominalization in code-switching, or (2) a transfer of English grammar to Korean-based code-switching. (MSE)