This paper investigates the issues related to code-switching/code-mixing in an ESL context. Some preliminary data on Bangla-English code-switching/code-mixing has been analyzed in order to determine which structural pattern of code-switching/code-mixing is predominant in different social strata. This study also explores the relationship of language use to the socioeconomic class of the language user. The redefined concept of modernization, the increased number of cross-cultural contact and the need for language innovation play a dominant role in the language user's application of code-switching/mixing that determine the socioeconomic rank. Out of four major types of code-switching/mixing, our focus of interest is on "inter-sentential mixing" and "intra-sentential mixing" i.e. where elements are mixed from both languages that are used in the same sentence and/or in the same conversation. The context and factors that lead to the motivation of using L1 and L2 in a social milieu are also explored in this paper. Our findings suggest that the users are concerned about the language during their speech in order to establish and/or to realize social function, pragmatic function, and metalinguistic function. Banu and Sussex (2001) conducted a research on code-switching/missing in the context of business names only. To our knowledge, no work has done on English-Bangla code-switching/mixing with predominant structural pattern in relationship with different social strata in the context of Bangladesh. That drive and motivation has worked on to conduct and explore the code-switching/mixing structural pattern in different social rung of Bangladeshi community. (A bibliography is included.)