Beyond Boundaries: Multilingual Code-Mixing in WhatsApp Conversations Among EFL University Students

Abstract

This study investigates the types and functions of code-mixing in the informal WhatsApp conversations of EFL university students at the English Tadris Study Programme (TBI 4C), IAIN Bengkulu. Using a qualitative descriptive approach, 34 naturally occurring utterances were collected from seven Semester 4 students through non-participant observation over one month (April 12 to May 12, 2026). The data were analyzed using Muysken's (2000) typological framework, which classifies code-mixing into Insertion, Alternation, and Congruent Lexicalization, and Hoffmann's (1991) functional framework, which identifies the communicative purposes behind code-mixing. The findings show that Insertion was the dominant type (85.3%), followed by Congruent Lexicalization (8.8%) and Alternation (5.9%), with students mostly embedding English words into an Indonesian or regional language base. In terms of function, lexical gap filling was the most frequent (50.0%), followed by affective and emphasis function (35.3%), group solidarity (11.8%), and prestige and affiliation (2.9%). These results suggest that code-mixing among these students is not a sign of linguistic deficiency, but a purposeful and sophisticated communicative strategy that reflects their trilingual identity across Indonesian, Bengkulu dialect, and English in everyday digital communication.