<![CDATA[Sociolinguistics examines how social norms influence daily language usage, including code-mixing. This study investigates the types of code-mixing used by first-semester English Language Education students at Universitas Maritim Raja Ali Haji during speaking activities, as well as the underlying factors driving its use. Utilizing a descriptive qualitative design, the study targeted a total class size of 23 students, among whom 17 students produced code-mixing instances totaling 30 occurrences across classroom observations. Eight students were selected via purposive sampling for in-depth semi-structured interviews. Data were analyzed using the interactive model of data condensation, data display, and conclusion drawing/verification by Miles et al. (2014), alongside Muysken’s (2000) theoretical framework of code-mixing typology. The results reveal that insertion was the most dominant type of code-mixing (21 occurrences), followed by alternation (7 occurrences) and congruent lexicalization (2 occurrences). The primary driver of code-mixing was identified as a lack of vocabulary, serving as a compensatory strategy, alongside secondary factors including language habits, confidence/anxiety, and interlocutor dynamics. Theoretically, these findings reinforce sociolinguistic perspectives that code-mixing functions as a natural communication strategy for early-stage language learners. Practically, the study highlights the importance of systematic vocabulary enhancement and supportive classroom environments to build students' speaking confidence.]]>
