<![CDATA[Code-mixing is a common phenomenon in bilingual and multilingual contexts, occurring not only in daily communication and on social media but also increasingly in educational settings, particularly in speaking-class activities. This study aims to investigate the types of Indonesian–English code-mixing and the reasons and factors influencing its use in students’ speaking class activities. The research employed a descriptive qualitative method. The data were collected from students’ speaking class activities through classroom video recordings and unstructured interviews. The recorded utterances containing code-mixing were transcribed into written data and subsequently analyzed by type of code-mixing. The data were then analyzed to identify the reasons and factors underlying students’ code-mixing. The findings reveal that students produced three types of code-mixing: insertion (34 data), alternation (26 data), and congruent lexicalization (3 data). Insertion emerged as the most dominant type, indicating that students tend to embed English lexical items within Indonesian sentence structures. The reasons for using code-mixing include talking about a particular topic, being emphatic about something, inserting sentence fillers or interjections, repetition for clarification, and clarifying speech content for the interlocutor. Furthermore, factors influencing code-mixing include speaker–interlocutor interaction, limited mastery of vocabulary, and classroom communication needs. These findings suggest that code-mixing functions as a communicative and pedagogical strategy that facilitates interaction, comprehension, and students’ active participation in speaking activities. This study contributes to the field of language education by highlighting the positive role of code-mixing in bilingual classroom contexts. ]]>
