A comprehensive examination of emoji usage in Mexican Spanish WhatsApp corpus: a mixed-methods Linguistic approach

The surge of emojis in computer-mediated communication (CMC) since 2011 presents a significant analytical challenge across various disciplines, such as linguistics, neurophysiology, psychology, computer science, and marketing. Departing from earlier works that exclusively utilized Natural Language Processing (NLP) techniques or engaged in the analysis of experimental or selectively curated social media texts, this study embraces a broader linguistic scope. By integrating NLP extraction and linguistic analysis of emojis within a specific language, in a one-to-one CMC corpus, this paper delivers an extensive examination of emoji usage in Mexican Spanish WhatsApp messages, uncovering underlying cultural nuances, gender differences, and a variety of pragmatic functions. The corpus contains 1,487,503 messages from a gender-balanced group of male and female participants finding that females used emojis in 17% of messages, whereas males used them in 11%. Comparing emoji usage trends revealed that while most commonly used emojis usually reflect global patterns, some culture-specific deviations exist. Out of 1049 distinct emojis, 34 were deemed significant in predicting gender-based usage, taking into account a minimum occurrence (0.8% of messages). Variables like message length, emoji positioning, and sentiment were also analyzed. This study also examines emoji Mexicanisms and employs a semiotic approach to emojis under the framework of Peirce’s theory. Additionally, a structured codebook that categorizing emoji functions is provided. Employing a mixed-methods approach, this study elucidated gender-based variations in emoji usage patterns across various functions, including discourse management, substitution, and reiteration.