Producing English lexical stress involves manipulating phonetic cues such as vowel duration, intensity, and fundamental frequency (F0). Existing literature presents diverse perspectives on how EFL learners utilize these prosodic features to realize stress (Fry, 1955; Modesto & Barbosa, 2019; Saha & Mandal, 2018; Zhang & Francis, 2010; Zuraiq & Sereno, 2021). However, inconsistencies in methodology and limited stimulus sets across previous studies hinder cross-study comparisons. Addressing this gap, the present study examines whether the use of these cues by Arab EFL learners remains consistent across disyllabic and trisyllabic words or varies according to syllabic structure and word length. A total of 65 Yemeni undergraduates at two proficiency levels and 10 American native speakers were recruited to produce 21 disyllabic and 21 trisyllabic words in which the stressed syllable was clearly marked. The analysis revealed that similar to American speakers, Yemeni EFL learners employed vowel duration, intensity, and F0 to distinguish stressed syllables from unstressed ones. Results showed significant differences in the stressed-to-unstressed vowel ratio between Yemeni learners and American speakers. American speakers exhibited a more pronounced reduction in vowel duration, intensity, and F0 for unstressed syllables compared to Yemeni EFL learners. Additionally, the study found that the use of phonetic correlates varies between disyllabic and trisyllabic words and across different proficiency levels, indicating that the production of English lexical stress is influenced by syllable pattern, proficiency level and word length. These findings highlight the complexity of stress production in EFL learners and suggest that instructional strategies should consider these variabilities to improve learners’ pronunciation skills.