Although investigations of bilingual effects on cognitive control have proliferated in recent years, results remain inconsistent. A prevailing explanation for these mixed results lies in the ignorance of the complex, multidimensional nature of bilingualism across previous studies. To address this issue, the present study examines bilingual effects from a holistic perspective, accounting for multiple levels of linguistic and non-linguistic factors including individual, interpersonal, social, and ecological. 148 undergraduate Mandarin-English bilinguals were asked to complete the Flanker task and Wisconsin Card Sorting Test to measure conflict monitoring, inhibition, and mental set shifting. Linguistic and non-linguistic characteristics were assessed through questionnaires and interview. Results of stepwise multiple regression analyses revealed that years of English use predicted conflict monitoring, English proficiency predicted conflict monitoring and inhibition, and English-speaking time in class predicted mental set shifting. These results support that factors at multiple levels impact differently on separate dimensions of cognitive control, highlighting a need to better measure and account for linguistic and non-linguistic variability in bilingual samples.