METHODS AND FINDINGS: Stratified random sampling design was used to select adolescents from 15 urban and rural secondary schools in Selangor, Perak and Kuala Lumpur, Malaysia. Data collection was carried out from 1st April 2014 to 30th June 2014. Information regarding socio-demographic characteristics, sun exposure and sun protective behaviours, clinical data and environmental factors were collected. Blood for total vitamin D was sampled. Descriptive and multivariate logistic regressions were performed. Total 1061 participants were analyzed (62% were female; mean age 15.1 ± 0.4 years). The prevalence of vitamin D deficiency was 33%. Mean vitamin D was lower in female (53 ± 15 nmol), obese (body fat percentage (≥25%m; ≥33.8%f) (56 ± 16 nmol/L), Malays (58 ± 18 nmol/L) and Indians (58 ± 15 nmol/L). In multivariate analysis, female (OR = 5.5; 95% CI: 3.4-7.5), Malay (OR = 3.2; 95% CI: 1.3-8.0), Indian (OR = 4.3; 95% CI: 1.6-12.0) and those always wearing long sleeve (OR = 2.4; 95% CI: 1.1-5.4) were more likely to have vitamin D deficiency. For female participants, ethnicity {Malays (OR = 6.7; 95% CI: 2.0-18.5), Indian (OR = 4.5; 95% CI: 1.8-19.3)} was an important risk factors. Cloud cover, school residence, skin pigmentation, sun-exposure and sun-protective behaviours were not significant risk factors. The limitation of this study was recall bias as it relied on self-reported on the sun exposure and protective behaviours. The diet factors were not included in this analysis.