Volatile organic compounds (VOCs) are an important public health problem throughout the world. Many jmportant questions remain to be addressed in assessing exposure to these compounds. Because they are ubiquitous and highly volatile, special techniques must be applied in the analytical determination of VQCs. Personal exposure measurements are needed to evaluate the relationship between microenvironmental concentrations and actual exposures. It is also important to investigate exposure frequency, duration, and intensity, as well as personal exposure characteristics. In addition to air monitoring, biological monitoring may contribute significantly to risk assessment by allowing estimation of absorbed doses, rather than just the external exposure concentrations, which are evaluated by environmental and personal monitoring. This study was conducted to establish the analytic procedure of VOCs in air, blood, urine and exhaled breath and to evaluate the relationships among these environmental media. The subjects of this study were selecled because they are occupationally exposed 10 high levels of VOCs. Environmental, personal, blood, urine and exhalation samples were collected. Purge & trap, thermal desorber, gas chromatography and mass selective deteclor were used to analyze the collected samples. Analytical procedures were validated with the ""break through test"", ""recovery test for storage and transportation"", ""method detection limit test"" and ""inter-laboratory QA/QC study"". Assessment of halogenated compounds indicled that they were significantly correlated to each other (p value < 0.01). In a similar manner, aromatic compounds were also correlated, except in urine sample. Linear regression was used to evaluate the relationships between personal exposures and environmental concentrations. These relationships for aromatic and halogenated are as follows: Halogens_(personal) = 3.875 +0.068Halogens_(environmet) (R² = .930) Aromatics_(personal) = 34217.757 - 31.266Aromatics_(environmet) (R² = .821) Multiple regression was used to evaluate the relationship between exposures and various exposure determinants including, gender, duration of employment, and smoking history. The results of the regression modeling for halogens in blood and aromatics in urine are as follows: Halogens_(blood) = 8.181 + 0.246 Halogens_(personal) + 3.975Gender (R² = .925), Aromatics_(urine) = 249.565 + 0.135Aromatics_(personal) - 5.651D.S (R² = .735), In conclusion, we have established analytic procedures for VOC measurement in biological and environmental samples and have presented data demonstrating relationships between VOCs levels in biological media and environmental samples.