PURPOSE: A low birth weight (LBW) rate is associated with perinatal morbidity and mortality. Many different factors are related to birth weight and LBW. This study examined the effects of various factors on birth weight. METHODS: The study was conducted from January 1997 to August 1997 through direct interview with mother and medical history. Sample were drawn from 1,248 pairs of mothers and infants. The variables assessed included birth weight, gender, gestation age, birth order, maternal age, education, frequency of antenatal care, maternal complication and illness, parental smoking habit, alcohol and caffeine intake. The obtained data were coded and inputted into a computer using a SAS statistic package and estimated by multiple regression and logistic regression. RESULTS: The rate of LBW was 7.9% and premature rate was 3.4%. The variables affecting birth weight were maternal smoking habits, maternal complication and illness, gestation age, caffeine intake and gender. After adjusting for extraneous factors through multiple regression, we found that shortened gestation age, female infant, maternal complication and illness significantly played an important role in the decrease of birth weight. After logistic regression in the group of LBW infants, it was shown that the variables associated with LBW were maternal complication and illness, shortened gestation age, old age over 35 years, inadequate prenatal care, and female infant. CONCLUSION: Birth weight and LBW were more related to gestation age, gender, maternal complication and illness than maternal smoking habit, alcohol and caffeine intake.