In this article, we describe a comprehensive model for exploring the determinants of birth-weight outcomes among Mexican-American women from the Arizona Perinatal Acculturation Project. Data for this article came from a longitudinal study consisting of two phases. In phase one, a detailed prenatal survey was administered to 500 pregnant women. Phase two consisted of a postnatal survey administered to the women at least three months after they delivered (N = 269). Subjects who provided data were recruited from two health care agencies. Separate model building processes were conducted for a continuous measure of birth weight, and a dichotomous indicator of low-moderate birth weight (<2900 grams) using multiple linear and logistic regression analyses, respectively. The potential predictor variables for the models were divided into twelve predictor sets. The results showed that both final models included a combination of biological/behavioral factors, as well as protective sociocultural factor indicators. Acculturation status, one of the primary variables of interest in the study was found to be important for predicting birth weight and low-moderate birth weight. This result did not change when low birth weight (<2500 grams) infants were removed from the analyses. Low acculturation status was found to be associated with better birth-weight outcomes than high acculturation status. Surprisingly, length of US residence had an opposite effect in predicting both birth-weight indicators when compared to acculturation status. These results suggest that the relationships between acculturation and birth outcomes should be redefined to take into account the complexity of the phenomenon of acculturation in addition to the measurement of an array of family and sociocultural factors.