Introduction:
Values of birth weight and percentage contribution of newborn babies with body weight below 2,500g (Low Birth Weight LBW) are mentioned among the indicators assessing the biological state of society.

Aim:
The objective of the study was assessment of the probability of low birth weight in newborns according to risk factors, such as maternal smoking during pregnancy, mother’s calendar age and education, as well as gestational age, using logistic regression.

Material and Methods:
The study was conducted during 1999–2007 in a group of 3,071 healthy and physically fit male and female students (and their mothers) of the first year of full studies at the Jędszej Śniadecki University of Physical Education in Gdańsk. The data was obtained from medical records (retrospective study), and a questionnaire designed by the author. The following tests were applied: Shapiro-Wilk, Mann-Whitney, χ2, Kolmogorov-Smirnow and Spearman rank-order correlation coefficient and analysis of the logistic regression.

Results:
Statistically signification relationships were found between maternal smoking during pregnancy and low birth weight of the baby (OR=0.36), and between times of pregnancy and low birth weight of the baby (OR=2). No significant relationship was observed between calendar age and education of the mother and the probability of occurrence of LBW in newborns. The created logistic regression model takes the form: Logit P=-25.57+1. x1 (gestational age) -1.02. x2 (smoking during pregnancy

Conclusions:
Special attention should be paid to the prevention of LBW, including the shaping of health promoting behaviours, popularization of knowledge concerning causes of LBW, and health consequences of the LBW. Logistic regression is perfectly suited to estimate the probability of LBW.