The association between various sociodemographic variables and the presence of anti-Shigella sonnei lipopolysaccharide (LPS) antibodies was examined in a random sample (N = 383) of male Israeli conscripts. Of the male conscripts, 190 (49.6%) had pre-existing antibodies against S. sonnei LPS (defined as HA titres of greater than or equal to 1:10 after treatment of sera with 2-mercaptoethanol). Univariate analysis revealed a significant positive association between the presence of humoral anti-S. sonnei LPS antibodies and sociodemographic variables including Eastern origin (p = 0.007), low socioeconomic status (p = 0.0016), and the number of siblings (p = 0.023). When multiple logistic regression was used to control simultaneously for the effects of the other variables, ethnic origin emerged as the strongest correlate of anti-S. sonnei LPS antibodies. On the other hand, the association of the sociodemographic variables in subjects suffering from S. sonnei infection during their military service, was in the opposite direction (p less than 0.001 for both socioeconomic status and ethnicity). These findings suggest differences between subpopulations in acquired immunity to S. sonnei due to differences in exposure to the homologous organism prior to military service.