The association of herd- and sample-level factors with the isolation of Salmonella group B from cattle fecal samples was analyzed. Study farms were 65 dairy herds with a recent history of laboratory-confirmed clinical salmonella infections. Herds were visited once per month for three months to collect data and samples for bacteriological culture. Herd size varied widely from 34 to 3700 total cattle on the farm (median=370). Salmonella serogroup B was isolated from 270 of 2726 samples tested. The predominant serotypes identified were S. Typhimurium and S. Typhimurium var. Copenhagen. Logistic regression was used to analyze the relationship between potential risk factors and isolating Salmonella serogroup B. The only herd-level factor which was significantly associated with fecal shedding was total herd size (hundreds of cattle OR=1.09; 95% confidence interval (CI): 1.05, 1.14). The probability of a positive sample decreased substantially for longer intervals between the initial clinical case and sampling (interval in months OR=0.5; 95% CI: 0.3, 0.6). The presence of diarrhea increased the risk of shedding (OR=2.1; 95% CI: 1.4, 3.0). The effect of recent treatment with antimicrobial agents depended on age group. For heifers and cows, recent antimicrobial treatment increased the probability of isolating Salmonella (heifers OR=11.8; 95% CI: 2.9, 48.8; cows OR=4.1; 95% CI: 2.0, 8.4), but this effect was not statistically significant for calves before weaning. Among animals without recent antimicrobial treatment, preweaned calves were more likely to have positive samples than cows (OR=3.5; 95% CI: 1.8, 6.9; heifers OR=4.7; 95% CI: 2.3, 9.6).