A cross-sectional study was conducted to evaluate the association between herd-level characteristics, management routines and farm-level antibody status of bovine coronavirus (BCV) and bovine respiratory syncytial virus (BRSV) in 257 Swedish dairy herds. The possible spatial clustering of positive herds compared with negative herds was also investigated. For each herd, a pooled milk sample from five primiparous cows was analysed for the presence of antibodies to BCV and BRSV. Herd-level information was obtained by a questionnaire. Logistic regression was used to evaluate the association between predictors and antibody status to BCV and BRSV. Large herd size, being located in southern Sweden, and not providing boots for visitors were found to be associated with being antibody-positive to BCV and BRSV. A short distance to the nearest cattle herd was an additional risk factor for BCV. One of the studied areas was suitable for spatial analysis. Positive herds were not spatially autocorrelated when compared with negative herds as estimated by the K-function regarding both BCV and BRSV. This indicates that local factors such as daily visiting milk trucks and wild animals were unlikely to be important sources of infection in this area. Moran' I statistics and semivariogram showed no evidence of spatial autocorrelation in the residuals, indicating that remaining unidentified factors are not spatially dependent in the areas under study.