We conducted a prevalence case-control study to investigate the relation between family composition, infection, and development of asthma at age 7-9 years. Potential cases (399) and controls (398) were selected from the Wellington, NZ, arm of the International Study of Asthma and Allergies in Childhood, a population-based prevalence study. Further screening questions restricted cases to children with a diagnosis of asthma and current medication use (N = 233) and restricted controls to children without a history of wheezing and no diagnosis of asthma (N = 241). After controlling for confounders (including infections, atopy, and socioeconomic status), family size was strongly related to asthma. Having no siblings [prevalence odds ratio (POR) = 2.51; 95% confidence interval (CI) = 1.05-6.01] or one sibling (POR = 1.86; 95% CI = 1.14-3.03) was associated with an increased risk of asthma compared with having more than one sibling. Parent-reported rubeola infection (and possibly other similar viral exanthems) was independently associated with a decreased risk of asthma (POR = 0.48; 95% CI = 0.27-0.83), but reported pertussis infection (POR = 1.57; 95% CI = 0.58-4.24) and day care attendance in the first year of life (POR = 1.81; 95% CI = 0.93-3.51) were not strongly associated with increased risks of asthma.