The relation of educational level to obstructive lung disease, spirometric airflow limitation, and respiratory symptoms was examined in a two-phase cross-sectional study of a Norwegian general population aged 18-73 years in 1985-1988. The first phase was a questionnaire survey. In the second phase, a stratified sample of those who responded in the first phase was invited to a clinical and respiratory physiologic examination. Altogether, 714 subjects attended, representing 84% of those invited. The prevalences of obstructive lung disease and spirometric airflow limitation were 7.8% and 4.5%, respectively. A total of 18% of the population had completed college, a further 60% had completed secondary school, and 21% had obtained a primary school education alone. The prevalence of both smoking and occupational airborne exposure decreased with increasing educational level. The sex-, age-, smoking-, and occupational exposure-adjusted odds ratio of obstructive lung disease in primary-versus university-educated subjects was 2.9 (95% confidence interval (CI) 1.3-6.5); in secondary- versus university-educated subjects it was 1.4 (95% CI 0.7-2.8). The corresponding values for spirometric airflow limitations were 5.2 (95% CI 2.0-13.4) and 1.8 (95% CI 1.2-2.7). All of the respiratory symptoms except breathlessness grade 2 were significantly associated with educational level after allowing for sex, age, smoking, and occupational airborne exposure. The survey indicates that educational level is a risk factor for airway disorders independent of smoking and occupational airborne exposure.