A cross-sectional survey of tuberculin skin reactivity was conducted in Al-Ain, United Arab Emirates (U.A.E.) between January and June 1994, to find out the prevalence rate of tuberculosis infection. A pre-designed questionnaire was used to collect details of BCG scar, age, sex, residence area, nationality, education, type of house, number of rooms, family size and household contact history of tuberculosis. A total of 785 students were screened, of whom 547 gave a history of BCG vaccination in the past and 238 were BCG-negative. Among BCG-negative children aged 5-11 years and 12-15 years, only 6.5% and 9.3%, respectively, had a positive Mantoux reaction--a rate lower than most Third World countries, but higher than developed countries where under 2% of children are tuberculin reactors. A general linear model with positive Mantoux reaction as the dependent variable was fitted to the data to examine the joint effect of age, sex, residential area, number of rooms at home, family size and BCG vaccine history. The Mantoux reaction was entered as positive and a number of statistically significant associations were found between positive Mantoux test > 10 mm and: age (P = 0.0018); sex (P = 0.0281); residential area (P < 0.0001); number of rooms (P = 0.0017); and BCG vaccine history (P < 0.0001). However, family size did not have any statistical effect on tuberculin testing (Mantoux test > 10 mm). The prevalence (8%) in the 5-14 years age group puts U.A.E. between low (2%) and middle (14%) prevalence countries, according to the classification of the International Union Against Tuberculosis. This calls for continuation of free treatment of active cases and increased efforts towards screening of contacts.