The purpose was to investigate the relationship between noise exposure and tinnitus among workers with normal hearing and hearing loss, respectively. We conducted a cross-sectional survey of 752 workers employed at 91 workplaces, that were investigated by means of full work-shift noise levels, questionnaire data, and bilateral pure-tone audiometry. Tinnitus was not associated with the present noise level, the duration of noise exposure, or the cumulative noise exposure if participants had normal hearing. As expected, such trends were demonstrated if participants had a hearing handicap. Based on these data, we will be cautious in ascribing tinnitus to noise exposure in our patients' workplaces if they have a normal audiogram. Furthermore our data indicates no risk of noise-induced tinnitus at exposure levels where no hearing loss would be expected, e.g. as usually encountered in non-industrial workplaces.