This article investigates the effect of survey modes on the reliability of answers. Web questionnaires are increasingly used to complement traditional data collection. The utilization of web data, however, raises concerns about the reliability of answers when compared with other survey modes. This analysis compares web and telephone surveys and shows that reliability again, of what depends on the specific content of the questions. Based on the Luxembourgish Labour Force Survey 2015- 2017, we find that answers to objective questions, such as employment status, are not affected by the survey mode while answers to subjective questions, such as job satisfaction, are.