The present study was carried out among a sample of 667 Dutch nurses and assesses three aspects of the construct validity of the two most widely used self‐report burnout questionnaires: The Maslach Burnout Inventory (MBI) and the Burnout Measure (BM). Although the factorial validity of the three‐dimensional structure of the MBI was convincingly demonstrated by confirmatory factor analysis with LISREL, four weak and ambiguous items were identified. Regarding the BM, some doubts have arisen about its one‐dimensionality. The congruent validity of the questionnaires was well established: They both refer to the core element of the burnout syndrome (i.e. exhaustion). Moreover, linear structural analyses suggested that burnout is a two multi‐dimensional construct consisting of an affective component (i.e. exhaustion) and an attitudinal component (i.e. a negative attitude towards recipients and towards one's job performance). However, the discriminant validity of the first component is rather poor since it considerably overlaps with self‐reported somatic complaints and psychological strain. It is concluded that the MBI can be employed as a reliable and valid multi‐dimensional indicator of burnout in professionals who work with people. The BM assesses the non‐specific affective component of burnout (i.e. exhaustion) and should therefore be supplemented by a scale that measures the attitudinal component of the syndrome.