Using data from the first three waves of the Danish Longitudinal Survey of Children (DLSC), this working paper develops measures of child well-being with the aim of allowing comparisons in time. These measures are inspired by the SDQ-index (from the Strengths and Difficulties Questionnaire), a relatively easy-to-use index enabling a grouping of children into three groups – normal, borderline, and abnormal – according to levels of psycho-social problems. The SDQ-questions were part of the third wave of the DLSC-survey in 2003 and will continue to be so in the future. The first two waves of the DLSC, in 1996 and 1999, used different questions to describe the children's well-being. Consequently, the issue in this working paper is whether it is possible, on the basis of the questions in the early questionnaires, to construct measures of child well-being, similar to SDQ, allowing analysis of developments in well-being in time. We construct different potential measures and evaluate their usefulness as indicators of child well-being. In this process, we also discuss different questions in relation to the measurement of child well-being, e.g. the level of problems deemed to be normal or abnormal. The use of the term “well-being” is convenient because it is a familiar concept to most people but because the definition of well-being is very individual it is also a vague concept. Here, well-being is defined as absence of particular problems, and although it is plausible that children with many of these problems have low well-being, there is no way to be sure that children without these specific problems have high levels of well-being. Section 2 of the working paper describes some of the measures of child well-being used in the literature, particularly the SDQ-index. Section 3 describes the DLSC and proposes different measures of child well-being based on factor-analysis of selected questions in the 1996 and 1999 DLSC-questionnaires. Section 4 roughly describes developments in time and section 5 discusses the measures.