While multi-method research is currently provoking much interest, there is little reflection on the legitimacy of this kind of research and on the conditions for achieving high quality research. This article first describes the scientific and socio-political contexts from which this movement towards multi-method research has emerged. It then goes on to discuss why comparative analysis is central to the triangulation of methods and why the notion of triangulation itself requires an external point of reference. It is suggested that the reason why we produce only half-hearted or even illigitimate comparisons is because the reference points are hidden. For multi-method research to be of high quality it must clearly externalize valid inferences at each moment in the spiral of knowledge, identify its analytical logic and establish its internal reference points. (This article is the result of joint research and the respective positions of the authors are reflected in the way in which the article is presented.)
