In the course of the past decades, a research paradigm has emerged und
er the label ''network analysis'' that is impressive primarily because
of its manifold analytical concepts and methods of analysis of relati
onal - as distinguished from attributive - data. This article aims at
giving an overview of some of the methodological problems and possible
solutions tied to social network analysis. Besides the better known d
escriptive analytical strategies of data reduction we also discuss mor
e recent approaches that allow statistical testing and make it possibl
e to analyse relational and attributive data at the same time. Finally
, some practical concerns are touched on such as, for instance, furthe
r readings and available software.