Research project Social integration of the elderly in Slovenia deals with studying factors of changes in time affecting the structure and composition of social support networks in a longer period of time. Part of this research also consists of secondary data analysis of two studies, in which social networks were also measured - in 1987 (before the transition) and in 2002 (after the transition). In this paper mainly methodological problems encountered during secondary data analysis are presented. Firstly, characteristics of the 1987 and 2002 studies are presented, followed by evaluation of factors that could threaten comparability of chosen indicators and the comparability of both samples. Analyses have shown that methodological differences do not have a large effect on substantive interpretations. Possible methodological biases may be softened by stricter statistical criteria and more tentative interpretations in substantive analyses
|