Lazarsfeldův přínos k metodologii panelové analýzy

This paper explores causal explanations that use panel data and describes the contribution of Paul Lazarsfeld to the methodology of panel analysis. The introductory part describes the concepts of ‘panel data’ and ‘panel analysis.’ The second section is devoted to the history of panel studies. The main part of the paper focuses on the contributions of Paul Lazarsfeld to panel data analysis. The term ‘panel study’ generally denotes any data collection that involves the same respondents who are questioned repeatedly in consecutive waves of a survey. In contrast, ‘panel analysis’ refers to the quantitative analysis of changes in the distributions of responses among the same respondents across two waves of a panel data set. Paul Lazarsfeld developed panel analysis during the late 1930s and early 1940s. The main aim of this early work was to test for causal relationships, and to outline some explanation for the intra-personal changes observed. Lazarsfeld outlined three important panel data analysis procedures: 1) analysis of turnover tables, 2) analysis of qualified change also known as the “analysis of qualifiers”, and 3) analysis of concurrent changes. The latter was often referred to by Lazarsfeld in his methodological papers as the problem of the “sixteen-fold table”. The final section of this paper discusses of the use of control groups in panel studies and problems associated with panel attrition rates.