Suchergebnisse

The horse race of election campaigns, built on trial-heat polls of citizens, is subject to errors of methodological differences & sample bias. It is very difficult to really know whether & how the race changes as the campaign moves along. This article examines the full set of available polls conducted during the 2000 US presidential campaign by multiple survey organizations using different methodologies. The results show that it is possible to extract reasonably valid results regarding the candidates as the campaign unfolds, ie, to "watch the horse race 'live' over time." However, there are areas of interest that cannot be reliably assessed even on election day, much less before. What the polls reveal & what they do not must be taken into consideration in interpreting poll data. Tables, Figures, References. J. Stanton

This follow-up to an earlier article, "Testing Visual Signals in Representative Surveys" (Peterson, 2005), which described a new methodological concept of combining survey research & quantitative media content analysis to interpret what is happening in election campaigns, applies the concept to the 2002 German federal election campaign. The concept involves defining the relevant visual elements in media reporting, testing the effect of these elements via representative population surveys, & weighting the elements & including them in quantitative media analyses. Figures are presented depicting what took place during this particular German campaign using this method. The results support the notion that it is possible to measure at least this one facet of the mass media's effect on public opinion. Figures, References. J. Stanton

This paper investigates the reliability & validity of real-time response measurements (RTR). It is based on a comparison of two quasi-experimental studies independently conducted on the second televised debate of the two major candidates for chancellor in the 2002 German federal election campaign. Participants in Bamberg & Mainz -- two mid-sized German cities -- followed the debate on a large-scale screen. The viewers' immediate reactions to the candidates were measured in real-time. In terms of technicalities & substance, both quasi-experiments differed in several important respects. For example, the system used in Mainz was based on a control unit with a 7-point scale & yielded one merged metric dimension for both candidates. The Bamberg system measured positive & negative impressions of the candidates independently, yielding categorical data. Despite these operational differences, the results show that both methods render reliable results that also meet the criteria of face, construct, & criterion validity. Hence, RTR measurements provide valuable, unique insights into subjective immediate reactions to candidates in televised debates & help to explain post-debate perceptions & attitudes. Tables, Figures, References. Adapted from the source document.

This article examines the question of developing the European Neighborhood Policy (ENP) particularly in relation to the eastern neighborhood of the EU, and proposes to make use of the thematic dimension suggested by the Commission to overcome apprehensions about focusing on a particular geographic dimension. It is also proposed that much of what needs to be done to enhance the Union's relations with its eastern neighbors can in fact be accomplished through already existing policy and mechanisms. Adapted from the source document.