Suchergebnisse

Questionner les programmes actuels, réfléchir à ce que pourraient être des objectifs de sciences sociales, cet effort signifie aussi un changement des pratiques pédagogiques. Avoir des idées de transformation, c'est se condamner à les appliquer dès maintenant, tant bien que mal - c'est aussi en retour enrichir la démarche générale par des interrogations nées d'une expérience ponctuelle, voici un exemple d'incarnation de cette réflexion théorique sur le terrain, avec des élèves concrets dans un collège banal de Seine-Saint-Denis.

Cover Page -- Title Page -- Copyright -- Preface -- I Simple Recursive Models and Path Analysis -- 1 Spurious Correlation: A Causal Interpretation -- 2 Four-Variable Causal Models and Partial Correlations -- 3 Path Coefficients and Path Regressions: Alternative or Complementary Concepts? -- 4 Path Analysis: Sociological Examples -- 5 Causal Inferences, Closed Populations, and Measures of Association -- II Simultaneous-Equation Techniques -- 6 Identification Problems in Economic Model Construction -- 7 Recursive versus Nonrecursive Systems: An Attempt at Synthesis -- 8 The Application of a System of Simultaneous Equations to an Innovation Diffusion Model -- 9 The Choice of Instrumental Variables in the Estimation of Economy-Wide Econometric Models -- 10 Two Theorems on Ceteris Paribus in the Analysis of Dynamic Systems -- 11 Sensitivity Analysis of Arbitrarily Identified Simultaneous-Equation Models -- III The Causal Approach to Measurement Error and Aggregation -- 12 Theory, Deduction, and Rules of Correspondence -- 13 Estimating the Parameters of Path Models Involving Unmeasured Variables -- 14 Multiple Indicator Models Using Lisrel -- 15 Respecification in Multiple Indicator Models -- 16 Multiple Indicators and Complex Causal Models -- 17 Problems of Aggregation -- Author Index -- Subject Index

This paper studies the relationship between incumbents' performance and political polarization, both with theory and data. The theory is based on a spatial model of political competition in which the voters use the incumbent's performance in office to update their beliefs about his competence. A better performance leads to the incumbent's electoral advantage and so allows him to announce a more extreme platform closer to his bliss point. Therefore, conditioned on reelection, a better incumbent's performance leads to higher political polarization. We use the data on the incumbents' performance in natural disaster relief and the ideological positions they take in the U.S. House of representatives in 1953-2010. The empirical evidence shows that a better performance in post-disaster recovery is associated with more extreme ideological positions of the corresponding incumbents. These and other empirical results are in line with the model predictions.

WHAT IS RESEARCH IN THE SOCIAL SCIENCES FOR? SHOULD (much of) it be financed by the state? How should it be organized? Where should it take place? In the United Kingdom, these questions thud down each morning on the desk of the Chairman of the Economic and Social Research Council, which now has a new home (in Swindon, with its big sister Research Councils) and a new Chairman, Professor Howard Newby. Six years, and more than one chairman, have come and gone since I sat at that desk; I have a short memory, and a full recognition of the duty of a retired bureaucrat — 'get out, and shut up'. I am not willing or able to bore readers with a description or a critique of recent policies. But my present function gives me a new standpoint; things look different from Strasbourg, and an international perspective helps; above all, my responsibilities now spread widely (although jolly thinly) over the whole range of learning, scholarship, and science. So these are home thoughts from abroad.

'Social-science research has been transformed over the last generation by the advent and expansion of the general social surveys (GSS). The GSS model of research has created a infrastructure for the social sciences designed to address the interests and research agenda of scholars and their students; cover a wide range of topics; utilize reliable, valid, and generalizable measurement; and provide data both across nations and across time. This design in turn has generated widespread analysis and notably contributed to our understanding of social processes and societal change.' (author's abstract)|