Suchergebnisse

THIS ARTICLE DISCUSSES LONG-TERM CHANGE IN THE POLITICAL PARTY SYSTEM. THE AUTHORS PROPOSE A THEORY OF ISSUE EVOLUTION THAT EXPLAINS NORMAL PARTISAN CHANGE, AND ILLUSTRATE THE THEORY BY EXAMINING THE IMPACT OF RACIAL ISSUES ON PARTY IDENTIFICATION. NEW ISSUES ARE SEEN AS A STIMULUS OF PARTISAN CHANGE.

Cover -- Half Title -- Title -- Copyright -- Dedication -- Contents -- List of Tables and Figures -- Preface to the Second Edition -- Preface to the First Edition: An Unfinished Essay? -- Acknowledgments to the First Edition -- 1 PUBLIC OPINION? -- On Politics at the Margin -- On Public Opinion Leadership -- On the Plan of the Essay -- Notes -- 2 THE CONCEPT OF POLICY MOOD -- The Dynamics of Public Policy Preference: A Model of Public Opinion -- Issue Dynamics: Trends, Cycles, Stationarity -- Notes -- 3 DEVELOPING A MEASURE OF MOOD -- The Survey Marginals Data Base -- The Conundrum -- A Regression Model for Measuring Mood -- An Algorithm for Estimating Mood -- Notes -- 4 THE COMPONENTS OF MOOD -- Estimating Policy Mood: A Search for Political Eras -- Political Eras? -- The Scale of Attitude Change -- Mood and Self-Identification -- The Policy Components of Mood -- Notes -- 5 AN ELECTORAL CONNECTION? -- Public Opinion, Elections, Outcomes -- Policy Mood and Partisanship -- Presidential Elections -- Congressional Elections -- Mood and Mandates -- Notes -- 6 REFLECTIONS ON THE PRESENT AND FUTURE OF AMERICAN POLITICS -- A Model of Opinion Response -- On Revolutions, Contracts, and Normal Politics: Public Opinion in the 1990s -- The Future -- Notes -- Appendix 1: An Algorithm for Estimating Mood -- Appendix 2: Selected Data Series -- Appendix 3: Selected Domestic Policy Survey Questions -- References -- Index

Regressions on data jointly structured in space & time, commonly referred to as the pooling of cross sections of time series, can be formidable both in the strength of their design properties & in the number of special statistical problems encountered with them. The potential applications of pooled design are explored, as well as the special statistical problems commonly associated with such analysis. Four estimators -- ordinary least squares, least squares with dummy variables, error components, & an adaptation of the G. E. P. Box-G. M. Jenkins ARMA models (Time Series Analysis: Forecasting and Control, San Francisco: Holden-Day, 1976) to the pooled estimation problem -- are reviewed, with an effort to suggest where each may find application in political science research. The four estimators are illustrated by analysis of regional dynamics in party issue polarization over issues of racial desegregation in the US House of Representatives, 1957-1980. 7 Tables, 33 References. HA