ebook img

Doing Research in Political Science: An Introduction to Comparative Methods and Statistics PDF

335 Pages·2005·1.37 MB·English
Save to my drive
Quick download

Preview Doing Research in Political Science: An Introduction to Comparative Methods and Statistics

Doing Research in Political Science An Introduction to Comparative Methods and Statistics Paul Pennings, Hans Keman and Jan Kleinnijenhuis Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page i DDooiinngg RReesseeaarrcchh iinn PPoolliittiiccaall SScciieennccee Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page ii Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page iii DDooiinngg RReesseeaarrcchh iinn PPoolliittiiccaall SScciieennccee sseeccoonndd eeddiittiioonn PPaauull PPeennnniinnggss,, HHaannss KKeemmaann aanndd JJaann KKlleeiinnnniijjeennhhuuiiss SAGE Publications London ●Thousand Oaks ● New Delhi Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page iv ©Paul Pennings, Hans Keman and Jan Kleinnijenhuis 2006 First edition published 1999 This edition published 2006 Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, this publication may be reproduced, stored or transmitted in any form, or by any means, only with the prior permission in writing of the publishers, or in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms should be sent to the publishers. SAGE Publications Ltd 1 Oliver’s Yard 55 City Road London EC1Y1SP SAGE Publications Inc. 2455 Teller Road Thousand Oaks, California 91320 SAGE Publications India Pvt Ltd B-42, Panchsheel Enclave Post Box 4109 New Delhi 110 017 British Library Cataloguing in Publication data Acatalogue record for this book is available from the British Library ISBN 1-4129-0376-9 ISBN 1-4129-0377-7 (pbk) Library of Congress Control Number Available Typeset by C&M Digitals (P) Ltd., Chennai, India Printed on paper from sustainable resources Printed and bound in Great Britain by Athenaeum Press, Gateshead Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page v Contents Preface ix PART 1 COMPARATIVE METHODOLOGY 1 1 Comparative Methodology and Statistics in Political Science 3 1.1 Introduction 3 1.2 The comparative approach to political and social science: theory and method 6 1.3 Comparing data: selecting cases and variables 8 1.4 Developing empirical-analytical comparative analysis 13 1.5 How to use this book 15 1.6 Endmatter 16 2 The Comparative Approach: Theory and Method 18 2.1 Introduction 18 2.2 Comparative research and case selection 19 2.3 The use of comparative analysis in political science: relating politics, polity and policy to society 23 2.4 Endmatter 28 3 Meaning and Use of the Comparative Method: Research Design 30 3.1 Introduction 30 3.2 The problem of variables, cases, and interpretations 32 3.2.1 Context matters 35 3.2.2 Logics of comparison 36 3.3 The role of space and time 39 3.3.1 Time and history 40 3.3.2 Space and Cross-Sections 41 3.4 Developing a research design 42 3.5 Transforming concepts into units of measurement 48 3.6 Conclusion 50 3.7 Endmatter 51 Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page vi vvii Contents PART 2 STATISTICS IN POLITICAL SCIENCE 53 4 Concepts, Cases, Data and Measurement 55 4.1 Data and data collection in political science 56 4.1.1 Data obtained from official statistical agencies 56 4.1.2 Verbal and visual accounts, content analysis 58 4.1.3 Questionnaires and surveys 59 4.2 Sampling and the basics of statistical testing 60 4.2.1 Statistical inference from a random sample 60 4.2.2 Random samples and non-random samples 61 4.3 Operationalization and measurement: Linking data with concepts and units 62 4.3.1 Handling missing data 65 4.4 Criteria to evaluate the quality of operationalization and measurements 66 4.4.1 Multiple indicators: the scalability (reliability) problem 69 4.5 Scalability and cluster analysis 70 4.5.1 Likert scales and Cronbach’s alpha 74 4.5.2 Factor analysis 75 4.5.3 Principal axis factoring and confirmative factor analysis 78 4.5.4 Digression: an unknown number of dimensions 80 4.5.5 Explorative cluster analysis 82 4.5.6 Summary 85 4.6 Conclusion 86 4.7 Endmatter 86 5 Explorative and Descriptive Statistics 88 5.1 The univariate distribution of a nominal variable 89 5.1.1 Measures of central tendency for nominal variables: the mode 90 5.1.2 Measures of dispersion for nominal variables: entropy and the Herfindahl Index 91 5.2 The univariate distribution of ordinal, interval and ratio variables 92 5.2.1 Measures of central tendency 93 5.2.2 Measures of dispersion 94 5.2.3 The shape of the entire distribution of a variable with interval measurement 97 5.3 Relationships between variables with nominal measurement levels 99 5.3.1 The chi-square measure of association in a cross-table 100 5.4 The bivariate distribution of two ordinal, interval or ratio variables 103 Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page vii Contents vviiii 5.4.1 Exploring the bivariate distribution the scattergram 104 5.4.2 Bivariate regression analysis 106 5.5 The relation between an interval or ratio variable and a nominal variable 114 5.5.1 An interval variable and a bivariate nominal variable: the comparison of two means 114 5.5.2 Analysis of variance: an interval variable by a nominal variable with j values 115 5.6 Populations, samples and inferential statistics 119 5.6.1 The urn model 120 5.6.2 Unbiasedness, efficiency and robustness of an estimator 121 5.6.3 The general procedure used in hypothesis testing 123 5.6.4 Four common probability distributions of test statistics 124 5.6.5 Degrees of freedom 127 5.6.6 Sense and nonsense of statistical tests 128 5.7 Summary 128 5.8 Endmatter 129 6 Multivariate Analysis and Causal Inference 132 6.1 Causality and multivariate relations 133 6.2 Overview of multivariate data analysis techniques 136 6.3 The case-oriented approach 137 6.4 Nominal dependent and independent variables 141 6.4.1 Cross-table elaboration 142 6.5 Nominal dependent variable, interval independent variables 145 6.5.1 Discriminant analysis example: explaining the type of government 146 6.6 Interval dependent variable, nominal independent variables: analysis of variance 150 6.7 Interval dependent and independent variables: regression analysis 152 6.7.1 Themultiple regression model 153 6.7.2 Assumptions of the ordinary least squares estimation method 157 6.7.3 Direct causes, intervening variables and antecedent variables 164 6.7.4 Interactions in the multivariate regression model 164 6.7.5 Time series analysis: the autocorrelation problem 166 6.7.6 Pooled time series analysis: autocorrelation and heteroscedasticity 174 6.7.7 Reciprocal causal relations: linear structural equation models 180 6.8 Epilogue 180 6.9 Endmatter 181 Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page viii vviiiiii Contents PART 3 DOING POLITICALRESEARCH 183 Introduction to Part III: Doing political research 183 7 How problems arise 187 7.1 Processes of electoral change 187 7.1.1 The problem of change 187 7.1.2 Measuring electoral change 189 7.1.3 Modelling change 193 7.2 Processes of party change 198 7.2.1 The role of parties 198 7.2.2 Parties and ideology scales 200 7.2.3 Parties and issues 206 7.2.4 Public opinion and party responsiveness 209 7.3 Conclusions 213 7.4 Endmatter 214 8 How Decisions are Made 217 8.1 Introduction 217 8.2 Types of democracies 219 8.3 Party systems 225 8.4 Cabinet formation and duration 229 8.5 Interest intermediation 239 8.6 Federalism, centralism and institutional autonomy 243 8.7 Presidentialism 248 8.8 Conclusions 252 8.9 Endmatter 252 9 How Problems are Solved 258 9.1 Introduction 258 9.2 Welfare-related outputs and performance 260 9.3 Actors and socio-economic problem-solving 263 9.4 Institutions and socio-economic problem-solving 270 9.5 Electoral cycles and macro-economic policy 276 9.6 Democratic performance 279 9.7 Parties and accountability 289 9.8 Conclusions 293 9.9 Endmatter 295 Appendix 299 Bibliography 305 Index 317 Pennings (Research)- Prelims.qxd 10/13/2005 11:41 AM Page ix Preface This second edition of Doing Research in Political Scienceis a thoroughly revised and updated version of the book that was originally published in 1999. In revising this edition we have benefited from several constructive and positive reviews and per- sonal communications. One comment in particular made us reconsider the target readership for which this textbook is intended. Apparently – so some of the critics maintained – the level of information makes the book especially suitable for advanced students (e.g. in the final year of BAtraining, during MAstudies and in the prelimi- nary stages of a PhD). With this caveat in mind we have rewritten parts of the book and attempted to improve the presentation. The book maintains its original structure consisting of three parts representing in our view the basic stages of any theory-driven empirical-analytical research in the social and, in particular, the political sciences. In each chapter there is an introduction to its contents, and at the end there is a list of the main topics covered, which may help both teacher and student to find the information she or he needs. In addition, each chapter contains examples that are taken from existing comparative research and are partially based on data made accessible by us via the World Wide Web (http:// research.fsw.vu.nl/DoingResearch). In Part 1 we present our own arguments concerning the comparative approach in the social sciences: namely, that any empirical research ought to be theory-driven and must be formulated in a well-elaborated research design. Part 2 is essential reading for those who wish to understand the use of (advanced) statistics in order to be able to conduct an explanatory analysis (including its caveats and pitfalls!). Part 3 can be seen as an attempt to pull together the threads of our way of doing comparative research and will be of interest to any reader, whether a freshman or an advanced student of comparative politics and social sciences at large. Without claiming that this approach is the one and only way to teach comparative methods and statistics in political science, we are certain that it offers a valuable ‘springboard’ to judging the comparative information with which most, if not all, students are confronted. It will help the student to shape a theory-inspired research design in such a way that it leads to plausible and adequate results. These are valu- able skills that are lacking in too many textbooks that focus on methodology. During the process of writing this book, we have benefited from contributions many institutions, scholars and students, to whom we wish to express our thanks. First of all, the Essex Summer School in Social Science Data Analysis and Collection gave us the chance to test the draft version of the book on an international group of graduate students. We thank a number of colleagues for their detailed and helpful corrections to the manuscript. Linde Wolters, an assistant in our department, care- fully organized the references and bibliography. Sabine Luursema has been helpful

See more

Similar Doing Research in Political Science: An Introduction to Comparative Methods and Statistics