Making Social Sciences More Scientific: The Need for Predictive Models

Hardcover | September 14, 2008

byRein Taagepera

not yet rated|write a review
In his challenging new book Rein Taagepera argues that society needs more from social sciences than they have delivered. One reason for falling short is that social sciences have depended excessively on regression and other statistical approaches, neglecting logical model building. Science isnot only about the empirical 'What is?' but also very much about the conceptual 'How should it be on logical grounds?' Statistical approaches are essentially descriptive, while quantitatively formulated logical models are predictive in an explanatory way. Why Social Sciences Are Not ScientificEnough contrasts the predominance of statistics in today's social sciences and predominance of quantitatively predictive logical models in physics. It shows how to construct predictive models and gives social science examples. Why Social Sciences Are Not Scientific Enough is useful to students whowish to learn the basics of the scientific method and to all those researchers who look for ways to do better social science.

Pricing and Purchase Info

$152.36 online
$201.00 list price (save 24%)
Ships within 1-3 weeks
Ships free on orders over $25

From the Publisher

In his challenging new book Rein Taagepera argues that society needs more from social sciences than they have delivered. One reason for falling short is that social sciences have depended excessively on regression and other statistical approaches, neglecting logical model building. Science isnot only about the empirical 'What is?' but ...

Rein Taagepera has B.A.Sc. in engineering physics plus M.A. in physics from University of Toronto and Ph.D. in solid state physics plus M.A. in international relations from University of Delaware. After 6 years of industrial research at DuPont Co., he has taught political science at University of California, Irvine since 1970 and also...

other books by Rein Taagepera

Format:HardcoverDimensions:232 pages, 9.21 × 6.14 × 0.79 inPublished:September 14, 2008Publisher:Oxford University PressLanguage:English

The following ISBNs are associated with this title:

ISBN - 10:0199534667

ISBN - 13:9780199534661

Look for similar items by category:

Customer Reviews of Making Social Sciences More Scientific: The Need for Predictive Models

Reviews

Extra Content

Table of Contents

PrefaceR Duncan Luce: ForewordPart I. The Limitations of Descriptive Methodology1. Why Social Sciences Are Not Scientific Enough2. Can Social Science Approaches Find the Law of Gravitation?3. How to Construct Predictive Models: Simplicity and Non-Absurdity4. Example of Model Building: Electoral Volatility5. Physicists Multiply, Social Scientists Add--Even when It Doesn't Add up6. All Hypotheses Are Not Created Equal7. Why Most Numbers Published in Social Sciences Are Dead on ArrivalPart II. Quantitatively Predictive Logical Models8. Forbidden Areas and Anchor Points9. Geometric Means and Lognormal Distributions10. Example of Interlocking Models: Party Sizes and Cabinet Duration11. Beyond Constraint-Based Models: Communication Channels and Growth Rates12. Why We Should Shift to Symmetric Regression13. All Indices Are Not Created EqualPart III. Synthesis of Predictive and Descriptive Approaches14. From Descriptive to Predictive Approaches15. Recommendations for Better Regression16. Converting from Descriptive Analysis to Predictive Models17. Are Electoral Studies a Rosetta Stone for Parts of Social Sciences?18. Beyond Regression: The Need for Predictive ModelsReferencesIndex