Forecasting the 2006 Elections for the U.S. House of Representatives
| Autor: | Stan Buchanan, Carl E. Klarner |
|---|---|
| Rok vydání: | 2006 |
| Předmět: | |
| Zdroj: | PS: Political Science & Politics. 39:857-861 |
| ISSN: | 1537-5935 1049-0965 |
| DOI: | 10.1017/s1049096506061051 |
| Popis: | o arrive at some understanding ofwhat is going on is hard enough,”said Abraham Kaplan, “without havingalso to meet the demand that we antici-pate what will happen next” ~1964, 351!.Political scientists have been taught todescribe and to explain phenomenarather than to predict them. Kaplan, forone, appeared to think that this wasenough. But within the rich soil of expla-nation, Kaplan admitted ~346!, lay theseed of prediction. Indeed, Carl G.Hempel and Paul Oppenheim ~1948,138!, whom Kaplan took to task for say-ing so, flatly stated that “an explanationis not fully adequate” unless it alsoserved as the basis for prediction ~quotedin Kaplan, 346!.We admit that many challengespresent themselves when we attempt topredict complicated social phenomena.But we argue that there is a danger inlimiting our research to description andexplanation. That danger is that repeatedanalysis of the same datasets will trackpatterns again and again that do not existin the reality outside our samples. Oneguard against this is to take theoriesabout politics and employ them to pre-dict events that have not yet happened.In this spirit, this paper utilizes well-known insights about congressional elec-tions to predict the 2006 elections in theU.S House of Representatives.Methods of predicting congressionalelection outcomes have fallen into twocategories. In the first, experts useddistrict- or state-level information to tryto call elections. Sometimes these predic-tions were aided by district-level polls~of varying quality!. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|