English  |  正體中文  |  简体中文  |  Items with full text/Total items : 51921/87065 (60%)
Visitors : 8473703      Online Users : 85
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/101465


    Title: 影響美國總統表現滿意度指標之關鍵變數 : 運用緃橫迴歸方程式分析
    Other Titles: The determinants of US presidential job approval : a panel data regression approach
    Authors: 劉文傑;Liu, Wen-Chieh
    Contributors: 淡江大學美洲研究所博士班
    柯大衛;Kleykamp, David
    Keywords: 總統滿意度;緃橫迴歸方程式;虛擬變數;社會心理學;民意調查;Presidential Job Approval;Presidential Approval Rating;Panel Data Regression;Dummy Variable;Social psychology;Survey Method
    Date: 2014
    Issue Date: 2015-05-01 13:31:26 (UTC+8)
    Abstract: 總統滿意度是評估工作表現的指標,它是民主政治中獨有的特色,不論在選舉期間或平常任期,已成為反應美國總統政治實力的最佳指標,特別是在選舉期間,若能充份運用影響滿意度的關鍵變數,將有利掌握政治權力的根源。這個領域自1970年代由慕勒教授創始後快速發展,研究主軸亦不斷演進。
    本論文研究資料始自1948年1月至2012年10月,共約64年,前後包括杜魯門至歐巴馬等12位總統。各總統任期內每月滿意度數據經相連結後,即成為迴歸方程式之因變數,其他經理論探討後可能具有影響滿意度之各類變數,則視為方程式自變數。這些自變數可區分為4大類,第1類是處理時間序列特性的周期變數與遲滯變數;第2類是各類經濟變數;第3類是消費者信心指數;最後1類則是為掌握個別獨特效應而設計之虛擬變數。藉由建立客觀一致的篩選標準,將發生在總統任內的重要事件,彙整成為重要事件的資料庫,並依屬性歸入政治、軍事、社會等3大類,據以編成虛擬變數,帶入迴歸方程式分析,如此一來即可考量原無法以計量觀點分析的因素,有效擴大分析層面。
    本研究以每位總統在位期間的月份取代實際的日期,並視為各總統彼此間關連之因素, 12位總統均視為具有2任96個月的任期,如此始得以運用緃橫迴歸方程式概念,做為計量分析各類自變數與因變數間的技巧。納入方程式內分析的自變數,都有具體的實證結論;另本研究其他的特色,包括將社會心理學領域中探討受訪者心智作用過程的相關理論,充份運用於篩選自變數及解讀實證結果的重要考量;另亦深入瞭解蓋洛普民意調查公司執行民意調查的程序,俾利驗証足可獲得可信度與準確度的民意調查理論。
    本論文共分為六章。第1章緒論闡述研究背景、動機、目的、研究問題及限制;第二章文獻探討完整收錄研究領域之最新發展趨勢,並將具體研究主題分類描述;第三章收錄社會心理學之研究主軸及民意調查程序的相關理論;第四章彙整本研究所使用之各類相關理論;第五章彙整經過統計分析之實證結果,並對實證結果加以解讀。第六章則比較本研究之成果與其他研究結論相同與相異之處,並提出未來可能的研究方向。
    Presidential job approval(presidential approval rating), a survey indicator, falls into the category of performance evaluation for incumbent presidents. This indicator is considered as one of the signature features for democratic politics in modern times. In fact, it has been widely used during both elections and ordinary times as an indicator of actual presidential popularity in America. It is well recognized that the presidential candidate who can make the best use of the determinants of presidential approval ratings during the elections will have the edge to grasp the source of political power. The studies in this field, founded by Professor John Muller in the 1970s, have rapidly developed ever since. The main themes of this research are continuously evolving with the time.
    The timeline for these studies originated on January 1 1948 and ended in October 2012, making a total study length of 64 years. If we put the said timeline into context, the studies covered 12 presidents in all, starting with President Truman and ending with President Obama. The presidential approval ratings of each of the 12 presidents were concatenated into a single time series, and treated as the dependent variable of panel data regression. As far as the independent variables are concerned, all the potential factors are divided into four different categories, including the factors dealing with the features of time series, a wide range of economic variables, consumer confidence indexes, and various dummy variables. The way we create dummy variables is to select critical events that attracted the US public''s attention in the 12 presidents, and separate them into three categories. These categories are political affairs, military events, and social affairs based on nature of event. By doing this, the study successfully broadens the perspectives by treating non-quantitative factors as independent variables of panel data regression. Moreover, we create a chronological database for important events.
    In this dissertation, we replace the actual date with the month in office of each of the 12 presidents, and restructure one-dimensional time series into a three-dimensional format. Furthermore, we investigate the theories of Social Psychology to consider the mind workings of each individual who takes the survey. Meanwhile, we consider issues from Gallup''s perspective, and seek confirmation between theory and actual procedures.
    This dissertation is divided into six chapters. Chapter one introduces the study background, research motivation, research questions, and limitations. Chapter two provides the results of a comprehensive literature review. Chapter three includes the main theories in the fields of social psychology and survey method. Chapter four consists of a wide range of statistics theories that were applied in this study. Chapter five is composed of the results of the statistics empirical results, and the last chapter includes conclusions made by comparing the achievements of this study to other research results and potential topics for future work.
    Appears in Collections:[美洲研究所] 學位論文

    Files in This Item:

    File SizeFormat
    index.html0KbHTML101View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback