本文主旨為探討官方數據(治安及交通)與第一線的110報案數據(治安及交通)所含資訊，何者較貼近民眾感受，希望民眾對政府的政策與決策「有感」，也提供公務機關及學者於未來數據及資訊蒐集運用之參考。經蒐集104至106年間桃園市官方公布的刑事案件發生數、交通事故發生數、警察局勤務指揮中心110刑事案件報案數、交通案件報案數、內政部實價登錄之房屋交易數據資料及人口數，進行實證分析。運用經濟學供需理論模型進行初步比較分析後，以線性迴歸篩選出不同數據中所包含的資訊，隨時間趨勢進行比較。究竟是否第一線未公布數據中所含的資訊，較官方公布數據所含的資訊更貼近民眾，更能反映出人民感受，亦或因第一線數據所含資訊有太多雜訊，致官方未公布數據？本文就客觀事實與實務現況分析，發現當第一線數據是完整的，確實更貼近民眾，更能反映出人民感受，惟若其並非完整，因不具代表性，官方統計數據所含資訊影響反而較為顯著；此外，研究發現民眾對近期生活環境治安好壞的重視程度比交通順暢更高；若長期觀察，交通順暢相較於生活環境治安的好壞影響更為長久。另外，改善治安對房價的效益較交通大，但一般認為改善治安成本較高，所以若房價真的反映人民對居住品質的真實感受，則需進一步的成本/效益分析始能擇定較具效率的施政方向。

The purpose of this paper is to study whether the information contained in the official data (security and traffic) or the first-line 110 report data (security and traffic) is closer to the public's feelings, hoping that there are true feelings in people on the government's policies and decisions, and also providing future application for institutions, scholars, and government official services for reference. The empirical research was conducted by collecting the number of criminal cases, traffic accidents, criminal cases reported by the police bureau's service command center, the number of traffic cases reported, the housing transaction registered on the Ministry of the Interior at the actual selling price data and the population number between 2015 and 2017. After using the economic supply and demand theory model for preliminary comparative analysis, the information contained in different data is screened by linear categorization, and then compared with time trends. Whether the information contained in the unpublished data on the first line is closer to the public and more likely to reflect the feelings of the people than the information contained in the officially published data? Or because there is too much noise in the information contained in the first line of data, the official data was not released? This paper analyzes the objective facts and actual conditions and finds that when the first-line data is complete, it is closer to the people and has better reflection on the people's feelings. However, if it is not complete, because it is not representative, the information contained in official statistics will affect it significantly instead. In addition, the study found that the public pays more attention to the safety of the recent living than the traffic; if observed for a long time, the traffic has longer affection than the quality of the living. Moreover, the improvement of public security on housing prices is more efficient than transportation, but it is generally believed that the cost of improving public security is high, so if housing prices really reflect the people's true feelings about the quality of living, further cost/benefit analysis is needed for the government to governance more efficiently.

一、 中文文獻
〔1〕 Jeffrey M. Wooldridge著，胥愛琦譯(2017)，「計量經濟學(6版)」，華泰文化。
〔2〕 N. Gregory Mankiw著，王明正譯(2015)，「經濟學原理」，新加坡商聖智學習。
〔3〕 林惠玲 陳正倉 合著(2014)，「基礎統計學-觀念與應用」，雙葉書廊。
〔4〕 丹˙賈德納、菲利普˙泰特洛克著，蔡裴驊譯(2016)，「超級預測：洞悉思考的藝術與科學，在不確定的世界預見未來優勢」，寶鼎出版社。
〔5〕 馬克˙湯普森著，王審言譯(2018)，「為什麼我們要懂公眾語言？」，臉譜出版社。
〔6〕 陳正倉 林惠玲 林建甫 林世昌 合著(2018)，「經濟學理論與實際」，雙葉書廊。
〔7〕 陳明傳著(2000)，「警察行政專題(二)」，中央警察大學。
〔8〕 張清溪 許嘉棟 劉鶯釧 吳聰敏 合著(2016)，「經濟學(5版)」，雙葉書廊。
〔9〕 劉彩卿 陳欽賢 合著(2012)，「STATA基礎操作與統計模型應用」，雙葉書廊。
 
二、 英文文獻
〔1〕 Granger, Clive W. J. (1969), "Investigating Causal Relations by Econometric Models and Cross-spectral Methods." Econometrica, 37, p.424–438.
〔2〕 Jan K. Brueckner (2011) Lectures on Urban Economics, p.296.
〔3〕 Pearl, Judea (2009), Causality: Models, Reasoning and Inference, Cambridge University Press.