第六十五期人口學刊2022.12 出刊


本期目錄
Contents


本期共刊登三篇研究論文,前兩篇為空間人口學(spatial demography)系列的論文,以空間統計的概念應用於人口議題,包括 生育、罹病和死亡、遷移等數值的空間和時空之特性。這類型方法根據相對位置建立資料間的關係,藉由鄰區決定分析時的權重,將鄰 近資料通常較為相似的特性加入考量,以統計方法從看似雜亂無章的資料中找出規律。當資料加上時間、空間等座標紀錄後,其相對關 係更為多元複雜,而空間統計的分析大多非常注重視覺化的呈現,以圖像之類的工具顯示資料的重要特性。這個步驟屬於探索性資料分析 (exploratory data analysis)的一部分,正是大數據分析的關鍵步驟,這類型分析除了熟悉統計等方法外,也需對資料所屬專業領域有一定 的認知,才能確定圖像化的過程是否有過度解讀之虞。本期前兩篇論文均與罹病有關,讀者可從中觀察資料圖像化的過程,瞭解作者們的 切入點與思維重點,以及實務應用時如何在問題需求及分析方法間取得協調。

本期刊登的第一篇文章由陳威全撰寫,屬於空間點分布型態的議題研究,可以應用於探討疾病是否存在群聚現象,並衍生至判斷疾 病的擴散方向及趨勢。空間資料主要分成三種類型:空間點分布型態(spatial point patterns)、空間格點資料(lattice data)、地理統計資 料(geostatistical data),空間點分布的資料多半來自於自然科學與生物科學,像是地震發生地點、交通肇事地點等,目標在於分析其形成 原因和可能產生的影響,第一篇文章即是以高雄市最近20年的登革熱病例為目標,探究是否存在高罹病率的高風險地區。藉由疾病發生點 分布資料來決定高風險地區,尺度是非常重要的判斷關鍵,用於評估點分布為隨機分布或是具有某種規則。本文作者根據兩種經典概念, 基於資料驅動原則提出臨界尺度的判斷方式,研究發現新方法可用於有效限縮臨界尺度的可能範圍。

第二篇文章則由陳怡如與楊澤全撰寫,以地理加權分量迴歸(geographically weighted quantile regression, GWQR)為題,可用於 探討與各地區死亡率有關的因素。GWQR同時考量統計和空間異質性,統計異質性假設迴歸模型的誤差項未必服從相同的分配(如常態 分配),而空間異質性則假設模型參數值隨地點改變。現今多數人口研究仍以常見迴歸模型為主,但不少文獻顯示空間和統計異質性普 遍存在,引進GWQR可更精確地描繪出變數間的關係,本文分析美國新型冠狀病毒(coronavirus disease)病例數、疫苗接種率,發現在 接種率較低州郡,兩者關聯性較高。不過,這類型迴歸模型的估計較為複雜,以常見的地理加權迴歸(geographically weighted regression, GWR)為例,假設目標變數與解釋變數間的關係因地而異,因自由度不足使得估計方法較為複雜,GWR引進距離愈遠、權重愈小的概念,各地點參數在估計時會加入鄰近地區資料,再給予不同權重作為依據,但參數估計值大多會有不夠平滑的問題。

除了兩篇專題論文,本期亦收錄劉千嘉所撰寫的研究論文〈高齡租屋族群的特性與租屋歷程:奠基於住宅普查與住宅需求動向調查的發現〉。這篇論文探討屋宅租賃市場中高齡獨居者的處境狀 況,呈現臺灣租屋的高齡者及其家庭的樣貌。該文利用2010年人口與住宅普查資料,分析高齡租屋族群的人口、家庭型態與其住宅的 特性,並以同年度內政部住宅需求動向調查,勾勒其租屋的歷程及其面臨的處境。該研究發現多數高齡租屋族群處於獨居或近似獨居 的狀態,同時,相較於與親人同住的高齡者,獨居或近似獨居者承租到的屋宅狀況較差,而獨居者的屋宅狀況更差。這篇論文透過媒 體所報導的蝸居老人現象,由於個人與家庭的經濟能力,或受租屋市場的年齡歧視之故,使這群高齡者在晚年階段依舊處於輾轉流徙 的不安態勢中。這篇論文可提供政府思考如何透過政策的協助,避免讓高齡的經濟弱勢族群,陷入雙重困境,並從社會與社區資源面向提出思考的觀點。

《人口學刊》自本期起陸續出版「空間人口學」專刊的論文,介紹利用空間統計方法在理解人口分布與歷程的最新前沿議題與應 用,以期提供人口學領域的學術界與政府單位參考。未來本刊持續作為國內提供相關議題研究成果與政策意涵的重要平臺,並期望能 引領人口學研究的議題設定與跨領域研究,提供研究者一個更多元的對話空間。因應電子化時代,本刊投稿方式以繳交電子檔為主, 檔案格式為word文件檔、pdf檔。與編輯部聯絡的E-mail: psc@ntu.edu.tw;以及電話:(02)3366-3405。感謝您對本刊的支持與愛護,敬請舊雨新知繼續支持,惠賜大作。

專題論文:空間人口學

DOI : 10.6191/JPS.202212_(65).0001

全文下載


spatial scaling ; fractal dimension ; dengue fever ; clustering phenomenon ; critical scale ; 空間跨尺度 ; 碎型維度 ; 登革熱 ; 群聚現象 ; 臨界尺度
中文摘要
疾病的空間群聚分析中常用點資料型態對病例分布位置進行標記與分析。因為點資料的無尺度特性與可調整面單元問題,計算出適當的分析尺度對進行點分布的群聚分析非常關鍵。本研究提出一個資料導向的新分析架構以計算出臨界尺度。此架構是基於兩項經典分析概念:一、點區塊四分樹空間索引方法,二、碎型型態分析中的計算格子法。這兩者都具有捕捉空間跨尺度過程的特性,而這跨尺度特性即為此分析架構的核心概念。本研究應用臺灣高雄市過去近20年的登革熱病例資料作為分析對象,應用前述分析架構計算出每一波疫情年的臨界尺度。透過兩種群聚分析方法,包括核密度推估法與基於密度群聚與噪點分析法,本研究對臨界尺度所計算出來的群聚結構與其他尺度計算所得之群聚結構進行比較與討論。這兩項群聚分析都需設定一距離參數,即其搜尋半徑。因此,本研究透過設定不同的搜尋半徑,借用兩項群聚分析方法所得的群聚型態作為比較不同尺度下的點分布型態以進行比較討論。兩項群聚分析的比較結果都顯示在臨界尺度所計算的群聚結構特性相對較可突出其分布型態。
關鍵詞:空間跨尺度、碎型維度、登革熱、群聚現象、臨界尺度
Abstract
In a disease outbreak context, disease cases are usually presented by using point distribution data. Due to the scale-invariant issue of point data and the scaling issue of the modifiable areal unit problem, identifying a critical scale for the analysis of point patterns, such as the clustering phenomenon, is important. This study proposes a novel data-driven framework for calculating the critical scale based on two traditional concepts: (1) the point-region quadtree spatial indexing method and (2) the box counting method for fractal pattern analysis. Both concepts capture the spatial scaling process and serve as the core concepts of the proposed framework. Using dengue fever cases in Kaohsiung City, Taiwan, during the past two decades, the critical scale was identified for each outbreak year. Two clustering analysis approaches were used to test the resulting critical scales, including kernel density estimation and density-based spatial clustering application with noise. Both clustering analyses involved distance parameter settings. Therefore, through the setting of search radii, the two clustering methods were used as a tool to explore the clustering patterns under different scale levels. In summary, the identified critical scales can better capture the spatial patterns of point data.
Keywords: spatial scaling, fractal dimension, dengue fever, clustering phenomenon, critical scale

DOI : 10.6191/JPS.202212_(65).0002

全文下載


heterogeneity ; homogeneity ; geographically weighted regression ; quantile regression ; spatial demography ; 異質性 ; 同質性 ; 地理加權迴歸 ; 分量迴歸 ; 空間人口學
中文摘要
愈來愈多人對於探索自變數與反應變數間在不同分量下(使用分量迴歸〔quantile regression〕)或在空間上(使用地理加權迴歸〔geographically weighted regression〕)之異質關係感到興趣。前者通常被視為一種統計異質性,而後者則指的是空間異質性。然而,人口研究在採用這些方法方面進展略顯緩慢。本研究首先簡要討論了為何人口研究者需要多加關注異質性的概念,並介紹了一種同時考量統計和空間異質性的方法,即地理加權分量迴歸(geographically weighted quantile regression, GWQR)。我們以美國各郡新型冠狀病毒(coronavirus disease, COVID-19)疫苗接種率之資料為例,說明如何應用GWQR,並解釋GWQR如何分析疫苗接種率及其因子之間在空間和疫苗接種率分布上的異質性。研究結果顯示,空間和統計異質性是普遍存在的。例如,COVID-19病例率在位於疫苗接種率低分位數的郡中其關聯性比在高分位數的郡中更強。這種關聯性的空間分布集中在山區州的郡,並轉移到中西部地區。因此,我們認為人口研究應考慮這兩種異質性。
關鍵詞:異質性、同質性、地理加權迴歸、分量迴歸、空間人口學
Abstract
There is a growing interest in exploring heterogeneous associations with independent variables across the distribution of either the dependent variable (using quantile regression) or across geographic space (using geographically weighted regression). The former is often known as statistical heterogeneity, whereas the latter refers to spatial heterogeneity. However, population research has been slow to adopt either of these methods. This study first briefly discusses why more attention to the concept of heterogeneity is needed and then introduces a method that simultaneously considers statistical and spatial heterogeneity, namely geographically weighted quantile regression (GWQR). We illustrate how to use GWQR with U.S. county-level coronavirus disease (COVID-19) vaccination data and explain how GWQR identifies significant heterogeneities in the relationships between the vaccination rate and its determinants across space and over the vaccination distribution. The results suggest that both spatial and statistical heterogeneity are a common occurrence. For example, the COVID-19 case rate has a stronger association in counties in the lower quantiles than in the higher quantiles. The spatial distribution of this relationship is focused on counties in the Mountain states and is shifted to the Midwest region. As such, we conclude that both heterogeneities should be considered in population research.
Keywords: heterogeneity, homogeneity, geographically weighted regression, quantile regression, spatial demography

研究論文

DOI : 10.6191/JPS.202212_(65).0003

全文下載


高齡租屋者 ; 獨居老人 ; 年齡歧視 ; 人口及住宅普查 ; 住宅需求動向調查 ; elderly tenants ; elders living in solitude ; age discrimination ; Population and Housing Census ; Housing Demand Survey
中文摘要
媒體數度以「蝸居老人」指稱一群都市中居住在簡陋出租屋宅的高齡者,2021年高雄城中城事件,讓弱勢群體居住問題再次浮上臺面,凸顯出屋宅租賃市場中,高齡獨居者的劣勢。本文運用2010年人口及住宅普查資料,呈顯高齡租屋族群的人口、家庭型態,與其住宅的特性,並以同年度內政部住宅需求動向調查,勾勒其租屋的歷程與處境。研究初步發現:一、2010年有七成七的高齡租屋族群處於獨居或近似獨居的狀態;二、相較於與親人同住的高齡者,獨居或近似獨居者承租到的屋宅狀況較差,而獨居者的屋宅狀況更差,顯示租屋年齡歧視對獨居者的影響更大;三、高齡者傾向於以親友介紹或傳統廣告的方式尋找租屋,在租屋歷程所經歷的困難較多,其承租的屋宅租金較便宜但坪數相對較小,且租金對於高齡者所造成的壓力亦較非高齡者高;四、未婚與經濟弱勢的高齡者較易承租到租屋市場中的次級產品;五、高齡租屋者所能承租到的房子,受其教育程度、家庭月收與家庭結構所影響。
關鍵詞:高齡租屋者、獨居老人、年齡歧視、人口及住宅普查、住宅需求動向調查
Abstract
The media has been repeatedly using the expression "elders in humble dwellings" to refer to elderly people who live in shabby rental housing in urban areas. A residential building fire in 2021 in Kaohsiung brought the housing problems of vulnerable groups back to forefront, highlighting the disadvantages experienced by the solitary elderly in the rental housing market. This study draws on the 2010 Population and Housing Census data to represent the population, family patterns, and housing characteristics of elderly tenants. It also outlines their renting experience and situation based on the Housing Demand Survey conducted by the Ministry of the Interior in the same year. The preliminary findings reveal that (1) In 2010, 77% of elderly tenants lived in solitude or quasi-solitude; (2) The housing conditions of the elderly who lived in solitude or quasi-solitude were poorer than for those who lived with relatives. The housing conditions of those who lived in solitude were even worse, showing that age discrimination in renting has a more significant impact on them; (3) The elderly tended to rent through the recommendation of relatives and friends or traditional advertisements. They experienced more difficulties in house renting. The housing they rented was cheap; however, its square footage was relatively small. Moreover, the rental pressure on the elderly was also higher than that on the non-elderly; (4) The unmarried and economically challenged elderly were more likely to rent secondary products in the rental housing market; (5) Finally, the housing conditions of elderly tenants were affected by their education level, family monthly income, and family structure.
Keywords: elderly tenants, elders living in solitude, age discrimination, Population and Housing Census, Housing Demand Survey