Carstairs deprivation scores by CATT2, 1981, 1991, 2001

Carstairs deprivation scores were used to measure relative deprivation differences between small areas and over time, as the scores are widely used and recognised (Carstairs and Morris 1989; Carstairs and Morris 1991). The Consistent Areas Through Time (CATTs, Exeter, Boyle et al. 2005) were used to provide a consistent geography between the 1981, 1991 and 2001 censuses, for ease of comparison. Specifically the CATT2 small area geography was use, giving 10,058 individual areas with an average population in 2001 of approximately 500 persons. Carstairs deprivation scores were unavailable for the CATT2 geography, therefore were calculated from raw census data. This also enabled a second set of scores to be calculated without the car ownership component (called Adjusted Carstairs hereafter), as car ownership is more of a necessity in rural areas compared with urban areas and can bias deprivation scores (Christie and Fonea 2003). Although unadjusted Carstairs scores have been used for national studies this was considered a useful opportunity to investigate how the car ownership component affected scores in rural areas. The Carstairs score is constructed from four components that have been shown to measure deprivation well (Carstairs and Morris 1989): 1. Overcrowding: the percentage of all persons living in private households with a density of more than one person per room. 2. Male Unemployment: the percentage of economically active males seeking or waiting to start work 3. Low Social Class: the percentage of all persons in private households with an economically active head with head of household in social class IV or V. 4. No Car: the percentage of all persons in private households which do not own a car.

本研究采用卡斯特尔斯剥夺得分（Carstairs deprivation scores）衡量小区域间及跨时间的相对剥夺差异，因该得分应用广泛且得到学界认可（Carstairs与Morris，1989；Carstairs与Morris，1991）。为便于比较，本研究采用跨时间一致性区域数据集（Consistent Areas Through Time，CATTs，Exeter、Boyle等，2005）以实现1981、1991及2001年人口普查间地理边界的统一。具体采用CATT2小区域地理划分方案，共包含10058个独立区域，2001年各区域平均人口约500人。由于CATT2地理划分方案下无现成的卡斯特尔斯剥夺得分，故需从原始普查数据重新计算。同时，本研究还可额外生成不含机动车拥有量维度的修正版得分（下文简称调整后卡斯特尔斯得分），原因在于：相较于城市地区，机动车拥有量在农村地区更属于生活必需条件，若纳入该维度会对剥夺得分产生偏倚（Christie与Fonea，2003）。尽管未经调整的卡斯特尔斯得分已应用于全国性研究，但本次研究仍借此契机，探究机动车拥有量维度对农村地区剥夺得分的影响机制。卡斯特尔斯剥夺得分由四个经证实可有效衡量剥夺程度的维度构成（Carstairs与Morris，1989）：1. 住房拥挤度：私人住户中人均居住密度超过1人/间的人口占比；2. 男性失业率：经济活动男性群体中正在求职或等待入职的人口占比；3. 低社会阶层占比：私人住户中经济活动户主社会阶层为IV或V类的人口占比；4. 无机动车占比：私人住户中未拥有机动车的人口占比。