Data, variables and analysis strategies

Data

Data for this study come from a survey conducted by the Shanghai Social Science Research Center of Shanghai University from November 2014 to October 2015 on the living conditions of megacities Beijing, Shanghai and Guangzhou.² The survey employed a two-stage sampling method. The first stage consisted of a random map sampling; 50 communities were selected in each city, 20 households were selected from each community, and from each family a resident aged 18-65 was selected as an interviewee. Sample data was obtained from a total of 3,004 regular residents. Due to the particular requirements of this study, the statistical analysis excluded students from the sample, leaving a total of 2,889 samples. These data served as the basis for analysis.³ The basic characteristics of the sample are shown in Table 5.1.

Operationalization of variables

Dependent variables

In this study, the dependent variable is the subjective class identity of urban residents. This variable is measured by "What do you think the level of your overall status is in the country?” The options are divided into "upper level”, "upper-middle level”, “middle level”, "lower-middle level” and "lower level”. These responses were converted to a gradient of 1 to 5 points; 1 point represented the "lower level” while 5 points represented the "upper level”. Housing residents with higher status identity had a higher score, and vice versa.

Independent variables

In this study, independent variables include two broad categories. The first category is "housing quality”, which includes the following four variables: (1) Housing property rights. For this variable, the current situation of urban residents' housing was taken into account; questions were asked about the property rights of the current residential houses (categorized as complete ownership, shared property rights, rental, and so forth). (2) Housing surface area. Although the surface area of a house can refer to either the area of use (residential area) or the building

Table 5.1 Basic characteristics of the sample

Note: Because some values are missing, the sum of all sections of the data may not equal 2,889.

area of a house, in the current commercial housing market, housing property rights usually use the building area as the basic unit of measurement for the area of a house. With the building area and the number of people living together, the indicator of “per capita housing area” can be obtained. (3) Housing expendi- ntre refers to the proportion of respondents' spending on home purchases, rent or mortgages in the past year as a percentage of total household income. (4) The market value of housing was calculated using the market value of the respondents' ownership in property housing.

The other category of independent variable is "symbolic distinction”, which specifically includes the following two variables: (1) The type of community in which the housing is located. Types of community include unrenovated old districts, single or mixed unit communities, affordable housing communities, common commercial housing estates, villa neighborhoods or high-end residential areas. (2) Property management fee payment standards. It was investigated whether the community in which the respondent lives requires payment of property management fees and the corresponding payment amount.

Analysis strategy

In this study, the authors attempt to reveal that great differences and disparities exist among residents in the key element of housing. To this end, descriptive analysis of the relevant variables in the fields of "housing quality” and "symbolic distinction” is carried out for the regions of Beijing, Shanghai and Guangzhou. On the other hand, by establishing a multivariate linear regression model containing the above-mentioned variables and examining from the perspective of wealth stratification, the study focuses on the extent to which housing, as an important symbol of wealth, affects the class identities of urban residents. The basic characteristics of the sample are shown in Table 5.1.