Factors Influcing Job-housing Balance

Job-housing balance is an important concept in urban planning, which contributes to easing traffic, reducing commuting and pollution, and improving the living environment. In the case study of Shanghai, regression analysis and multi-statistical methods for job-housing balance are discussed in order to explore its inherent operational mechanism.Job-housing balance influencing factors are complex, many of those factors are mutually influencing. Some factors can be quantitatively measured, and some can only be described qualitatively. Some belong to the individual micro level, while others are standing in the overall global perspective. Based on this, this paper discusses the job-housing balance influencing factors from two perspectives. First, list the influencing factors without considering the mutual interaction between factors, starting from a single factor to explore its impact on commuting time and commuting distance, including measurable statistical factors such as age, gender, education, income, housing property, employment, rail and bus density, and the ratio of total job-housing balance, while also consisting of qualitative factors like agglomeration economies, housing and land market, employment and living space reconstruction. Second, taking into account the mutual interaction between the factors, so the influencing factors were mainly divided into individual micro-level and inter-district property macro level. Based on preliminary survey data and related references, firstly conduct single-layer statistical analysis of the influencing factors, in other words conduct the traditional regression analysis. Then establish multilevel hierarchical linear statistical model, on the basis of demonstrating the necessity of multi layer analysis. The multilevel hierarchical linear statistical model contains three models, zero model, random effects model, and full model. The establishment of the three models not only controls the mutual influence between factors, but also looks into how different factors interacted with each other, for the sake of a more comprehensive and objective interpretation of the various factors that affect job-housing balance.In the single-layer regression analysis, sex, education, population and transportation density, and housing property all affect job-housing balance, while age, income, household employment do not have a great impact on job-housing balance. In the multi hierarchy regression analysis, the ratio of owner-occupied housing has a positive correlation with commuting distance; the commuting length of male is longer than the community length of female, people who have college degrees have longer commuting than those who do not have college degrees.In addition, the establishment of multi hierarchy statistical model also explores how the factors are mutually interacted. According to the analysis findings, the area population density will weaken the impact of education on job-housing balance. The greater the employment density is, the weaker the impact of sex and education on job-housing distance.