基于K-Means聚类和Logit回归的上海市非沪籍青年“反向就业”影响因素研究

Research on the Influencing Factors of “Reverse Employment” of Non-Shanghai Youth in Shanghai Based on K-Means Clustering and Logit Regression

摘要: 在当前一线城市就业市场呈现出“内卷化”趋势，就业形势也变得日益严峻的情况下，青年作为具有强创新性的群体，开始主动做出了就业方式上的转向——“反向就业”。基于此，本文以上海市为例，对上海市非沪籍青年“反向就业”意愿及影响因素进行研究。首先，本文根据预调查的结果，结合计算得到的样本量，利用分层抽样等方式对上海市18~35岁非沪籍青年随机发放问卷，获取样本数据；其次，在描述性统计分析上海市非沪籍青年反向就业情况的基础上，利用K-Means聚类算法对整体受访者进行特征分析，利用手肘法确定两类群体的最佳聚类中心个数为4后，分别刻画群体画像，得到典型反向就业和非反向就业青年的群体特征；再次，结合二元Logit回归模型定量分析不同基本特征对于青年群体反向就业选择意愿的影响；最后，提出对选择“反向就业”青年的支持对策，提升“反向”就业地的吸引力。

Abstract: In the current first-tier cities, the job market is showing a trend of “involution”, and the employment situation has become increasingly difficult. The youth, as a group with strong innovation, began to take the initiative to make a shift in the employment mode—“Reverse Employment”. Based on this, this paper takes Shanghai as an example to study the willingness of non-Shanghai youth to “reverse employment” and the factors affecting it. Firstly, based on the results of the pre-survey, combined with the calculated sample size, researchers randomly distribute questionnaires to non-Shanghai youths aged 18~35 in Shanghai using stratified sampling to obtain sample data. Secondly, on the basis of descriptive statistics to analyze the reverse employment of non-Shanghai youth in Shanghai, the K-Means clustering algorithm is used to characterize the overall respondents. After determining the optimal number of clustering centers for the two types of groups as 4 by using the elbow method, the group portraits were portrayed separately to obtain the group characteristics of typical reverse employment and non-reverse employment youths. Thirdly, a binary Logit regression model is used to quantitatively analyze the influence of different basic characteristics of the youth groups on their willingness to choose reverse employment. Finally, supportive countermeasures are proposed to enhance the attractiveness of the reverse employment sites for the youths who choose “reverse employment”.