杭州市基层保安职业认同分型与精准干预研究——基于聚类分析视角

A Study on the Typology and Precise Intervention of Professional Identity among Grassroots Security Guards in Hangzhou—Based on a Clustering Analysis Perspective

摘要: 现有底层劳动者职业认同研究多采用变量中心回归范式，默认群体同质，政策干预普遍“一刀切”。本文以杭州市430名小区保安为样本，承接集成学习识别的核心影响维度，采用K-means聚类划分高/中/低认同三类保安群体，结合ANOVA效应量锁定区分群体的三大核心诊断靶点，通过多群组结构方程模型检验干预效用的群体非对称差异。结果显示：工作不洁感知、职业焦点重聚、工作内容认同是群体分型核心指标；改善环境、认知培训两类干预对低认同群体边际提升效果显著高于另外两类。研究构建“异质性识别–靶点诊断–精准干预”完整分析框架，为物业、街道实施分层差异化保障政策提供实证支撑。

Abstract: Existing studies on the professional identity of lower-tier workers mostly adopt a variable-centered regression paradigm, assuming group homogeneity, and policy interventions are generally “one-size-fits-all.” This study uses a sample of 430 community security guards in Hangzhou, identifies key influencing dimensions through integrated learning, applies K-means clustering to divide the guards into high, medium, and low professional identity groups, and, combined with ANOVA effect sizes, identifies three core diagnostic targets distinguishing the groups. A multi-group structural equation model is then used to test the asymmetric effectiveness of interventions across groups. Results show that perceived job uncleanliness, occupational focus regrouping, and recognition of work content are core indicators for group typology; environmental improvements and cognitive training interventions yield significantly higher marginal benefits for the low-identity group compared to the other two groups. The study constructs a complete analytical framework of “heterogeneity identification-target diagnosis-precise intervention,” providing empirical support for property management and street offices to implement differentiated assurance policies by layers.