中国内地31省出生率差异及影响因素分析  

The variation of birth rate and its influencing factors in 31provinces of the Chinese mainland

作　　者：朱荣嘉 刘扬[1,2] 余洪钊 周玉博 李宏田[1,2,3,4] 刘建蒙 ZHU Rongjia;LIU Yang;YU Hongzhao;ZHOU Yubo;LI Hongtian;LIU Jianmeng(Institute of Reproductive and Child Health/National Health Commission Key Laboratory of Reproductive Health,Peking University,Beijing 100191,China)

机构地区：[1]北京大学生育健康研究所/国家卫生健康委员会生育健康重点实验室,北京市100191 [2]北京大学公共卫生学院流行病与卫生统计学系 [3]北京大学第三医院女性生育力促进全国重点实验室 [4]北京大学人工智能研究院智慧公众健康研究中心

出　　处：《中国煤炭工业医学杂志》2024年第2期207-212,共6页Chinese Journal of Coal Industry Medicine

基　　金：国家卫生健康委员会课题“开展孕产妇基本特征多源数据分析研究”;中央高校基本科研业务费专项资金(编号:BMU2021RCZX029)。

摘　　要：目的 分析中国内地31省(不含港澳台地区)出生率差异,探究经济、政策、文化等典型因素对出生率的影响,为优化完善生育相关政策提供依据。方法 基于全国人口普查公报及统计年鉴,收集各省2020年第七次全国人口普查(简称七普,六普、五普以此类推)出生率及相关影响因素数据,包括2020年各省人均国内生产总值、汉族人口占比、生育旺盛期女性占比、商品房平均销售价格、居民人均消费支出、老年人口抚养比以及七普、六普、五普出生人口性别比等。构建多重线性回归模型,采用逐步回归法筛选变量,以偏决定系数(Rv2)评价各因素对出生率的解释程度。对比泊松回归与多重线性回归结果差异,评价研究结果稳健性。结果 2020年中国内地31省总出生率8.52‰,平均出生率8.60‰(标准差2.45‰),范围为3.76‰~13.90‰。省内出生率变异度较省际更小。经多重线性回归筛选,六普出生人口性别比、生育旺盛期女性占比与出生率显著正相关,人均国内生产总值、汉族人口占比与出生率显著负相关,模型决定系数0.767。六普出生人口性别比与出生率相关关系最强(Rv2=0.570),生育旺盛期女性占比次之(Rv2=0.368)。泊松回归结果与多重线性回归基本一致。结论 中国内地各省出生率差别较大,变异系数超25%,最高出生率约为最低出生率的3.7倍。出生人口性别比、生育旺盛期女性占比、人均国内生产总值及汉族人口占比对出生率影响较大。从可干预的角度看,出生人口性别比与出生率相关性较强,提示营造重视生育的文化可促进出生率提高。Objective To analyze the variation of birth rates in thirty-one provinces of Chinese mainland(excluding Hong Kong,Macao,and Taiwan)along with the correlation between birth rate and its typical influencing factors in economic,political,cultural and other aspects,in order to provide reference for improving fertility policies.Methods The study was conducted based on data sourced from National Population Census and national and provincial statistical yearbooks.Main outcome was birth rate of each province in the seventh population census in 2020(referred to as the seventh census,as is the sixth census and the fifth census).The variables in the exploratory analysis included per capita gross domestic product(GDP),the proportion of Han population,the proportion of women in their peak fertility years,average selling price of commercial build ings,per capita consumption expenditure of households,elderly dependency ratio,and the sex ratio at birth(SRB)of the seventh census in 2020,the sixth census in 2010,and the fifth census in 2000.Stepwise regression was used to build a multiple linear model in order to calculate the partial determination coefficient(R2 v),and to evaluate the correlation between each factor and birth rate.The results were compared with those based on poisson regression to achieve robustness.Results The national birth rate of 31mainland provinces was 8.52‰,with the average of 8.60‰(standard deviation of 2.45‰),ranging from 3.76‰to 13.90‰.The variability of birth rates within provinces was smaller than that among provinces.It was found that birth rate was positively correlated with the proportion of women in their peak fertility years and SRB of the sixth census,while negatively correlated with the per capita GDP and the proportion of Han population.The coefficient of determination reached 0.767in the linear regression model.The SRB of the sixth census had the strongest correlation with birth rate(R2 v=0.570),followed by the proportion of women in their peak fertility years(R2 v=0.368).The result

关 键 词：出生率 出生人口性别比 回归分析 

分 类 号：C924.2[社会学—人口学]