上大学会多大程度上推迟初婚年龄?——基于IV-Tobit模型的估计

在估计高等教育会多大程度上推迟初婚年龄时会同时面临生存数据结构(Duration data)与内生性问题,以往文献无法很好地解决这个问题。基于已有研究方法,本文设计了一种计量策略来解决这一问题。首先将该问题等价转化为研究在给定年龄的条件下高等教育对已婚时间的影响,从而将数据转化为一个单方向截尾的数据结构并使用Tobit模型进行估计。然后用是否在扩招后参加高考作为高等教育的工具变量,用IV-Tobit模型估计高等教育对初婚年龄的推迟效应。估计结果显示,相对于高中毕业生,每多接受1年高等教育平均初婚年龄将推迟1.5年;相应地,大专、本科教育会将初婚年龄推迟4.5和6年。由于女性生育存在一个最佳的年龄阶段,因此该推迟效应可能会严重影响生育率和母婴健康。另外,高等教育对男性初婚推迟的效应大于女性,对非农业户口的影响大于农业户口,在东、中、西部地区没有差异。本文发现这种推迟效应主要是在校接受高等教育带来的顺延,高等教育并没有显著降低接受者的结婚意愿,上大学只是推迟婚姻,而不是终身不婚。人口普查数据显示我国高等教育人口与较低教育人口的最终结婚率几乎没有差异。不同于其他发达国家,我国的高等教育群体仍然具有较高的结婚率。政府应该采取积极措施降低他们步入婚姻的经济成本,延缓这种婚姻推迟的趋势。     

中国女性初婚进度的新变动、原因和趋势

文章采用队列分析视角，利用七普资料重构中国城乡女性出生队列的初婚进度。研究结果表明：（1）列克西斯平面图可以较为清晰地展现中国女性初婚变动历程中年龄、时期、队列3个维度的特征。从时期效应看，2010年以来女性初婚风险率快速下降、初婚年龄不断推迟，农村女性因起步晚而变动更快；从队列效应看，1986年以后出生的年轻队列初婚风险率整体降低，终身曾婚比例可能与较早队列存在较大差异。（2）七普数据表明2019年城乡女性初婚率大幅降低，调查误差不足以解释这一变动，婚姻市场性别失衡、女性教育提升及性别观念改变可能造成女性初婚率和终身曾婚比例的真实下降。（3）利用初婚年龄分布参数模型进行的外推预测结果表明，1980年之后出生队列终身未婚比例将较快上升，到1990年队列预计达到4.6%～9.0%；其中城市女性更高，预计达到4.5%～13.8%；乡村女性异质性更强，社会压力很难扩散而趋弱，终身未婚比例在1985年之后队列上升很快，将超过镇女性。较晚队列晚婚和不婚的趋势难以逆转，将深刻影响生育水平和婚姻家庭。     

1990～2020年中国的婚育推迟：变化趋势与典型特征

基于1990～2020年历次全国人口普查和人口抽样调查数据，对中国不同人群婚育推迟的变化趋势与典型特征进行分析。研究发现，中国平均初婚年龄不断提高，初婚推迟在城乡普遍发生但并不同步，2000～2010年城市初婚推迟快于农村，2010～2020年农村初婚推迟加快，导致城乡差距先拉大后缩小；受教育程度提高显著推迟人们的初婚时间，其中，高等教育的影响存在性别差异，其在推迟男女两性初婚时间的同时，降低了男性终身不婚的可能性；中国育龄妇女生育率曲线的峰值向右向下移动，城市育龄妇女生育峰值年龄推迟更快；育龄妇女平均生育年龄不断提高，2000～2010年生育推迟速度快于2010～2020年；各孩次平均生育年龄均不断推迟，其中，一孩和二孩生育推迟更明显；不同受教育程度的育龄妇女均表现出生育推迟趋势，且受教育程度越高，生育推迟越明显。     

婚前同居对初婚年龄的影响研究 ——基于Heckman二阶段模型的分析

在现代社会中,初婚年龄持续攀升,婚前同居现象越来越多地出现。那么婚前同居与初婚年龄之间有没有关系?本文基于婚前同居对离婚影响的选择假说与经历假说进一步探讨两种假说与初婚年龄之间的关系。选择假说认为同居者自身特征促使他们追求思想解放,消极地影响着婚姻稳定性,因此与其同居伴侣之间并不想建立一个长期的契约。由此推测这些积极影响同居的特征因素会给结婚带来消极影响,婚前同居就会推延初婚年龄;经历假说认为共同生活的经历将会改变同居者对婚姻的认识,他们不再强烈地致力于追求婚姻的身份。他们将会接受婚前同居这种暂时性的本质。很多夫妇认为同居提供了一个婚姻的替代品,结婚的欲望由此下降。因此这也将推延初婚年龄。除两个假说外,同居者想要通过同居搜集更多对方的信息,这种想要更多地了解对方的想法也将推延初婚年龄。在理论分析的基础上,本文基于2010年CFPS抽样调查数据,采用Heckman二阶段模型纠正婚前同居的自选择性带来的偏误并检验婚前同居推延初婚年龄的研究假设。研究结果发现婚前同居使女性与男性的初婚年龄分别推迟了0.415岁与0.868岁。在控制婚前同居的自选择性后,婚前同居对女性与男性初婚年龄的影响仍然很大且显著,初婚年龄分别推迟了0.431岁与0.890岁,证明婚前同居将会提高初婚年龄。这进一步明晰了婚前同居在我国家庭形成过程中的作用。婚前同居对初婚年龄的推延作用无疑为我国全面二孩政策下出生人口数量增长带来消极影响。     

甘肃省人口婚姻意愿调查分析

本文利用问卷调查数据,分析了甘肃省居民婚姻意愿现状,主要对择偶标准、期望初婚年龄、期望夫妻年龄差和期望婚居模式四个方面进行分析。调查发现,甘肃男性择偶时较看重身体健康,甘肃女性择偶时较看重经济收入;甘肃居民的平均初婚年龄逐步提高;绝大多数居民期望婚姻中丈夫的年龄大于妻子,且倾向于男大于女1-2岁;从夫居和独立居住是普遍接受的婚居模式。虽然现代婚姻观对甘肃居民有一定影响,但是这里的婚姻观仍较传统。     

结婚年龄之制约性条件研究——平均初婚年龄为什么推迟了

在法定初婚年龄较低的今天，西方发达国家的平均初婚年龄呈现出一个较高的年龄段．发展中国家的平均初婚年龄也处于一个不断提高的过程中，这是由经济发展水平所决定的，同时在内在因素上受结婚对象的机会成本比较、教育程度的提高、结婚成本的增加和性行为、同居现象的普遍化影响。婚姻与性及生育的逐渐背离使得结婚更成为一种社会化的标志，而不是生理上的需要。     

初婚年龄变迁的稳定因素——基于代际传递视角的实证探索

中国人口的初婚年龄呈现出明显的推迟趋势。推迟的原因是“促使变化”与“保持稳定”两方面因素共同作用的结果,但现有研究对后者的关注相对缺乏。从代际传递视角出发,利用2010—2018年中国家庭追踪调查(CFPS)数据,对初婚年龄变迁的“稳定因素”进行实证探索,以事件史分析为主要分析方法。研究发现：初婚年龄存在显著的代际影响,但父亲与母亲的影响存在差异。父亲初婚年龄显著影响子女(特别是儿子)的初婚年龄,呈现出代际传递特征;但母亲的初婚年龄无显著的独立影响,且不因子女性别而变化。除了差异,父母间的代际影响又彼此联系,父母初婚早晚的一致性较高,如均为早婚或晚婚,则代际传递影响会更加明显。进一步将代际影响置于宏观环境中发现,当父亲初婚年龄与周围他人的平均初婚年龄差距较大时,父亲的代际影响更为凸显。初婚年龄的代际传递体现了父母及家庭对于个体婚姻行为的重要意义,可视为初婚年龄变迁中重要的“稳定因素”。推而言之,代际传递影响可为进一步分析我国婚姻家庭变迁中变化与不变共存的现实提供解释路径和分析视角。     

中国分性别婚姻推迟及其补偿研究北大核心

自20世纪80年代中后期以来我国人口出现了明显的婚姻推迟现象。婚姻推迟与补偿问题不仅关系婚姻家庭本身,而且与生育水平紧密相关,因而受到广泛关注。然而,以往相关研究存在以下不足:第一,对于婚姻推迟仅仅是人口步入婚姻的“进度”上的减慢,还是随着结婚年龄的推迟最终导致相当比例的人终身不婚这一问题,现有研究持不同观点,需要进一步检验。第二,以往相关研究多关于女性,但在中国婚姻挤压非常严重的人口结构背景下,分析比较两性婚姻推迟与补偿的特点更具有意义。第三,以往研究在方法上多使用净婚姻表分析其主要指标,但净婚姻表因纳入了死亡因素反而不能反映婚姻的“净变化”。鉴于此,文章利用我国最新普查数据,采用粗婚姻表以及婚姻补偿指数等多个统计指标,从队列和时期视角探讨近30年来中国婚姻推迟趋势及其补偿效应,特别关注在婚姻挤压背景下婚姻推迟与补偿的性别差异。研究结果表明:中国人口的婚姻推迟程度不断加深,近20年男性婚姻推迟程度高于女性;男女婚姻推迟的补偿效应不断减弱,且男性婚姻推迟的补偿效应低于女性。根据男女婚姻推迟与补偿、终身未婚比例的变化趋势,未来15-20年我国男性将过渡到晚婚不婚模式,而女性还将保持在晚婚普婚模式。我国婚姻推迟与补偿的性别差异与婚姻挤压有密切关系。     

高校扩招对我国初婚年龄的影响——基于普查数据的分析

利用第五次和第六次全国人口普查数据,考察了自高校扩招以来我国人口受教育结构发生的巨大变化对我国初婚年龄的影响.结果显示,高校扩招后,我国各年份出生同期人的中位初婚年龄持续上升;在2010年,受益于高校扩招的适婚人口的已婚比例相比2000年同龄人的已婚比例显著下降,且受高等教育人口的已婚比例低于未受高等教育人口的已婚比例.高校扩招带来的人口受教育结构变化对我国初婚的推迟现象具有较强的解释力,解释力表现出随年龄增加逐渐减小的特点,并具有性别差异,其中对“六普”时点23岁的男性解释力高达78％,对23岁女性的解释力亦超过50％.     

20世纪90年代我国婚育模式的初步分析

20世纪90年代以来,中国女性人口的平均初婚年龄平稳缓慢上升,育龄妇女平均初育年龄在波动中小幅度上升,已婚育龄妇女的平均初婚初育间隔呈波动式扩大。     

The Gendered Double Standard of Aging in US Marriage Markets

Using data on marriages collected in most US states between 1970 and 1988, we show that the older men are when they marry, the more years senior to their brides they are, whether it is a first or higher‐order marriage. While older men with more education marry down in age slightly more than less educated older men, the pattern of men marrying further down if they marry later holds strongly for all education groups. We consider several possible explanations for the tendency of men to marry further down in age if they are older at marriage. While we have no direct measure of physical attractiveness, we argue that the most compelling interpretation is that men, more than women, evaluate potential spouses on the basis of appearance. Because the prevailing standard of beauty favors young women, the older men are when they marry, the less they find women their own age attractive relative to younger women, leading them to marry further down in age if they are older at marriage. The consequence for women of men's preference for youth is more often that they remain unmarried than that they end up married to much older or less educated men.     

Age at marriage and timing of the first birth

Summary An attempt has been made to measure the effect of age at marriage of brides on the timing of the first birth. In Australian vital statistics, first nuptial confinements have been tabulated by age of mothers and by single years of marriage duration in single months for the first two years and by single years for all other durations since 1916. A simple technique has been used to link such data with marriage cohorts. The study briefly reviews the prevailing patterns of the timing of first births by mothers' age at marriage and changes in this pattern since the marriages of the 1925/9 period. The analysis shows that after a period of relative stability of family formation patterns in the 1950s and early 1960s, women married in the late 1960s started postponing the first birth beyond the first two years of marriage. It is suggested that a fraction of the decline in total births recorded in Australia since 1972 can be attributed to the postponement of first nuptial confinements by women married in the late 1960s and early 1970s.     

Late Marriage and Less Marriage in Japan

Between 1975 and 1995, the singulate mean age at marriage in Japan increased from 24.5 to 27.7 years for women and from 27.6 to 30.7 years for men, making Japan one of the latest‐marrying populations in the world. Over the same period, the proportion of women who will never marry, calculated from age‐specific first‐marriage probabilities pertaining to a particular calendar year, increased from 5 to 15 percent for women and from 6 to 22 percent for men—behaviors sharply different from those characterizing the universal‐marriage society of earlier years. This article investigates how and why these changes have come about. The reasons are bound up with rapid educational gains by women, massive increases in the proportion of women who work for pay outside the home, major changes in the structure and functioning of the marriage market, extraordinary increases in the prevalence of premarital sex, and far‐reaching changes in values relating to marriage and family life.     

Child Marriage and Intimate Partner Violence in Rural Bangladesh: A Longitudinal Multilevel Analysis

Child marriage (before age 18) is a risk factor for intimate partner violence (IPV) against women. Worldwide, Bangladesh has the highest prevalence of IPV and very early child marriage (before age 15). How the community prevalence of very early child marriage influences a woman’s risk of IPV is unknown. Using panel data (2013–2014) from 3,355 women first married 4–12 years prior in 77 Bangladeshi villages, we tested the protective effect of a woman’s later first marriage (at age 18 or older), the adverse effect of a higher village prevalence of very early child marriage, and whether any protective effect of a woman’s later first marriage was diminished or reversed in villages where very early child marriage was more prevalent. Almost one-half (44.5 %) of women reported incident physical IPV, and 78.9 % had married before age 18. The village-level incidence of physical IPV ranged from 11.4 % to 75.0 %; the mean age at first marriage ranged from 14.8 to 18.0 years. The mean village-level prevalence of very early child marriage ranged from 3.9 % to 51.9 %. In main-effects models, marrying at 18 or later protected against physical IPV, and more prevalent very early child marriage before age 15 was a risk factor. The interaction of individual later marriage and the village prevalence of very early child marriage was positive; thus, the likely protective effect of marrying later was negated in villages where very early child marriage was prevalent. Collectively reducing very early child marriage may be needed to protect women from IPV.     

Cohabitation and Marital Expectations Among Single Millennials in the U.S.

Cohabitation has surpassed marriage as the most common union experience in young adulthood. We capitalize on a new opportunity to examine both marital and cohabitation expectations among young single women in recently collected, nationally representative data (National Survey of Family Growth 2011–2015) (N?=?1467). In the US there appears to be a ‘stalled’ second demographic transition as single young adult (ages 18–24) women have stronger expectations to marry than cohabit and the vast majority expects to, or has, already married. Among young women expecting to marry, the majority (68%) expect to cohabit with their future spouse but about one-third expect to follow a traditional relationship pathway into marriage (to marry without cohabiting first). In addition, women from disadvantaged backgrounds report the lowest expectations to marry, but there is no education gradient in expectations to cohabit. Marriage expectations follow a “diverging destinies” pattern, which stresses a growing educational divide, but this is not the case for cohabitation expectations. Our results, based on recently collected data, provide insight into the contemporary context of union formation decision-making for the millennial generation.

     

How female education affects reproductive behavior in urban Pakistan

Although Pakistan remains in a pretransitional stage (contraceptive prevalence of only 11.9% among married women in 1992), urban women with post-primary levels of education are spearheading the gradual move toward fertility transition. Data collected in the city of Karachi in 1987 were used to determine whether the inverse association between fertility and female education is attributable to child supply variables, demand factors, or fertility regulation costs. Karachi, with its high concentration of women with secondary educations employed in professional occupations, has a contraceptive prevalence rate of 31%. Among women married for less than 20 years, a 10-year increment in education predicts that a woman will average two-fifths of a child less than other women in the previous 5 years. Regression analysis identified 4 significant intervening variables in the education-fertility relationship: marriage duration, net family income, formal sector employment, and age at first marriage. Education appears to affect fertility because it promotes a later age at marriage and thus reduces life-time exposure to the risk of childbearing, induces women to marry men with higher incomes (a phenomenon that either reduces the cost of fertility regulation or the demand for children), leads women to become employed in the formal sector (leading to a reduction in the demand for children), and has other unspecified effects on women's values or opportunities that are captured by their birth cohort. When these intervening variables are held constant, women's attitude toward family planning loses its impact on fertility, as do women's domestic autonomy and their expectations of self-support in old age. These findings lend support to increased investments in female education in urban Pakistan as a means of limiting the childbearing of married women. Although it is not clear if investment in female education would have the same effect in rural Pakistan, such action is important from a human and economic development perspective.     

Attitudes of welfare recipients toward marriage and childbearing

The 1996 Personal Responsibility and Work Opportunities Reconciliation Act (PRWORA) put the need for attitudinal change at the center of efforts to help welfare recipients become economically independent, avoid out-of-wedlock childbearing, and embrace marriage. In this paper, we focus specifically on attitudes, analyzing both differences in values and attitudes between welfare recipients and other women on the cusp of reform, as well as the effects of TANF reforms in two states on the attitudes and behaviors of women subject to the reforms. National data reveal few differences in values and attitudes between welfare recipients and other women once background characteristics are held constant. A majority of both groups believes that prospective parents should marry, but single parents can raise a child as well as married parents. Personal aspirations for marriage and further childbearing also are fairly similar. These similarities may be one reason that the literature has shown TANF to have limited effects on marriage and childbearing. A second set of analyses investigates the degree to which welfare recipients in Delaware and Indiana report that reforms affected their aspirations for marriage and childbearing. Self-reported impacts are greater for fertility than marriage attitudes. Even among those who report their attitudes were affected by reform, recipients appear to have difficulty acting on their marital and childbearing desires, dampening any effects on behavior. These findings reinforce the current sense among researchers and policy makers that more direct reforms are needed to have a substantial effect on marriage and out-of-wedlock childbearing.     

Family transitions in Young Adulthood

Using the first (1995) and third (2001-2002) waves of the Add Health survey, we examine women 's family transitions up to age 24. Only a third of all women marry, and a fifth of those marriages dissolve before age 24. Three out of eight women have afirst birth, with a substantial majority of those births outside of marriage: 66% for whites, 96% for blacks, and 72% for Mexican Americans. Cohabitation is the predominant union form; 59% of women cohabit at least once by age 24. Most cohabitations are short lived, with approximately one in five resulting in a marriage. We summarize the family and relationship experience of women up to age 24 in terms offour categories, each accounting for roughly a quarter of all women. Category 1 has the women who remain single nonparents. Category 2 has the early marriers, women whose marriage is not preceded by a first birth. Category 3 has those who become single parents. Category 4 has the women who cohabit at least once, but who do not marry or have a birth by age 24. The strictly ordered transitions of the 1950s are long gone and have been replaced by a variety of paths to adulthood.     

The conflicting effects of delayed marriage and declining divorce rates on cumulative fertility in Indonesia

The impact of nuptiality patterns on fertility in Indonesia is examined with multivariate analysis controlling for 8 socioeconomic variables. Data were obtained from the 1987 Indonesian Contraceptive Prevalence Survey. Marriage is usually universal by age 35, and in this study all women 30 years had been married at least once. 20% were married at 15 years and 45% married at 18 years. For those married more than once, prevalence of 1st marriage was 7% for women 15-24 years, 15% for 25-34 years, and 29% for 35-49 years. In 1976 and 1987, the age at 1st marriage and number of times married were both strongly and negatively correlated. The % never marrying between 15-49 years rose from 21.5% to 26.4% between 1980-87. Cumulative fertility w as related to both age at 1st marriage and number of times married. Muslim women, women in Java and Bali, and rural women all marry at younger ages. 27% of the variance in age at 1st marriage is explained by women aged 25-34, current residence, region, religion, language, education, and work or not before marriage. The number of times married is also associated with socioeconomic characteristics without control, i.e., Muslim women 25-34 years were 3 times more likely to have been married more than once than in other faiths. With controls for socioeconomic factors, only 13% of the variance is explained and being Muslim has no statistically significant effect. The important net effects were being interviewed in Balinese, age, and age at 1st marriage. In the analysis of cumulative fertility, age at 1st marriage consistently is related to cumulative fertility in almost every socioeconomic group when age and number of times married is controlled for. Women married more than once have lower fertility. 36% of the variance is explained by all the variables. Being married more than once leads to having 2.1 fewer children. A 5-year delay in marriage leads to .75-1.1 fewer children. When other variables are controlled for, neither educational level nor prior work experience has a statistically significant effect on cumulative fertility. In the contraception analysis, women married more than once used contraception less. Among women 35-49, those marrying later had higher contraceptive use, but in general contraceptive use declined with age. More information is needed on why marriage patterns are changing.