What Are the Determinants of Intentions for Childlessness?
Anna Rybińska, University of North Carolina at Chapel Hill
Permanent childlessness can be a result of developing an intention to remain childless. Descriptive studies of the childbearing intentions indicate that women develop an intention for a childless life style well before their fecundity starts to decline due to their age (Berrington 2004; Rybinska and Morgan 2016). In addition, women who report a childless intention rarely “switch back” to intending and having children in the future (Rybinska and Morgan 2016). At the same time, childless intentions are shown to strongly predict permanent childlessness (Westoff 1990; Rovi 1994). Understanding how childless intentions are developed is thus an important step in understanding the high levels of childlessness in the United States.
In this paper, I investigate the development of intentions for childlessness over the course of women’s reproductive careers. I focus on how key changes in a woman’s life course can increase the likelihood of developing an intention for childlessness. I study three life domains: union history, employment, and education as they are strongly intertwined with childbearing (Gerson 1985; Cain 2001; Cannold 2005). Importantly, existing research is rich with studies of correlates of permanent childlessness but little is known to date about why women develop an intention to remain childless. This project is especially timely because the results from the National Survey of Family Growth indicate that the percent of childless women aged 20-24 who intend to remain childless has doubled between 1982 and 2015.
I estimate logistic regression models with fixed/random effects. A sample of 1,000 women was selected from the 1979 National Longitudinal Survey of Youth followed from the age of 14-15 until the age of 45; when 20% of the respondents were permanently childless.
The measure of intentions for childlessness is created using a questionnaire item “How many (more) children do you expect to have?”. Childless women who provide an answer of zero are classified as intending childlessness. Conceptually, childbearing expectations are different from childbearing intentions. Childbearing intentions can be viewed as purposeful plans to have children in the future and childbearing expectations are people’s best prediction of future outcomes. Thus, expectations might include one’s preferences but could also reflect one’s knowledge about her reproductive health or her ability to effectively use birth control. For the purpose of this study, I treat expectations as equivalent to intentions. My first premise of doing so is empirical evidence that respondents do not distinguish between intentions and expectations (Ryder and Westoff 1969; Morgan 2001, p.154). Secondly, women in this specific cohort could only assess their potential use of contraception to a limited extent due to a narrow choice of contraception methods.
Four key events are included in the analyses: post-secondary enrollment, professional advancement, marriage, and divorce. A respondent is classified as enrolled in post-secondary education if her educational attainment changes from 12th grade in year t to more than 12th grade in t+1. Next, if she enrolls in another post-secondary degree, this change is coded as a higher number on a scale. Occupational information is coded using the 1970 Census 3 digit system and conceptualized as the change in occupational codes between time t and t+1. Each subsequent promotion to a higher position is coded as a higher number on a scale. A respondent is classified as divorced if she is married at time t but divorced or separated at t+1. To account for the process of motherhood postponement, I will include a time-varying measure of the respondent’s age. In my analyses that include random effects, I will control for a number important confounding time-invariant covariates, such as race/ethnicity, family background, religiosity and attitudes toward gender roles.
I expect to find positive effects of the changes in educational and professional careers on women’s intentions for childlessness. Enrolment in higher education and occupational advancements will increase the likelihood of developing an intention for childlessness. Similar effects are expected for divorce – women will be more likely to develop an intention for childlessness following a divorce. Marriage is expected to decrease the likelihood of developing an intention for childlessness for women.
Presented in Session 1157: Fertility