2) According to Sayer and Bianchi, how might gender ideology and marital quality be linked? Becker’s specialization model and the income model (Oppenheimer) imply a "cost-benefit" framework for marriage. Can a gender ideology model fit into a similar framework?
3) Sayer and Bianchi argue that controlling for gender ideology is essential in determining the effect of women’s economic independence on divorce. Might we worry about inconsistencies between reported beliefs and exhibited behaviors? Further if surveys do not include specific survey questions about each partner’s attitudes regarding gender roles, can we capture this effect? What might be some proxies for gender ideologies within marriage?
4) Sayer and Bianchi find that increased earnings for women do not increase the odds of divorce when controlling for other characteristics. Do women’s earnings act as a preventative for divorce? In the face of other characteristics that are positively related to divorce, can women’s earnings act as a buffer? What implications does this have for conceptualizing an income effect for women’s earnings?
5) Does the effect of women’s earnings on the probability of divorce change relative to spouse’s earnings (each partners relative earnings)? Relative to absolute family income (i.e. low-income, middle-income, high-income)? Why might this be?
6) Smock finds broadly that the post-divorce disadvantage experienced by women is a function of their role as the primary caregiver of the children previous to and after the divorce. The mechanism for this effect seems to be lower human capital attainment. Smock’s sample is relatively young-- 23-30 years old in 1988-- thus limiting the potential duration of marriage. Understanding that human capital accumulation is a life course process, what hypotheses might we make regarding the economic effects of divorce for longer duration marriages? For marriages that take place at older ages?
7) Smock’s findings lead to a peculiar conclusion: low-income men’s
economic circumstances improve after a divorce. Do we buy this? Does this have
implications for low-income men’s decision to marry? What implications
does this have for each party’s incentives within marriage?
What about economic behaviors within a marriage?
8) Why does Smock use the log of income in the OLS analysis?
9) Smock explicitly focuses on the short-term economic consequences of marital disruption. Does this focus overestimate women’s economic disadvantage after divorce? Can we assume that the short- and long-term economic implications of martial disruption are similar, i.e. that the initial disadvantage persists indefinitely?
10) Smock excludes remarriage and post-divorce cohabitation from analysis. Would it be useful to compare the financial situation of women who remarry or cohabit after divorce with those who do not? Which insights could be provided by such comparisons?
11) Smock tests gender interactions in the effect of the number of children on post-divorce income. Given pronounced race differences in income, the number of children, and divorce rates, do you think it’s important to study the joint effects of race and gender by testing three-way interactions?
12) Smock uses personal income and per capita income as outcomes. Which other measures of economic standing can be used as outcomes in research on the economic consequences of marital disruption? Why?
13) How might personal income and the patterns of labor force participation of women who are married differ from those of cohabiting women? Are cohabiting women with children less disadvantaged than their married peers after the disruption of the union?
14) In studying women’s economic circumstances after divorce, researchers typically compare women’s and men’s post-divorce income, or women’s post-divorce income to women’s re-divorce income. What about comparing women’s post-divorce income to their premarital income? Would such comparisons help researchers address the issue of spuriousness in the association between divorce and economic disadvantage?
15) Men with children are more likely to remarry than women with children. How might post-divorce income of men with children change upon remarriage? Is it likely to increase or to decline? Why?
16) Smock concludes that “the conflict between parenting and labor force attachment… is one borne almost invariably by women” (p. 259). Does this pattern differ by SES? Women with higher levels of education are more likely to delay childrearing in order to establish their carriers, and better-educated men are more likely to participate in childcare.
17) Both Smock and Goldstein take into account marital duration prior to dissolution. Yet, if a marriage was preceded by cohabitation, the measure of marital duration may underestimate the true length of a union. To what extent not accounting for the duration of premarital cohabitation might have influenced the findings of Smock and Goldstein (if at all)?
18) Goldstein focuses on female martial histories. Is it because his analysis is based on the assumption that changes in women’s individual characteristics and behaviors are primarily responsible for trends in divorce? Would Goldstein’s conclusions be the same if his analysis were based on a male sample only?
19) Which social factors and processes (other than the explanations refuted by Goldstein) might account for the observed leveling of divorce rates?
20) In terms of the psychological impact of parental divorce on children, does it matter how old a child was when his or her parents divorced? Would it be useful to test divorce * age of a child interactions to tease out the effects of a child’s developmental stage? On a related note, to which extent reasons for, or causes of, divorce matter for children’s post-divorce well-being?
21) Sigle-Rushton et al. restrict their sample to children who were living with both biological parents at the time of the first follow-up. Do children who live in cohabiting unions or stepfamilies experience consequences of marital break-up comparable to those experienced by children living with two married biological parents?
22) Sigle-Rushton et al.’s main outcome variable is based on parental reports of their children’s temperament. Is it a good measure? How can it be improved?