Aquilino, William S. 1994. "Impact of Childhood Family Disruption on
Young Adults' Relationship with Parents." Journal of Marriage and the Family
This paper is an extensive exploration of childhood living arrangements and
their possible impacts on young adult men and women's relationships with mothers
and fathers. Aquilino tests the following hypotheses:
1. There is no difference between adults' relationships with single custodial
mothers and mothers of intact families. (support found with the exception of
the relationship quality rating which was negative)
2. Relationships between non-custodial mothers and adult children will be weaker
than those with mothers of intact families, but stronger than adult parent child
relationships with non-custodial fathers. (partial support found, i.e., geographic
3. There is no difference between adults' relationships with single custodial
fathers and mothers of intact families. (support found)
4. Non-custodial fathers will have lower quality relationships with adult children
than fathers of intact families. (support found)
5. Negative relationships effect between adults and non-custodial fathers will
be stronger for men than women. (no difference found)
6. Adults who grew up in a single mother family from birth will have worse
relationships with fathers than those whose fathers divorced their mothers.
(no difference found)
7. Mother-custody households will produce stronger relationships between mothers
and daughters than mothers and sons in adulthood, (no support found)
8. and father-custody households will produce stronger relationships between
fathers and sons than fathers and daughters in adulthood. (no support found)
9. Remarriage of a custodial parent has a negative impact on the relationship
with the non-custodial parent compared to intact and single-parent families.
(partial support for non-custodial fathers, strong support for non-custodial
10. Remarriage of a custodial parent will produce stronger negative effects
on adult children's relationships with non-custodial fathers than non-custodial
mothers (little support found; a negative relationship was observed between
frequency of contact and custodial mother's remarriage)
11. There is a positive relationship between child's age at divorce of his/her
parents and the relationship quality with the non-custodial parent. (support
12. There is a negative relationship between the number of family transitions during childhood and the relationship quality with parents and adult children (no support for relationships with mothers, weak support for relationships with fathers).
Aquilino used a sample of men and women ages 19-34 from the NSFH I. He used
relationship quality ratings, geographic distance, frequency of contact, and
the giving and receiving of support (services, emotional, financial) to measure
relationship quality between these adults and their parents.
Overall, this paper supports the idea that childhood living arrangements can
significantly impact intergenterational relations.
Bumpass, Larry. 1994. "A Comparative Analysis of Coresidence
and Contact with Parents in Japan and the United States." Pp.221-246 in
Tradition and Change in the Asian Family, edited by L.J. Cho and M. Yada.
Honolulu, HI: University of Hawaii Press.
This paper documents basic facts of coresidence and contact with
parents in Japan and the United States. Although the author begins with
the widely cited cultural difference of collectivist versus
individualistic orientations between Japan and the United States to
motivate his empirical inquiry, the analyses reported in this article
essentially follow the bottom-up strategy rather than the top-down
strategy. Personally, however, I would like to see more theoretical
speculation linked to these empirical findings.
There are two sections in this paper: one deals with coresidence
and the other with contact with parents. The data come from the 1987-88
U.S. National Survey of Families and Households and the 1988 National
Family Survey of Japan. Bumpass's thoughtful attention to the demographic
profiles (including selection effect) and data quality while explaining
the substantive results is particularly worth of our learning.
With respect to coresidence, the analysis confirms the well-known
difference in the nature of kin relations. The American kinship is
bilateral, meaning a similar coresidence pattern with both husband's and
wife's parent, while the Japanese is patrilocal with only 15 percent of
coresident couples living with wife's parents. Nonetheless, Bumpass
objects to the conventional cultural explanation for this difference that
filial obligation and coresidence to take care of parents only exist in
the Confucian Japanese society and not in the individualistic American
society. His argument is fleshed out in Table 12.3 by providing a
illustration of the coresidence patterns for adult children between age
55-64 in the United States. Table 12.4 examines the factors influencing
current parent-child coresidence in Japan-a replication of the excellent
Martin and Tsuya (1991) paper upon which Bumpass's analysis builds.
In regard to the contact frequency, the most intriguing finding to
me is the life-course differential between Japanese and Americans, as
demonstrated in Figure 12.1. In Japan, parent-child interaction declines
with age, and this decline cannot be statistically explained by parents
being taken care of by another sibling. In the United States, the
declines are less strong for men and curvilinear for women. Besides,
distance is the strongest predictor of interaction. However, this
relationship may be endogenous, namely living farther apart is because of
the child's intention to reduce interaction with parents or vise versa.
Ogawa, N., & Retherford, R.D.(1993). "Care of the Elderly in Japan: Changing Norms and Expectations". Journal of Marriage and the Family, V.55, pp.585-597.
Japan has experienced rapid socioeconomic development and demographic changes
over the past decades. Along with industrialization and modernization, a demographic
transition from high birth and death rates to low birth and death rates has
resulted in rapid population aging.
In this study, the authors analyze the trends and determinants of changes in norms of filial care for elderly parents and expectations of old-age support from children in Japan under the context of rapid socioeconomic and demographic changes. They distinguish between norm and expectation by defining "norms" as socially accepted standards about how people should behave and "expectations" as something less constrained by what other people think. According to the author, Japan is a highly integrated society that is homogeneous in such characteristics as language, ethnicity, and religion. In such a highly integrated society, the diffusion process often proceeds swiftly once a normative shift begins because the population tends to be function normatively as a unit. In this light, they hypothesize that norms of filial care for elderly parents remain relatively stable for a time, then weaken rapidly across the population, and then stabilize at a new plateau. On the other hand, they hypothesize that expectations of old-age support from children adjust more gradually and continuously to changes in underlying socioeconomic and demographic conditions because expectations are less sensitive to normative constraints.
Their analyses are based on data from the various rounds of the National Survey
on Family Planning, and the respondents are currently married women below age
50. First of all, the analysis of norms of filial care of elderly parents is
based on a question, "What is your opinion about children caring for their
elderly parents?". The response categories of this question are good custom,
natural duty, unavoidable, not a good custom, and others. The analyses of distribution
of responses over time in the various surveys taken between 1963 and 1990 indicate
a two-stage normative shift: first a gradual shift from "good custom"
to "natural duty", and second a sudden shift from "natural duty"
to "unavoidable" and "not a good custom". Norms remained
highly traditional between 1963 and 1986, and that a major shift from traditional
to modern occurred after 1986. Up to 1986, the changes in norms of filial care
for elderly parents have lagged behind changes in underlying socioeconomic and
demographic conditions, and then, the normative shift appears to have been precipitated
Second, the analysis of expectations of old-age support from children is based
on a question, "Are you planning to depend on your children in your old
age?". The response categories of this question are "expect",
"never thought about it", and "do not expect". The findings
show a different pattern from the analysis of norms. Expectations of old-age
support have declined steadily over time and show little evidence of lags or
Finally, they use logit regression to examine determinants of norms and expectations and include such predictor variables as wife's age, wife's education, husband's occupation, urban-rural residence, whether or not the woman is currently living with her own or her husband's parents (omitted in the analysis of norms), and annual income. For norms of elderly care, the effects of predictor variables are small cross-sectionally, and the percentages who view filial care favorably declined sharply after 1986 in all categories of all predictor variables. On the other hand, for expectations of old-age support, the predictor variables have fairly substantial cross-sectional effects, but the adjusted category-specific percentages who expect old-age support form their children change little over time, indicating that the substantial decline in the overall observed percentage who expect support is due almost entirely to compositional changes. Thus, they conclude that expectations, in contrast to norms, tend to adapt continuously to changes in underlying socioeconomic and demographic conditions, as they hypothesized.
Ogawa, Naohiro and Robert D. Retherford. 1997. "Shifting Costs of Caring for the Elderly Back to Families in Japan: Will it Work?" Population and Development Review 23:59-94.
The Japanese social security system includes old-age pension, medical plans, unemployment compensation and other smaller programs.
The Employees' Pension Scheme (EPS) and the National Pension Scheme (NPS) cover about 90% of the workforce. EPS and NPS were established in 1941 and 1961, respectively. In 1986 major pension reform was implemented by integrating the different pension schemes into NPS. Another change in pension schemes was that they were based on reserve financing when they were established, but are now almost entirely based on pay-as-you-go financing. This is not unusual and has been implemented in other countries, and happens when the benefits exceed the contributions. In the pay-as-you-go scheme, generations do not finance their own future, but pay for the pension of the current elderly population. This means that the benefit-contribution ratio of more recent cohorts is smaller, and that they will receive fewer benefits for more contributions than earlier cohorts.
The rising cost of the pension and medical plans have led the government to look for ways to shift some of the burden of caring for the elderly to the elderly themselves and their families. In 1990 a ten-year project "Golden Plan" was launched to improve social services for the elderly and their families. Golden Plan includes improvements in institutional care, expansion of nursing home capacity, and improvements in services for elderly who live at home (home-helpers, short-term stay facilities, and day-care centers). Support services for families taking care of elderly parents are expected to improve the quality of care as well as reduce the burden on the caretaker.
There have been considerable changes in the values about caring for the elderly. For example, between 1963 and 1996 the proportion of currently married women of reproductive age who responded either "good custom" or "natural duty" to the question "What is your opinion about children caring for elderly parents?" fell from 80 to 47 percent. Even though the values have changed, children are legally responsible for taking care of their elderly parents, and under some circumstances even more distant relatives are required to provide support. This law is enforced to some extent. There were over 100 court cases in 1995 involving the inadequate provision of support for elderly persons.
The authors suggest that the governments' attempt to shift responsibility for the care of the elderly to families will not be successful because of changing values, increases in women's education and employment, and declines in coresidence.
Spitze, Glenna and John R. Logan. 1991. "Sibling Structure and Intergenerational Relations." Journal of Marriage and the Family 53:871-884.
In this paper, Spitze and Logan asked how the behaviors of individuals, and relationships between parents and individual children were affected by the sibling structure of the family. It might be an every-day life experience that parent-child relationship was affected by the sibling structure of the family. However, there had been no systematic research in this area, and the results of this paper were not clear-cut either (or "frustrating" as the authors put it). The authors called for the collection of more suitable data with detailed information on family structure.
They used data from a local probability sample from a personal interview survey
conducted in the metropolitan area of Albany-Schenectady-Troy, New York from
September 1988 to February 1989. Their analyses consisted of two parts: one
from the perspective of the adult children, and the other from the perspective
of the parent.
There were three dimensions of the sibling structure: number of siblings, gender of siblings and birth order. Spitze and Logan used three alternative measures of sibling structure: 1) number of siblings (not including respondents); 2) number of respondents' sisters and number of brothers; 3) four dummies representing the number and gender of siblings (only children, one sister, one brother, and two or more siblings). And they had a measure and data for birth order in the second part of their analyses.
The first part of their analysis focused on respondents with living but non-coresident biological parents. First, they asked if respondents' filial attitudes or feelings of closeness were affected by the sibling structure. They found no effects of any measure of sibling structure on the attitude and closeness measures. Then they measured parent-child relationship by frequency of visiting, and phoning, and weekly hours of help, and found:
· Women reported about 50% more visits, phone calls and hours of help
to parents than men.
· There was a consistent negative effect of the number of siblings on parent-child contact.
· There was a significant interaction between gender and number of siblings for helping and visiting, but not for phoning. Women were more negatively influenced by number of other siblings in their helping and visiting behavior.
· The number of brothers significantly decreased visits and phone calls, and this effect did not differ by gender, while the number of sisters significantly decreased visits only.
· Only children visited and helped more, but the effects of measures capturing more details of the sibling structure were difficult to summarize.
The second part of their analysis looked into the reports of elderly respondents with non-resident children aged 40 and above. In this analysis, each respondent-child pair was a record in the data file and there was information on birth order.
· As in the first part, number of siblings had a significant negative
effect on visits, phone calls and help. But unlike in the first part, these
effects did not differ by gender.
· Also unlike the first part, number of sisters negatively affected visits, phones and help, but the number of brothers made no difference.
For the discrepancy between the parts of the analysis, Spitze and Logan ventured two explanations. One was the bias of the sample, which excluded institutionalized persons and those too frail to interview. Another possibility was that parents tended to view relations to children being more similar than they actually are.
Spitze, Glenna, John R. Logan, Glenn Deane, and Suzanne Zerger. 1994. "Adult Children's Divorce and Intergenerational Relationships." Journal of Marriage and the Family 56:279-293.
The rising divorce rate in the United States has led to fears among some that those who suffer from divorce will have fewer resources to aid the growing elderly population. This study evaluates the effects of divorce on the relationship between an adult child and his/her parents. The authors look at measures of feelings of closeness, patterns of contact, and instrumental help both to and from children, for both daughters and sons.
The data are from a personal interview survey that was conducted in the Albany-Schenectady-Troy, New York metropolitan area in the six months between September 1988 and February 1989. The sample was taken from 120 block groups with probabilities proportionate to size. Interviewers randomly selected one adult age 40 or over from all eligible adults present in the household. Each respondent was asked detailed questions about relationships with all parents and children, including stepchildren. Respondents were asked how often they visit with, talk on the telephone to, and how close is his/her relationship with each child. They were asked who performs each of the following tasks: cooking, cleaning, shopping, laundry, home repair, health care, and transportation and the amount of time spent on each task. Finally, they were asked whether they ever help specified other (including any adult children) and if so how often.
The analysis includes information on respondents with one or more married children. It is in two parts: descriptive and an analysis of the effects of a child's divorce. The descriptive analysis uses the respondent's (parent's) record as the unit of analysis. The analysis of the effects of a child's divorce uses each parent-child pair as the unit of analysis. Analysis was conducted separately for male and female children. Because the observations are not independent, the Liang and Zeger method is used. Tobit models were used to estimate the probability that help was received/given and the amount conditional on help being received/given.
The analysis includes the adult child's marital status (first marriage; remarriage; divorced/separated with custody of a child; and divorce/separated without custody of a child, never married, and currently widowed), employment status, and the number of children under 12 years of age, parent's age, marital status (married; unmarried), health, education, income, employment status, and gender, and geographical distance between parent and child, number of children, and history of custody (ever did not have custody of adult child).
The results show that about half of all parents over 60 have at least one child who is separated/divorced. There is some evidence that there is more contact between divorced daughters with child custody and their parents than with married daughters. Divorced daughters receive more babysitting help than other daughters, but they help parents as much as married daughters. Never married sons help parents most, while divorced and widowed sons help parents the least. The presence of grandchildren appears to increase parent-child contact and "reinforce reciprocal patterns of helping" (291). The findings suggest that a daughter's divorce does not weaken helping relationships, though a son' divorce may have a slight negative effect.
Wolf, D. A. and Beth J. Soldo (1994). "Married Women's Allocation of Time to Employment and Care of Elderly Parents". The Journal of Human Resources. Vol. 29, issue 4. Special Issue: The Family and Intergenerational Relations.
Wolf and Soldo use the NSFH (National Survey of Families and Households) data to analyze parental care and employment. The authors recognize the potential simultaneity of decisions regarding employment and parental care. In order to deal with this problem, the authors use simultaneous equations model to generate the estimates. The dependent variable in the multivariate analysis is the number of "usual" hours worked in the job held during the last week. The authors also explore the actual number of hours worked in the last week in the bivariate analysis. The exogenous variables are: age, age squared, race, education, income, homeownership, ADL, dummies indicating parents and parents in law that are sick and those that are healthy. The analysis is restricted to married women with at least one parent or parent-in-law or step-parent aged 65 or older.
The main results are:
1) Caregivers are less likely to hold a job than non-caregivers;
2) They are also less likely to have actually worked during the last week;
3) However, employed caregivers work no fewer hours than those not providing parent care;
4) Having a parent in good health reduces the effective claim on a married women to take care of an older parent in poor health;
5) Women highly likely to be employed, on the basis of unobserved traits, are also relatively unlikely to provide parent care.
It is important to mention that NSFH do not collect direct data if respondents
were actual caregivers. The study used an indicator of potential risk status
based on the respondent's judgment of parental health ('very poor', 'poor',
'fair', 'good' and 'excellent'). Therefore, results should be taken cautiously
because actual provided care was not considered in the analysis.
Finally, it is important to mention that adjustments were necessary in the
hours-of-work equation and this reduced the sample size in that equation.