Page 111 - IJPS-3-1

Zhang Z and Lin I-F

supported, with a few exceptions. The flow of upward transfers from adult children to
a widowed parent is most likely to occur when the parent is in need of support (either
financially or physically) and has a large number of children regardless of child’s
gender. The number of children who live close by is positively associated with the
odds of receiving some types of support including sick care and emotional support.
The transfer, however, is less likely to occur when the widowed parent had a complex
marital history. Our findings also suggest that the most vulnerable group, the widowed
with ADL disabilities, may face particular challenges in sick care and emotional
support as their disability worsens over time.
How are these results compared to those of the United States? There are some
key similarities as well as differences. Widowed parents’ needs in ADL assistance
are associated with coresidence in both countries (Liang, Brown, Krause et al.,
2005; Seltzer and Friedman, 2014). Nevertheless, while widowed older adults’
socioeconomic status and the number of adult children are positively associated with
coresidence in China, these factors are not associated with coresidence in the U.S.
Instead, having at least one daughter is significantly related to coresidence for widows
in the U.S. In addition, the rates of coresidence in widowhood vary widely in the two
countries: less than 10% of widows lived with their children in the U.S. (Seltzer and
Friedman, 2014) compared to 67% in China. This is partly due to different cultural
traditions and economic contexts in the two societies. As our measures of financial
support, sick care, and emotional support are not directly comparable to those used in
the U.S. studies, it is difficult to compare these results between the two countries.
Our study also confirms previous findings that sons (and daughters-in-law) still
play a very important role in all types of old-age support for widowed older adults,
due to the filial norms and the high prevalence of coresidence. In addition, we found
that daughters also participate in providing financial support, sick care, and emotional
support to their widowed parents. Grandchildren play a much smaller role in providing
care and support than sons, daughters, and their spouses. This key finding regarding
the role of sons and daughters-in-law is also different from the findings in the U.S.
where daughters provide more support than sons to older parents (Grigoryeva, 2017;
Laditka and Laditka, 2000; Lee, Jeffrey, and Coward, 1993).
Finally, we found that widowed older adults with ADL limitations are more likely
to receive care from children or their spouses of the same gender, with the exception
of sons-in-law. Specifically, whereas sons are the major caregivers for widowers,
daughters-in-law and daughters bear the major responsibility of caring for widows.
This finding is largely consistent with Lee et al.’s (1993) finding in the U.S. that
a same-gender preference exists in intergenerational caregiving relationships. The
notable difference is that daughters-in-law play a far more important role in caring for
disabled widows in China than in the United States.
Our study has some limitations that suggest avenues for future research. First,
because the survey does not ask respondents detailed information about each child in
the family, such as their marital status, educational attainment, occupation, and health,
we were unable to take into account each child’s abilities to assist their older parents
in the current study. Second, our study is based on cross-sectional data and reflects a
snapshot of intergenerational support in widowhood. Longitudinal and dyadic data will
provide more insights into the trajectories and determinants of caregiving to widowed
older adults over time. Third, our measures of intergenerational support are based on
reports from widowed adults; a recent study in the United States showed that parents
and adult children had a moderate level of agreement on intergenerational transfers,
and it is important to control for reporting bias in future studies when both parents’ and
adult children’s reports of same transfers are available (Lin and Wu, 2017). Fourth,
although we examined multiple dimensions of support, there are a few important
intergenerational transfers that we did not examine due to the lack of data. These
transfers include household help, agricultural fieldwork help, frequency of contacts,
etc. Lastly, we did not examine the quality of care provided by adult children. Recent
research in China showed that disabled older adults are more satisfied with the care

International Journal of Population Studies 2017, Volume 3, Issue 1 105