Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Sep 21.
Published in final edited form as: J Fam Issues. 2012 Apr 27;33(9):1143–1167. doi: 10.1177/0192513X12438685

Grandparents Providing Care to Grandchildren: A Population-Based Study of Continuity and Change

Ye Luo 1, Tracey A LaPierre 2, Mary Elizabeth Hughes 3, Linda J Waite 4
PMCID: PMC10512842  NIHMSID: NIHMS1928960  PMID: 37736111

Abstract

This study examines transitions in grandchild care and the characteristics of grandparents making these transitions, using longitudinal data from a nationally representative sample of 13,626 grandparents in the 1998–2008 Health and Retirement Study. More than 60% of grandparents provided grandchild care over the 10-year period; more than 70% of those did it for 2 years or more. Grandparents with fewer functional limitations and more economic resources were more likely to start or continue nonresidential care, whereas relatively disadvantaged grandparents were more likely to start and continue coresidential care. Grandparents who were African American, younger, married, living with fewer minor children of their own, or had more grandchildren were more likely to start care, particularly nonresidential care. African Americans and Hispanics were more likely than Whites to start and continue coresidential care. These findings demonstrate the heterogeneity of caregiving and point to the lack of resources among those who provide coresidential care.

Keywords: grandparents, grandchildren, caregiving, living arrangements, intergenerational relationship

Social policies in the United States typically assume that the family will provide a safety net for family members experiencing difficulties. Perhaps in response to the lack of public alternatives and contrary to the popular notions of declining family ties in America, family members in large measure do provide care for one another, offering practical and emotional support in response to both everyday problems and major misfortunes (Eggebeen & Hogan, 1990). Longer life expectancy means that many older adults now survive to see their grandchildren grow up and many children grow up with living grandparents (Bengtson, 2001). At the same time, social and economic changes have increased the number of parents who need help raising their children. This combination has led to an increase in both the incidence and prevalence of grandparent caregiving (Pebley & Rudkin, 1999; Szinovacz, 1998).

Grandchild care can take several forms. Some grandparents take sole responsibility for raising their grandchild when the grandchild’s parents are incapacitated by substance abuse, mental or physical illness, incarceration, death, or another reason. Parents are not present in these “skipped-generation households” (Goodman & Silverstein, 2002). In other cases divorce, financial need, work demands, or school commitments compel parents to seek help from their parents in raising the next generation. Generally, these adult children move back in with their own parents, who take on supplemental or coparenting responsibilities in “multigeneration households” (Goodman & Silverstein, 2002; Waldrop & Weber, 2001). Other grandparents provide child care or babysitting to grandchildren who do not live with them (Presser, 2004).

In 2000, almost 6 million grandparents lived with their grandchildren; 42% of these grandparents had primary responsibility for these minor coresident grandchildren (Simmons & Dye, 2003). These numbers continue to rise. From 2000 to 2008, the number of grandparents providing primary care to their grandchildren increased by 8%, 5% of which occurred from 2007 to 2008, the start of the great recession (Livingston & Parker, 2010). Some grandparents make a permanent commitment to caregiving whereas in other families the arrangement is only meant to be temporary. The majority (54%) of grandparents reporting primary care for a coresident grandchild have been providing this care for 3 or more years (Simmons & Dye, 2003).

We argue that transitions into and out of caregiving matter because they tell us about caregiving in the life course of the grandparent and tell us about the stability and instability of caregiving, which are shown to have important consequences. A growing body of literature has demonstrated the impact of continuity and change in grandparent caregiving on a number of grandparent outcomes including changes in personal freedom and privacy, intrafamily strain, stress, social and instrumental support, family functioning, preventive health behaviors, and mental and physical health (Baker & Silverstein, 2008; Blustein, Chan, & Guanais, 2004; Hughes, Waite, LaPierre, & Luo, 2007; Lumpkin, 2008; Musil et al., 2011; Szinovacz & Davey, 2006). Although it is clear that continuity and change have important effects, we know less about the rates of transitions into and out of caregiving and the predictors of these transitions.

Many studies report significant differences between grandparent caregivers and noncaregiving grandparents (e.g., Fuller-Thomson & Minkler, 2001; Fuller-Thomson, Minkler, & Driver, 1997; Musil et al., 2011). However, these comparisons are often cross-sectional, making it impossible to determine whether differences on attained characteristics occurred before or after a transition, and results could be biased if they are likely to catch only long-term caregivers. The only longitudinal study to date used the first two waves of the National Survey of Families and Households to identify characteristics that are predictive of individuals becoming primary caregivers for their grandchildren in skipped-generation households (Minkler & Fuller-Thomson, 2000). We thus know little about the characteristics of those most likely to start living in multigeneration households and even less about grandparents who provide regular care to grandchildren who live elsewhere. To our knowledge, no empirical research has longitudinally examined predictors of transitions out of grandchild care.

Using a national longitudinal survey, we examine the dynamics of grandchild care, focusing on three types of care: babysitting, living in a multigenerational household with grandchildren, and living in a skipped-generation household with grandchildren. Our analysis is framed around three sets of questions: (a) What are the dynamics of caregiving provided by grandparents to their grandchildren? (b) What is the probability over a 2-year period that a grandparent will initiate various types of grandchild care? That a grandparent will end each type of care? That a grandparent will switch care types? (c) What factors predict changes in grandchild care? Do these factors vary by type of care?

Literature Review and Hypotheses

Grandparent caregiving is a central component of family support in the broader context of intergenerational relationships. Life course scholars contend that grandparent–grandchild relationships are partially contingent on grandparents’, parents’, and grandchildren’s roles in other life spheres (e.g., marriage, parenthood, labor force participation) and on the sequencing of life transitions in each of these domains (Burton & Bengston, 1985; Hagestad & Burton, 1986; Szinovacz, 1998). The availability and willingness of grandparents as well as the needs and preferences of parents and their children change over the life course and are the primary determinants of the extent of caregiving provided by grandparents (Hagestad & Burton, 1986; Hank & Buber, 2008). Other scholars see grandparent role enactment as a social construct that varies across personal and historical time, as well as across cultural and regional contexts. From this perspective, cultural norms that emphasize or downplay the role of grandparents affect the type and level of grandparent involvement (Silverstein, Giarrusso, & Bengtson, 2003; Silverstein & Marenco, 2001). To these perspectives, we add the Resources and Demands Theoretical Framework developed by Waite and Hughes (1999; see also Hughes & Waite, 2002). In this perspective, individuals experience role-based household relationships as sets of resources and demands, including time, money, skills, and attention. We extend this to family relationships more generally, in this case grandparent, adult child, and grandchild relationships. Grandparents have variable sets of resources and face variable sets of demands from family members. These affect the care they are able and willing to provide and the care that grandchildren need. These perspectives suggest a number of factors that might influence the likelihood that a grandparent will provide grandchild care, including demographic characteristics, other role obligations and family demands, socioeconomic resources, and health conditions.

Demographic characteristics.

Differences in kin relationships among those in racial and ethnic subgroups have long been apparent to researchers (Burton & Bengtson, 1985; Fuller-Thomson & Minkler, 2000; Minkler & Fuller-Thomson, 2005). Both a positive cultural tradition emphasizing the grandparents’ role as guardians and caregivers across the generations and current contextual problems, such as maternal incarceration, AIDS, and substance abuse, suggest a considerably higher likelihood that African American grandparents become surrogate parents to their grandchildren (Minkler & Fuller-Thomson, 2005). Empirical studies suggest this is the case (Fuller-Thomson & Minkler, 2001; Minkler & Fuller-Thomson, 2000; Szinovacz, 1998). Although Hispanic children are less likely to be living in grandparent-headed families than African American children, a larger proportion of Hispanic grandparent-headed families are multigenerational households than White or African American grandparent-headed families (Fields, 2003), reflecting the high value placed on intergenerational living in Hispanic culture (Tienda & Angel, 1982). White families have the lowest proportion of children living in grandparent-headed families (Fields, 2003). Thus, we hypothesize that African American grandparents are more likely than both Hispanic and White grandparents to move into and remain in skipped-generation households and that Hispanics are more likely than African Americans and Whites to move into and remain in multigeneration households. Among non-coresident grandparent caregivers, however, the small amount of empirical evidence available suggests that African Americans or Hispanics are about as likely to provide babysitting to their grandchildren as Whites (Cherlin & Furstenberg, 1986; Fuller-Thomson & Minkler, 2001).

The importance of gender in kinship relationships has been stressed by many scholars (Rossi, 1995; Uhlenberg & Hammill, 1998). The pronounced sexual division of household labor also suggests grandmothers’ greater share of caregiving responsibilities of all types (Brody & Saperstein, 2004). Empirical studies of coresident grandchild care consistently show disproportionate representation of grandmothers as primary caregivers and as extensive secondary caregivers to their grandchildren (Fuller-Thomson & Minkler, 2001; Minkler & Fuller-Thomson, 2000). We expect that grandmothers are more likely than grandfathers to start and continue all types of grandchild care.

Most survey research found that younger grandparents are more likely to take on caregiving responsibilities than their older counterparts (Fuller-Thomson & Minkler, 2001; Minkler & Fuller-Thomson, 2000). This finding prompted some researchers to suggest that caregiving may be more intrusive in the lives of older grandparents because they are more likely to be caring for their own elderly parents and/or dealing with personal health problems and limitations (Minkler & Fuller-Thomson, 2000). However, qualitative studies show that younger grandparents experience high levels of stress over the role conflicts that arise when they care for grandchildren because they are more likely to be in the paid labor force and/or to have their own youngest children still at home (Burton & Bengtson, 1985). All else equal, young grandparents have adult children who became parents at early ages, when they are most likely to need help.

Other role obligations and family demands.

The ability to take on the care of a grandchild may be contingent on the other roles the grandparents hold, the demands these roles make, and the resources they provide. We focus on four key roles: grandparent, spouse, parent, and worker. The more grandchildren an older adult has, the more opportunities they have to become a caregiver, although the chance of caring for any particular grandchild may decrease (Uhlenberg & Hammill, 1998). Married grandparents not only have resources available in the time and energy of their spouse that could facilitate caregiving but also have demands on their own time and attention from that same spouse; the net effect on caregiving for grandchildren is indeterminate (Fuller-Thomson & Minkler, 2001; Uhlenberg & Hammill, 1998). Grandparents who have their own minor children in the household may face limits on their ability to provide care to their grandchildren since their children demand their time and attention. Research in Europe has found that employed grandparents are less likely to provide regular care than nonemployed grandparents (Hank & Buber, 2008). Employed grandparents have more money but less time, all else equal, suggesting that they might subsidize their grandchildren rather than provide care directly. We expect that grandparents are more likely to become a caregiver and remain a caregiver across all categories of care if (a) they have more grandchildren, (b) they have fewer minor children of their own, and (c) they are not in the paid labor force. Being married both increases and decreases the likelihood of providing care to grandchildren, so we make no hypotheses about this role.

Socioeconomic resources.

Grandparents with higher socioeconomic status have more resources that allow them to keep close contact with their grandchildren without coresiding (Uhlenberg & Hammill, 1998). High income also allows grandparents to provide financial support to their children for the costs of child care. Grandparents who are poor, however, do not have the same financial resources to assist with payment for formal child care. Coresidence allows poor families to capture the economies of scale from joint living, albeit at the expense of privacy (Tienda & Angel, 1982). Prior research consistently shows a higher likelihood of coresidence with grandchildren among less educated and low-income grandparents (Blustein et al., 2004; Fuller-Thomson et al., 1997), and grandparents with less than high school education have a higher probability than others of becoming a primary caregiver for their grandchildren (Minkler & Fuller-Thomson, 2000) and of providing extensive but noncustodial care (Fuller-Thomson & Minkler, 2001). These disparities may reflect fewer options for both child care and living arrangements among low-income families. We hypothesize that grandparents with higher socioeconomic status are less likely than those with lower socioeconomic status to begin babysitting and to continue to provide it. We also expect that grandparents with lower socioeconomic status are more likely to start and continue coresidential care than those with higher socioeconomic status.

Health conditions.

Although a number of studies assess the health impact of providing care to grandchildren, few take into consideration the role that health may have played in determining who becomes and remains a caregiver. On one hand, people who are willing and able to become caregivers are likely to be healthier than others of their age. In fact, prior research has shown that the likelihood of assistance to adult children is positively associated with individual health (Henretta, Grundy, & Harris, 2002). On the other hand, caring for grandchildren often begins in response to a crisis in the adult child’s life such as job loss, divorce, serious illness, addiction, or incarceration (Dellmann-Jenkins, Blankemeyer, & Olesh, 2002; Pinson-Millburn, Fabian, Schlossberg, & Pyle 1996). Poorer health and the need to care for a grandchild may both be outcomes of a disadvantaged life course (Hughes et al., 2007; Strawbridge, Wallhagen, Shema, & Kaplan, 1997). Although grandparent caregivers have been shown to have poorer health than other grandparents (Hank & Buber, 2008; Fuller-Thomson & Minkler, 2000), the extent to which poor health influences transitions into and out of caregiving is unclear.

In sum, our study is among the few that look at transitions in caregiving provided by grandparents to grandchildren over time. We identify a number of factors that may affect these transitions based on life course, social construction, and resources and demands perspectives. We hypothesize that grandparents are more likely to move into and remain in all types of care if they are female, have more grandchildren, have fewer minor children of their own, are not in the paid labor force, have fewer socioeconomic resources, and have better health. We hypothesize that African American grandparents are more likely than both Hispanic and White grandparents to move into and remain in skipped-generation households and that Hispanics are more likely than African Americans and Whites to move into and remain in multigeneration households.

Method

Data

Our data come from Waves 4 to 9 (1998–2008) of the Health and Retirement Study (HRS), a nationally representative, longitudinal study of people over age 50. The HRS consists of five birth cohorts who entered the study in different calendar years and were interviewed every 2 years thereafter. Spouses of age-eligible respondents were interviewed regardless of their age. The sample for each cohort was derived from the same stratified, multistage area probability design in which African Americans, Hispanics, and Floridians were oversampled. The initial cohort response rates ranged from 70% to slightly more than 80%; reinterview rates for all cohorts at each wave have been between 92% and 95% (HRS, 2011). The HRS now includes more than 30,000 respondents. We use data from age-eligible members of the Oldest Old Cohort (born 1923 or earlier), the Children of the Depression Cohort (born 1924–1930), the Original Cohort (born 1931–1941), and the War Baby Cohort (born 1942–1947). These cohorts form a nationally representative sample of the U.S. population over age 50 in 1998, the first year in which they were all interviewed.

Our analytic sample is composed of 15,663 noninstitutionalized, non-Hispanic White, African American, and Hispanic grandparents born between 1918 and 1947. Of these, 2,037 (13%) provided no data about grandchild care; thus, our final sample size is 13,626. The large sample size is a key strength of our study because relatively few grandparents live with grandchildren (Pebley & Rudkin, 1999).

Dependent Variable: Care for Grandchildren

In each wave of the HRS, respondents were asked whether they spent 100 or more hours taking care of grandchildren over the previous 2 years. Respondents answering “yes” were asked how many hours they spent on grandchild care. Respondents also listed the people living in their household and their relationship to each person. Using this information, we identified grandchild care status for each respondent at each interview. We distinguished three kinds of care: personally caring for at least one grandchild who did not live in the same household for 100 or more hours in the last 2 years, approximately 50 hours or more per year (babysitting); living with at least one adult child and grandchild(ren) (multigeneration household); and living with a grandchild(ren) with no adult child present (skipped-generation household). We also categorized grandparent babysitters by hours of care per year: 50 to 99 hours, 100 to 199 hours, 200 to 499 hours, and 500 or more hours.

Independent Variables

All independent variables are measured at the beginning of each 2-year interval. The distributions of these variables in 1998 are shown by race/ethnicity in Table 1.

Table 1.

Measures of Care for Grandchildren and Covariates in Baseline Year (1998)

Characteristic All (N = 13,626), Mean (SD)/% White (N = 10,616), Mean (SD)/% Black (N = 1,905), Mean (SD)/% Hispanic (N = 1,105), Mean (SD)/%
Grandchild care status
 Not a grandparent in I998a 8.8**c 9.3 4.3 7.8
 Grandparent in 1998 91.2 90.7 95.7 92.2
  Not providing careb 65.5 67.0 55.7 61.8
  Providing care
   50–99 hours a year 7.5 8.1 4.6 4.5
   100–199 hours a year 9.7 10.2 7.9 6.3
   200–499 hours a year 5.2 5.3 5.0 4.0
   500+ hours a year 5.3 5.1 6.9 5.5
  Multigeneration household 5.1 3.3 13.8 14.7
  Skipped-gene rati on household 1.6 0.9 6.0 3.3
 White 83.9
 Black 9.4
 Hispanic 6.7
 Female 56.2** 55.6 61.5 55.9
 Age 63.14 (8.43)** 63.45 (8.48) 61.7 (8.08) 61.22 (7.75)
 Married 72.2*** 75.1 50.0 67.0
 Number of own children <18 in household 0.07 (0.33)** 0.05 (0.29) 0.10 (0.41) 0.18 (0.56)
 Number of grandchildren 6.21 (6.44)** 5.74 (5.73) 8.92 (9.18) 8.28 (8.54)
 Working full-time 35.5* 35.5 35.4 36.1
 Working part-time 10.0 10.4 8.3 8.2
 Not working 54.5 54.2 56.3 55.7
 Years of education 12.28 (3.05)** 12.72 (2.62) 11.04 (3.35) 8.57 (4.42)
 Household income ($1,000) 55.88 (126.01)** 59.56 (94.78) 31.57 (36.62) 43.91 (348.68)
 Household net worth ($1,000) 339.48 (1211.57)** 387.76 (1313.61) 85.20 (172.11) 91.27 (272.42)
 Number of functional limitations 2.35 (2.83)** 2.25 (2.74) 3.02 (3.31) 2.69 (3.07)
 Functional limitations missing 16.3** 15.1 23.0 22.4
 Number of chronic conditions 0.98 (1.01)** 0.96 (1.01) 1.22 (1.05) 0.88 (0.95)
Open in a new tab

Note: Data are weighted to represent the U.S. noninstitutionalized older adults. Respondents not interviewed in 1998 are excluded.

a.

These respondents became grandparents after the 1998 interview and then reported care status in at least two consecutive interviews.

b.

“No care” includes grandparents who spent less than 50 hours a year personally caring for grandchildren.

c.

Indicates levels of significance of the chi-square or ANOVA tests of racial/ethnic differences:

*

p < .05,

**

p < .01.

Demographic variables.

These include race/ethnicity (non-Hispanic White, African American, or Hispanic), gender (female or male), and age measured in years.

Other role obligations and family demands.

We include indicators of whether the respondent is married/partnered, the number of own children under 18 in the household, the total number of grandchildren (both in and outside the household), and whether the respondent is working full-time, part-time, or not working. We also include dummy variables indicating either fewer or more grandchildren at the end of a 2-year interval than at the beginning.

Socioeconomic resources.

These include the respondent’s education measured in years of schooling and the logarithms of total household income and net worth.

Physical health.

We include two measures of physical health. Number of functional limitations is calculated by summing responses to 12 items assessing whether the respondent has difficulty with specific forms of ambulation, such as walking a block and climbing a flight of stairs, or muscle movements, such as moving a large chair or picking up a dime. Because a large number of respondents indicated they did not do certain activities, we used multiple imputation to assign values for these cases. A dummy variable indicates these cases. Number of chronic conditions is the total number of conditions reported out of these six: diabetes, heart disease, lung disease, cancer, hypertension, or a stroke.

Statistical Procedures

Because the HRS oversampled African Americans, Hispanics, and Floridians, all analyses were weighted. Our analyses included two components. First, we estimated the prevalence of the various types of grandchild care and the frequency of care transitions for the entire sample and for each race/ethnic group. We assessed care transitions by comparing care status at the beginning and end of each 2-year interval, using these intervals as the unit of observation. Precise estimation of the duration of caregiving spells is not possible because we do not know when caregiving began or ended within the 2-year interval. However, we provide a rough estimate of duration by counting the number of waves (1998 to 2008) at which the respondent reported he or she provided care to grandchildren.

Second, we estimated four multivariate regression models using the 2-year intervals as the unit of analysis. We separated respondents who did and did not provide grandchild care at the beginning of the interval. For grandparents who were not caregivers at the beginning of the interval, we estimated a binary logistic regression model predicting the start of any care over the next 2 years and a multinomial logistic regression model predicting starting each of the three types of grandchild care. For grandparents who were caregivers at the beginning of the interval, we estimated a binary logistic regression model predicting continuation of care over the next 2 years and a multinomial regression model predicting each type of care at the end of the interval. Because each respondent could contribute up to five 2-year periods of observation to the analysis, the observations are not independent. We used the Huber–White variance estimator to ensure that clustering in our sample (due to sample design and the use of multiple intervals per person) does not inflate test statistics (Greene, 1997). A dummy variable indicating the calendar year in which each interval began was also included in the models.

Results

Prevalence of Grandchild Care and Frequency of Care Transitions

Of the older adults who were grandparents in 1998, the first year we observe respondents, 66% of grandparents provided less than 50 hours of care a year for grandchildren over the preceding 2 years (Table 1). Twenty-eight percent of grandparents provided at least 50 hours of care a year for grandchildren they did not live with. Nearly two thirds of these caregivers provided between 50 and 199 hours of care. However, 5% of all grandparents provided 500 or more hours of care a year (10 or more hours a week). About 7% of grandparents lived with grandchildren. Most of these households included three generations. Less than 2% of grandparents lived with grandchildren in skipped-generation households. Race/ethnic differences in prevalence and type of grandchild care are evident in Table 1, and chi-square test shows these differences are statistically significant (p < .001). The prevalence of grandchild care is highest among African American grandparents (44%) and lowest among Whites (33%), with Hispanics in between (38%). Among caregiving grandparents, non-Hispanic Whites provided more babysitting than African Americans and Hispanics and, conversely, provided less coresidential care. African American and Hispanic caregivers provided about the same amount of nonresidential care (55% and 53%), but African Americans were more likely to live in a skipped-generation household than Hispanics (14% and 9%) while Hispanics were more likely to live in multigenerational households than African Americans (39% and 31%).

Table 2 shows the number of interviews at which respondents reported either that they had provided at least 100 hours of care to their grandchildren in the past 2 years or that they were living with their grandchildren at the time of the interview, among the respondents who were grandparents in 1998 and provided data about grandchild care in all six waves. About 61% of grandparents reported care at least once; 16% reported care in one interval, 32% reported care during two to four intervals, and 14% reported care in five or six intervals. A higher proportion of African American and Hispanic than non-Hispanic White grandparents reported care in five or more intervals. The table also presents proportions of grandparents who provided the same type of care in all the intervals in which they were caring for grandchildren. Among grandparents who provided care in two intervals, 91% provided the same type of care.

Table 2.

Number of Interviews From 1998 to 2008 at Which Grandparents Were Caring for Grandchildren and Percentages of Grandparents Providing the Same Type of Care in All Interviews

Number of Interviews All White Black Hispanic
0 38.6 39.9 29.1 34.4
1 16.1 16.3 15.2 14.5
2 12.4 (91.1) 12.3 (93.1) 14.6 (79.3) 12.1 (84.1)
3 10.2 (87.2) 10.1 (90.5) 11.0 (70.2) 10.8 (69.3)
4 9.0 (78.5) 8.8 (82.3) 8.9 (59.7) 11.0 (57.6)
5 7.1 (72.8) 6.7 (80.2) 9.9 (37.4) 8.7 (49.1)
6 6.5 (67.9) 5.9 (75.1) 11.2 (48.0) 8.6 (36.9)
N 7,692 6,166 946 580
Open in a new tab

Note: Numbers are weighted percentages of respondents in each category of number of interviews. Numbers in parentheses are percentages of grandparents who provided the same type of care in all the interviews at which they were providing care. Sample includes respondents who were grandparents and provided information about grandchild care at every interview between 1998 and 2008.

We observe respondents six times at 2-year intervals between 1998 and 2008, thus most respondents contribute five intervals of observation. In Table 3, we treat these intervals as the unit of observation and examine the frequency of various transitions in grandchild care. We see that in more than half of the intervals, grandparents were providing no care (i.e., less than 50 hours a year) for grandchildren at either the beginning or the end of the 2-year interval. Whites were more likely than African Americans or Hispanics to provide no care. In 9% of the intervals, grandparents began some kind of care, mainly babysitting. Nearly 19% of intervals showed grandparents continuing some kind of care, again, primarily babysitting, and in 11% of intervals, grandparents ended their caregiving responsibilities. In about 1% of intervals, grandparents provided more care (i.e., moved from babysitting to multigenerational or skipped-generation household or from a multigenerational to skipped-generation household); similarly, about 1% of intervals showed grandparents reduced their caregiving.

Table 3.

Frequency of Transitions in Grandchild Care Over 2-Year Intervals

All White Black Hispanic
No care both interviews 59.2 60.7 50.1 52.9
Start babysitting 8.3 8.4 7.4 7.6
Start multigeneration household 0.7 0.5 1.1 2.1
Start skipped-generation household 0.2 0.1 0.6 0.4
Continue babysitting 14.6 15.3 11.8 8.9
Continue multigeneration household 2.9 1.8 8.3 9.8
Continue skipped-generation household 0.8 0.5 3.2 1.6
Stop babysitting 10.3 10.5 9.7 8.7
Stop multigeneration household 0.8 0.5 1.7 2.9
Stop skipped-generation household 0.3 0.2 1.1 0.8
More care 1.1 0.8 2.4 2.6
Less care 0.9 0.7 2.5 1.8
Number of intervals 52,883 41,657 7,040 4,186
Open in a new tab

Note: Unit of observation is 2-year interval between interviews. Numbers are weighted percentages. Sample includes respondents who were grandparents and provided information on grandchild care in at least two consecutive interviews from 1998 to 2008.

Correlates of Grandchild Care Transitions

Table 4 presents odds ratios derived from regressions of 2-year transitions in grandchild care on variables measuring demographic characteristics, social roles, socioeconomic status, and health. Results for grandparents who were caregivers at the beginning of the interval are presented in the left panel, and the results for those who were not caregivers are presented in the right panel. The four babysitting categories are combined for this analysis.1

Table 4.

Odds Ratios From Binary and Multinomial Logistic Regressions of 2-Year Transitions in Grandchild Care, by Initial Caregiving Status

Start Care Versus Continue No Care, Non-Caregiving Grandparents
 Continue Care Versus Stop Care, Caregiving Grandparents
Start Carea
 Start Babysittingb
 Start Multigeneration Householdb
 Start Skipped- Generation Householdb
 Continue Carec
 Continue Babysittingd
 Continue Multigeneration Householdd
 Continue Skipped- Generation Householdd
(Column 1) (Column 2) (Column 3) (Column 4) (Column 5) (Column 6) (Column 7) (Column 8)
Blacke 1.26** 1.16* 1.69** 5.22** 1.04 0.94 1.43** 1.66**
Hispanice 1.28** 1.11 2.59** 2.58* 0.85+ 0.70** 1.32+ 1.13
Female 1.15** 1.15** 1.16 1.10 1.35** 1.49** 1.20+ 1.04
Age 0.93** 0.93** 0.94** 0.94** 0.96** 0.96** 0.96** 0.93**
Married 1.29** 1.33** 0.95 1.35 1.09 1.12+ 0.82+ 1.21
Number of own children <18 in household 0.76** 0.68** 1.34* 1.07 0.73** 0.67** 1.20 0.48**
Number of grandchildren 1.01** 1.01* 1.01 1.05** 0.99* 0.99** 0.99+ 1.00
Fewer grandchildrene 0.86* 0.83* 1.29 0.62 0.91 0.88 1.00 1.21
More grandchildrene 1.22** 1.22** 1.35* 1.20 1.07 1.09+ 0.96 1.11
Working part-timee 1.08 1.11 0.80 0.50 1.14+ 1.12 1.04 1.76**
Not workinge 0.96 0.99 0.67* 0.67 1.28** 1.26** 1.25+ 1.60**
Years of education 1.01 + 1.02** 0.95+ 0.97 1.02+ 1.02** 0.99 0.97
Household income (log) 1.02 1.02 0.98 1.01 1.04* 1.05* 0.98 1.02
Household net worth (log) 1.02** 1.04** 0.96* 0.99 1.00 1.01 0.99 0.97
Number of functional limitations 0.98* 0.98* 1.03 1.03 0.97** 0.97** 0.99 0.97
Number of chronic conditions 0.99 0.99 1.00 1.11 1.00 0.99 1.05 1.00
Number of observations 36,310 36,310 16,368 16,368
χ2 1,000.66 1,394.69 720.31 4,902.49
df 21 63 23 69
Open in a new tab

Note: Unit of analysis is 2-year interval between interviews. Standard errors adjusted for intraperson correlation. Results are weighted. A dummy variable indicating missing data on number of functional limitations and four dummy variables indicating the beginning year of the 2-year intervals also included in the models but not shown in the table.

a.

Logistic regression model with “continue no care” as comparison group.

b.

Multinomial logistic regression model with “continue no care” as comparison group.

c.

Logistic regression model with “stop care” as comparison group (type of care at the beginning of the interval is also included in the model, but not shown in the table).

d.

Multinomial logistic regression model with “stop care” as comparison group (type of care at the beginning of the interval is also included in the model, but not shown in the table).

e.

Reference categories are non-Hispanic White, no change in number of grandchildren, and working full-time.

†

p < .1.

*

p < .05.

**

p < .01, two-tailed tests.

Among grandparents who were not caring for grandchildren at the beginning of the interval, the odds of starting any type of grandchild care by the end of the interval are 26% higher for African Americans than for Whites, 28% higher for Hispanics than for Whites, 15% higher for grandmothers than for grandfathers, and 7% lower for every 1-year increase in the grandparent’s age (Column 1).2 Looking at transitions into different types of care (Columns 2–4), we see that, compared with Whites, African Americans and Hispanics are much more likely to start a skipped-generation or multigeneration household. Hispanics are also more likely to start multigeneration households than Blacks, but they are less likely to start skipped-generation households than Blacks. Grandmothers are 15% more likely than grandfathers to start nonresidential care. Among grandparents who were caring for grandchildren at the beginning of the interval, the odds of continuing any type of care are 35% higher for grandmothers than for grandfathers and 4% lower for every 1-year increase in age (Column 5). These effects are generally consistent across different types of care, but the effect of race/ethnicity varies by care type (Columns 6–8). Although African Americans are no more likely to continue babysitting than Whites, their odds of continuing a multigeneration or skipped-generation household are 43% and 66% higher, respectively, than Whites. Hispanics are less likely to continue babysitting than Whites or Blacks. They are marginally more likely to continue a multigeneration household than Whites. Blacks are more likely than Hispanics to continue skipped-generation households, but their difference in the odds of continuing multigeneration households is not significant.

Grandparents’ role obligations and family demands also affect transitions in grandchild care. Compared with their unmarried counterparts, married grandparents are 29% more likely to start caring for their grandchildren (Column 1). The odds that they start babysitting are 33% higher (Columns 2). Married grandparents are also marginally more likely to continue babysitting care and less likely to continue living in a multigenerational household (Columns 6 and 7). However, grandparents with their own children under 18 living in the household are less likely to start babysitting, though they are more likely to start living in a multigeneration household (Columns 2 and 3). Having their own minor children at home also reduces the odds that grandparents will continue to provide babysitting or live in a skipped-generation household (Columns 6 and 8). Having more grandchildren and the addition of new grandchildren increases the likelihood of starting grandchild care. However, initial number of grandchildren is associated with decreased likelihood of continuing babysitting (Column 6). Part-time employment only significantly increases the likelihood of continuing care in a skipped-generation household, but grandparents who were not working are more likely to continue all types of care than grandparents employed full-time, though they are less likely to start a multigeneration household.

Socioeconomic status appears to have opposite effects on babysitting care and coresidential care. Grandparents’ education increases the odds of beginning and continuing babysitting. Grandparents with more assets are more likely to begin babysitting, and grandparents with more income are more likely to continue babysitting. In contrast, the odds of beginning to live in a multigeneration household significantly decrease with higher household assets and marginally decrease with higher education.

The impact of health on the likelihood of different types of caregiving transitions is most clearly reflected in the effect of functional limitations on babysitting. Each additional functional limitation reduces the odds of starting babysitting and continuing to babysit by 2% to 3%. Overall, health status does not influence coresidential care. Neither the number of functional limitations nor the number of chronic conditions is significantly associated with the odds that a grandparent starts or stops living with their grandchildren.

Discussion

This study examined the dynamics of grandchild care among grandparents and the predictors of grandparents’ transitions into and out of grandchild care using the HRS, a large, longitudinal, nationally representative survey. Each of these features is fundamental to our efforts. Only large data sets contain sufficient cases for detailed categorization of types of care. It should be noted that, even the HRS, which is quite sizeable, contains few respondents reporting some types of care. Only longitudinal data allow us to examine transitions into and out of grandchild care and between types of care. And data that are representative of the population of older adults allow us to discuss the prevalence of various types of grandchild care separately for White, African American, and Hispanic grandparents.

Grandparent caregiving, especially grandparent-headed households, is an increasing phenomenon in the United States (Livingston & Parker, 2010; Pebley & Rudkin, 1999; Simmons & Dye, 2003). Coresidence with grandchildren is relatively rare in the population at any given time. Over 2-year periods, we observed a similar number of transitions into and out of coresidential households, even though we did not capture any movements that may have occurred in the 2 years between interview waves; if there are many short periods of coresidence, the proportion of grandparents who ever lived with a grandchild will be higher (Caputo, 2001). According to the 2000 census, 54% of grandparent primary caregivers had been providing care for 3 or more years (Simmons & Dye, 2003), suggesting we have captured a majority of these transitions. In addition, babysitting for grandchildren among grandparents age 50 and older is fairly common; in 1998 about 28% of grandparents provided at least 50 hours of care per year and 5% provided 500 hours of care per year. This type of grandchild care needs more research attention in the future.

We examined transitions in grandchild care among grandparents who provide care to grandchildren who do not live with them and among grandparents whose grandchildren moved into their household, with or without the child’s parent. As expected, African American and Hispanic grandparents are more likely than Whites to begin and continue a multigeneration household or start a skipped-generation household, with African American grandparents more likely than both groups to start and maintain a skipped-generation household and Hispanic grandparents more likely than both groups to start a multigeneration household. These differences may reflect cultural norms of family support and the role of grandparents and contextual factors in the middle generation that influence the availability of parents to be present in the home (Cox, Brooks, & Valcarcel, 2000; Ruiz, 2000).

We found mixed support for our hypothesis regarding gender. Grandmothers are more likely than grandfathers to start and continue babysitting, and the effects of gender on transitions into and out of coresidential care are in the same direction, but they are mostly not significant. When asking about babysitting it is more obvious who is doing the care work because respondents were specifically asked about care they provided in the past 2 years. With coresidence the grandchild may be living with both the grandmother and the grandfather, but the responsibility for care may fall primarily on the grandmother. Future research should consider an interaction between gender and marital status. Our results showed consistent age effects: the younger the grandparent, the more likely she or he will start or remain in any type of care. This finding is consistent with other quantitative studies and seems to suggest that despite the higher levels of stress over role conflicts associated with grandchild care younger grandparents still do the caring work when it is needed.

Our hypotheses on the effects of other role obligations and family demands receive mixed support. As we hypothesized, with increasing number of their own minor children living in the household, grandparents are less likely to start or continue most types of care with the exception that they are more likely to start a multigeneration household. With more grandchildren and the addition of new grandchildren, grandparents are more likely to start care, though with more grandchildren they are also more likely to stop babysitting. Grandparents who are not working are more likely to continue all types of care, but they are less likely to start a multigeneration household. Although we offered no specific hypotheses about marital status, it appears that being married functions more as a resource than as a competing demand for time, as married grandparents are more likely to begin and continue babysitting. However, married grandparents are either no different or less likely to continue coresidential care than other grandparents.

We also found mixed support for our initial hypotheses regarding socioeconomic resources and health conditions. Consistent with our hypothesis, grandparents with higher levels of education and household assets are less likely to start multigeneration households. However, contrary to our expectations, grandparents with higher levels of education, income, and assets are more likely to start and continue babysitting. Functional limitations significantly reduce the odds of starting and continuing babysitting, but do not impact coresidential care.

These results suggest that nonresidential care is discretionary; grandparents seem to undertake such care when they are willing and able to provide it. Grandparents who were African American, female, younger, married, had fewer minor children of their own in the household, had more education, had higher levels of income or assets, and fewer functional limitations were more likely than others to begin and continue babysitting. Grandparents who provide coresidential care, however, may be forced more often by circumstances to undertake that care; African Americans, Hispanics, and those with lower levels of education and relatively few economic resources face greater odds of beginning and continuing to provide care to grandchildren who live with them. Grandparents who neither provided care initially nor began providing care seem to fall between babysitting caregivers and coresidential caregivers in education, income, and assets. They are substantially older, less likely to be married, and less likely to be working than grandparents caring for grandchildren who live elsewhere. These grandparents do not seem to have the resources to provide care for grandchildren who do not live with them and have children who either do not need the help that coresidence would provide or who are managing on their own.

The mixed findings from our longitudinal analysis suggest more complex dynamics of caregiving by grandparents to their grandchildren than what was portrayed in cross-sectional snapshots. Our results suggest substantial heterogeneity among grandparents who provide care to grandchildren. These initial differences mean that researchers must make comparisons among grandparent caregivers carefully. In addition, the type of care involved varies a great deal between these situations. These findings point to the substantial disadvantages faced by grandparents who begin coresidential care.

Limitations

Although the HRS has many strengths for the study of grandchild care, it also has some important weaknesses. One limitation of the HRS for our purposes is the exclusion of those under age 50. A sizeable share (28%) of grandparents who live with grandchildren in either three-generation or skipped-generation households are younger than the HRS sample (Simmons & Dye, 2003), and their experience was not captured in our study. The proportion of grandparents under age 50 who provide babysitting remains unknown. In addition, the skipped-generation households may be underestimated because we cannot identify the parent of the child in the data. Some households classified as multigenerational in our analysis may contain the aunt or uncle of the grandchild but not the parent.

Our measure of care also has limitations. The HRS asks respondents only about babysitting care totaling 50 hours a year or about an hour a week. We do not view this as highly problematic, since it excludes only those who provide few hours of care, but its arbitrary nature should be kept in mind. More seriously, we have no measure of the timing and distribution of this care. The reference period is the entire 2-year period prior to the interview; we do not know if the hours of care were bunched together or spread out over the interval. Similarly, we are not able to control for the duration of this care. Also, it is not possible to distinguish which grandchildren and how many of them the grandparent cared for, and we have no information about the specific activities done or services provided. Among grandparents residing with grandchildren, we do not identify the primary caregiver. We assume that grandparents coresiding with grandchildren are providing some form of care, either directly or indirectly. Although this is consistent with previously published studies, looking at grandchild care using the HRS (e.g., Hughes et al., 2007; Szinovacz & Davey, 2006), future studies including detailed information on grandparent–grandchild interactions may provide greater insights on this issue.

We acknowledge that our measures of role obligations and demands, such as marital and employment status and children in the home, cannot fully capture family dynamics, quality of relationships, or changes that might cause transitions in caregiving. We know nothing about reasons for caregiving, characteristics of grandchildren or their families, although these are important. Finally, because of the small sample size for certain types of care, we pooled data from six waves of the HRS to produce more robust results on the transitions into and out of different types of care. Future research may examine cohort differences in caregiving dynamics with larger samples and more waves of data.

Implications for Social Policy

Our findings that different groups of grandparents are at risk of different types of care, and more important, grandparents of low socioeconomic status or who are minority are likely to take on extra care burden, have implications for social policy. Child welfare agencies are increasingly relying on family members, especially grandparents, to provide care to children when birth parents are unable to do so (Geen, 2004; Schwartz, 2002). This is within the tradition of U.S. policy, which expects families to provide a safety net. However, our findings suggest that grandparent caregivers who provide the most demanding care—which may well be involuntary—are already vulnerable, and so this “safety net” strategy is likely not without costs. Kin foster parents receive less financial support and fewer services than non-kin caregivers (Geen, 2004; Letiecq, Bailey, & Porterfield, 2009; Schwartz, 2002). Lack of resources may increase the burdens they experience and erode the quality of child care. Social programs should pay more attention to improving the financial and health conditions of family caregivers, especially grandparents who provide intensive coresidential care to their grandchildren. Daycare assistance may be particularly needed by the middle-aged grandparents who are juggling multiple role obligations, as a parent, a grandparent, and a paid employee. The heterogeneity of grandchild care and its risk factors suggest that a broad range of practices and services with targeted outreach hold greater promises in helping grandparents who provide care to their grandchildren.

Acknowledgments

We thank Pamela J. Smock and John T. Cacioppo for their helpful comments.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by Grants P01 AG 18911 and R37 AG030481 from the National Institute on Aging.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

1.

Additional analyses using more refined babysitting categories show that the significant predictors vary somewhat between babysitting more than 500 hours a year and babysitting less than 500 hours a year, although the directions of their effects are consistent across these categories.

2.

To test whether age has a nonlinear effect on transitions in grandchild care, we also recoded age into a series of dummy variables for decade and entered these into the regression models. The results show a gradient effect of age for all types of care transitions except for starting a skipped-generation household. The likelihood of starting a skipped-generation household appears to be higher among grandparents aged 60 to 69 than those aged 50 to 59, though the difference is not significant. Grandparents aged 70 to 90 are significantly less likely to start a skipped-generation household than those aged 60 to 69.

References

  1. Baker LA, & Silverstein M (2008). Preventive health behaviors among grandmothers raising grandchildren. Journal of Gerontology: Social Sciences, 63, S304–S311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bengtson VL (2001). Beyond the nuclear family: The increasing importance of multigenerational bonds. Journal of Marriage and Family, 63, 1–16. [Google Scholar]
  3. Blustein J, Chan S, & Guanais FC (2004). Elevated depressive symptoms among caregiving grandparents. Health Services Research, 39, 1671–1689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brody EM, & Saperstein AR (2004). Women in the middle: Their parent care years. New York, NY: Springer. [Google Scholar]
  5. Burton LM, & Bengtson VL (1985). Black grandmothers: Issues of timing and continuity of roles. In Bengtson VL & Robertson JF (Eds.), Grandparenthood (pp. 61–77). Beverly Hills, CA: Sage. [Google Scholar]
  6. Caputo RK (2001). Depression and health among grandmothers co-residing with grandchildren in two cohorts of women. Families in Society, 82, 473–483. [Google Scholar]
  7. Cherlin AJ, & Furstenberg FF Jr. (1986). The new American grandparent: A place in the family, a life apart. New York, NY: Basic Books. [Google Scholar]
  8. Cox CB, Brooks LR, & Valcarcel C (2000). Culture and caregiving: A study of Latino grandparents. In Cox CB (Ed.), To grandmother’s house we go and stay: Perspectives on custodial grandparenting (pp. 218–232). New York, NY: Springer. [Google Scholar]
  9. Dellmann-Jenkins M, Blankemeyer M, & Olesh M (2002). Adults in expanded grandparent roles: Considerations for practice, policy, and research. Educational Gerontology, 28, 219–235. [Google Scholar]
  10. Eggebeen DJ, & Hogan DP (1990). Giving between generations in American families. Human Nature, 1, 211–232. [DOI] [PubMed] [Google Scholar]
  11. Fields J (2003). Children’s living arrangements and characteristics: March 2002. In Current population reports, population characteristics (Series p20–547). Washington, DC: Government Printing Office. [Google Scholar]
  12. Fuller-Thomson E, & Minkler M (2000). African American grandparents raising grandchildren: A national profile of demographic and health characteristics. Health & Social Work, 25, 109–118. [DOI] [PubMed] [Google Scholar]
  13. Fuller-Thomson E, & Minkler M (2001). American grandparents providing extensive child care to their grandchildren: Prevalence and profile. The Gerontologist, 41, 201–209. [DOI] [PubMed] [Google Scholar]
  14. Fuller-Thomson E, Minkler M, & Driver D (1997). A profile of grandparents raising grandchildren in the United States. The Gerontologist, 37, 406–411. [DOI] [PubMed] [Google Scholar]
  15. Geen R (2004). The evolution of kinship care policy and practice. Children, Families and Foster Care, 14, 131–139. [PubMed] [Google Scholar]
  16. Goodman CC, & Silverstein M (2002). Grandmothers raising grandchildren: Family structure and well-being in culturally diverse families. The Gerontologist, 42, 676–689. [DOI] [PubMed] [Google Scholar]
  17. Greene WH (1997). Econometric analysis. Upper Saddle River, NJ: Prentice-Hall. [Google Scholar]
  18. Hagestad GO, & Burton LM (1986). Grandparenthood, life context, and family development. American Behavioral Scientist, 29, 471–484. [Google Scholar]
  19. Hank K, & Buber I (2008). Grandparents caring for their grandchildren: Findings from the 2004 Survey of Health, Ageing, and Retirement in Europe. Journal of Family Issues, 30, 53–73. [Google Scholar]
  20. Health and Retirement Study. (2011). Sample sizes and response rates. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/sampleresponse.pdf
  21. Henretta JC, Grundy E, & Harris S (2002). The influence of socio-economic and health differences on parents’ provision of help to adult children: A British-United States comparison. Ageing & Society, 22, 441–459. [Google Scholar]
  22. Hughes ME, & Waite LJ (2002). Health in household context: Living arrangements and health in late middle age. Journal of Health and Social Behavior, 43, 1–21. [PMC free article] [PubMed] [Google Scholar]
  23. Hughes ME, Waite LJ, LaPierre TA, & Luo Y (2007). All in the family: The impact of caring for grandchildren on grandparents’ health. Journal of Gerontology: Social Sciences, 62, S108–S119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Letiecq BL, Bailey SJ, & Porterfield F (2009). “We have no right, we get no help”: The legal and policy dilemmas facing grandparent caregivers. Journal of Family Issues, 29, 995–1012. [Google Scholar]
  25. Livingston G, & Parker K (2010). Since the start of the Great Recession, more children raised by grandparents [Electronic version] (Pew Research Center Social & Demographic Trends; ). Retrieved from http://pewsocialtrends.org/2010/09/09/since-the-start-of-the-great-recession-more-children-raised-by-grandparents/ [Google Scholar]
  26. Lumpkin JR (2008). Grandparents in a parental or near-parental role: Sources of stress and coping mechanisms. Journal of Family Issues, 29, 357–372. [Google Scholar]
  27. Minkler M, & Fuller-Thomson E (2000). Second time around parenting: Factors predictive of grandparents becoming caregivers for their grandchildren. International Journal of Aging & Human Development, 50, 185–200. [DOI] [PubMed] [Google Scholar]
  28. Minkler M, & Fuller-Thomson E (2005). African American grandparents raising grandchildren: A national study using the Census 2000 American Community Survey. Journal of Gerontology: Social Sciences, 60, S82–S92. [DOI] [PubMed] [Google Scholar]
  29. Musil CM, Gordon NL, Warner CB, Zauszniewski JA, Standing T, & Wykle M (2011). Grandmothers and caregiving to grandchildren: Continuity, change, and outcomes over 24 Months. The Gerontologist, 51, 86–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Pebley AR, & Rudkin LL (1999). Grandparents caring for grandchildren: What do we know? Journal of Family Issues, 20, 218–242. [Google Scholar]
  31. Pinson-Millburn NM, Fabian ES, Schlossberg NK, & Pyle M (1996). Grandparents raising grandchildren. Journal of Counseling & Development, 74, 548–554. [Google Scholar]
  32. Presser HB (2004). Working in a 24/7 economy: Challenges for American families. New York, NY: Russell Sage Foundation. [Google Scholar]
  33. Rossi AS (1995). Wanted: Alternative theory and analysis modes. In Bengtson VL, Shaia KW, & Burton LM (Eds.), Adult intergenerational relations: Effects of societal change (pp. 264–276). New York, NY: Springer. [Google Scholar]
  34. Ruiz D (2000). Guardian and caretakers: African American grandmothers as primary caregivers in intergenerational families. African American Research Perspectives, 6, 1–12. [Google Scholar]
  35. Schwartz AE (2002). Societal value and the funding of kinship care. Social Service Review, 76, 430–459. [Google Scholar]
  36. Silverstein M, Giarrusso R, & Bengtson VL (2003). Grandparents and grandchildren in family systems: A socio-developmental perspective. In Bengtson VL & Lowenstein A (Eds.), Global aging and challenges to families (pp. 75–102). New York, NY: Aldine de Gruyter. [Google Scholar]
  37. Silverstein M, & Marenco A (2001). How Americans enact the grandparent role across the family life course. Journal of Family Issues, 22, 493–522. [Google Scholar]
  38. Simmons T, & Dye JL (2003). Grandparents living with grandchildren: 2000. Washington, DC: U.S. Census Bureau. [Google Scholar]
  39. Strawbridge WJ, Wallhagen MI, Shema SJ, & Kaplan GA (1997). New burdens or more of the same? Comparing grandparent, spouse, and adult child caregivers. The Gerontologist, 37, 505–510. [DOI] [PubMed] [Google Scholar]
  40. Szinovacz M (1998). Grandparents today: A demographic profile. The Gerontologist, 38, 37–52. [DOI] [PubMed] [Google Scholar]
  41. Szinovacz M, & Davey A (2006). Effects of retirement and grandchild care on depressive symptoms. International Journal of Aging & Human Development, 62, 1–20. [DOI] [PubMed] [Google Scholar]
  42. Tienda M, & Angel R (1982). Headship and household composition among Blacks, Hispanics, and other Whites. Social Forces, 61, 508–531. [Google Scholar]
  43. Uhlenberg P, & Hammill B (1998). Frequency of grandparent contact with grandchild sets: Six factors that make a difference. The Gerontologist, 38, 276–285. [DOI] [PubMed] [Google Scholar]
  44. Waldrop DP, & Weber JA (2001). From grandparent to caregiver: The stress and satisfaction of raising grandchildren. Families in Society, 82, 467–472. [Google Scholar]
  45. Waite LJ, & Hughes ME (1999). At risk on the cusp of old age: Living arrangements and functional status among Black, White and Hispanic adults. Journal of Gerontology: Social Sciences, 54, S136–S144. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES