Full text of DOI:10.1093/geront/40.6.718

The Gerontologist
Copyright 2000 by The Gerontological Society of America
Vol. 40, No. 6, 718–728
Comparisons of African American and White
Women in the Parent Care Role
Tracela M. White, PhD,1 Aloen L. Townsend, PhD,2
and Mary Ann Parris Stephens, PhD3
Little is known about African American women’s experiences
providing care to impaired older relatives. This study investi-
gated potential differences in depressive symptomatology,
parent care stress and rewards, parent care mastery, and the
quality of the parent care relationship between 261 White and
56 African American daughters and daughters-in-law who
were providing care for an impaired parent or parent-in-law.
Multivariate analysis of variance, controlling for significant
background characteristics and interrelationships among care-
giving experiences, revealed that African American women
reported less stress and more rewards in the parent care role
than White women did. Race did not have a significant effect
on caregivers’ depressive symptomatology, parent care mas-
tery, or the quality of their relationship with the parent. How-
ever, this research demonstrates the importance of examining
a broad range of caregiving experiences in order to detect
both similarities and differences between African American
and White caregivers.
children (especially daughters) who will assume the
caregiving role. Yet, research on family care to elderly
relatives historically has focused on samples exclu-
sively or predominately composed of White caregiv-
ers. As a result, little is known about the caregiving
experiences of African American families.
The studies of African American caregivers that
have begun to appear indicate that there may be dif-
ferences in the psychological well-being and experi-
ences of family caregivers by race. The purpose of the
present study is to extend knowledge about potential
differences in well-being and caregiving experiences
between African American and White daughters and
daughters-in-law who are providing care for an im-
paired parent or parent-in-law. The study investigates
caregivers’ depressive symptomatology, parent care
stress and rewards, parent care mastery, and the qual-
ity of their relationship with the impaired relative.
Prior research has shown that there is a relation-
ship between providing care for an impaired elderly
relative and depressive symptomatology. For instance,
caregivers have been found to report significantly
higher levels of depressive symptomatology than non-
caregivers (Dura, Stukenberg, & Kiecolt-Glaser, 1991;
Haley, Levine, Brown, Berry, & Hughes, 1987; Schulz,
Tompkins, & Rau, 1988). In addition, much of the re-
search comparing the effects of caregiving on African
American and White caregivers’ depressive symp-
tomatology has found that African American caregiv-
ers report lower symptomatology than White caregiv-
ers (Haley et al., 1995; Lawton, Rajagopal, Brody, &
Kleban, 1992; Miller, Campbell, Farran, Kaufman, &
Davis, 1995; Mintzer & Macera, 1992). However, a
study by Young and Kahana (1995) suggests that con-
trolling for caregiving context variables—such as the
age of the caregiver, impairment of the care recipi-
ent, and hours spent providing care—eliminates dif-
ferences in depressive symptomatology between White
and African American caregivers.
Key Words: Caregiver, Race, Depression, Stress, Women
By the year 2050, it is estimated that the number
of African Americans aged 65 and over will more
than triple and their proportion of the total elderly
population will increase from 8% to 10% (U.S. Bu-
reau of the Census, 1996). Because African American
elders frequently rely on their children for assistance
with physical and psychological needs (see review
by Wykle & Kaskel, 1991), the changing demograph-
ics among the aging population have significant im-
plications for the number of African American adult
This research was part of a larger study, “Multiple Roles of Middle-
Generation Caregiving Women,” funded by National Institute on Aging
Grant RO1 AG11906; a National Institutes of Health Research Supplement
for Underrepresented Minorities; and the Department of Psychology at
Kent State University. We acknowledge the assistance of the Women’s
Multiple Roles Study office staff, interviewers, and participants. We also
appreciate the valuable feedback that Dr. Powell Lawton provided on an
earlier version of the manuscript. Support for manuscript preparation was
provided by T. M. W.’s postdoctoral fellowship that was funded by NRSA
Training Grant 5T32 MH19331 in Psychiatric Medical Comorbidity at the
University of Pennsylvania, Section on Geriatric Psychiatry.
Although depression is the most commonly stud-
ied outcome in the caregiving literature, considerable
research also has examined the impact of providing
care on caregivers’ subjective stress. Many studies
have found that providing care for an aging relative
can be stressful (Lawton, Moss, Kleban, Glicksman,
& Rovine, 1991; Townsend, Noelker, Deimling, &
Address correspondence to Tracela M. White, PhD, University of
Pennsylvania Department of Psychiatry, Section on Geriatric Psychiatry,
3600 Market Street, Room 759, Philadelphia, PA 19104. E-mail: tmwhite@
mail.med.upenn.edu
1University of Pennsylvania School of Medicine, Philadelphia.
2Mandel School of Applied Social Sciences, Case Western Reserve
University, Cleveland, Ohio.
3Department of Psychology, Kent State University, Ohio.
718
The Gerontologist

Bass, 1989; Zarit, Reever, & Bach-Peterson, 1980).
Most research that focuses on African American care-
givers suggests that levels of caregiving stress may
vary by race (Aranda & Knight, 1997). For instance,
African American caregivers report, on average, less
subjective burden, lower levels of caregiver role strain,
and less sense of intrusion on their lives from caregiv-
ing responsibilities than White caregivers (Hinrichsen
& Ramirez, 1992; Lawton et al., 1992; Macera et al.,
1992; Miller et al., 1995; Mui, 1992). Alternatively,
three studies found no difference in stress or burden
between African American and White caregivers (Cox,
1993; Morycz, Malloy, Bozich, & Martz, 1987; Wood
& Parham, 1990).
Fewer studies recognize the potential benefits or
rewards of the caregiving role than those that focus
on the stresses associated with caregiving (Stephens
& Townsend, 1997). Only two studies are known to
have investigated the relationship between race and
caregiving rewards. One is a study by Lawton and
colleagues (1992), which revealed that African Amer-
ican caregivers expressed greater caregiving satisfac-
tion (e.g., benefits from caregiving) than White care-
givers did. The second study (Picot, Debanne, Namazi,
& Wykle, 1997) also found that race was significantly
related to perceived rewards, with African American
caregivers reporting higher levels of rewards than
White caregivers.
A growing interest in caregivers’ sense of mastery
is accruing. Global measures as well as role-specific
measures of mastery have been used in the caregiv-
ing literature. Global measures of mastery assess in-
dividuals’ feelings of control over forces affecting
their lives generally (Aneshensel, Pearlin, Mullan, Zarit,
& Whitlatch, 1995). Role-specific caregiving mastery
has been defined as caregivers’ beliefs in their ability
to influence or control care-related events or to man-
age these events in a competent or effective manner
(Lawton et al., 1992; Pearlin, Mullan, Semple, & Skaff,
1990). Research utilizing primarily White samples
has consistently shown that greater caregiving mas-
tery is significantly related to better psychological
well-being of caregivers (Coppel, Burton, Becker, &
Fiore, 1985; Haley, Bartolucci, Levine, & Brown,
1987; Pagel, Becker, & Coppel, 1985). Two studies
are known to have examined the relationship be-
tween race and mastery. One concluded that African
American caregivers espouse a greater sense of role-
specific mastery in caregiving than White caregivers
(Lawton et al., 1992). The other study found no sig-
nificant difference in African American and White
caregivers’ global sense of mastery or in their role-
specific caregiving mastery (Miller et al., 1995).
The quality of family relationships has been a rela-
tively neglected topic in the caregiving literature, al-
though the detrimental influence of negative ties and
the beneficial influence of positive ties in the caregiv-
ing relationship have become increasingly evident
with samples composed exclusively or primarily of
White caregivers (Creasey, Myers, Epperson, & Tay-
lor, 1990; Fiore, Becker, & Coppel, 1983; Townsend
& Franks, 1995). To our knowledge, no study has in-
vestigated the association between caregivers’ race
and the quality of their relationship with care recipi-
ents. However, one study (Mui, 1992) has explored
race as a moderator of the association between the
quality of the caregiving relationship and caregivers’
role strain. In that study, White daughters reported
more caregiving role strain when their relationship
with the impaired parent was poor, but the quality of
the parent–daughter relationship was not a significant
predictor of caregiving role strain for African Ameri-
can daughters.
A variety of explanations have been offered to ex-
plain why caregiving experiences differ by race.
Some authors (e.g., Dilworth-Anderson & Anderson,
1994; Haley et al., 1996) suggest that racial and eth-
nic differences in values and beliefs about aging may
account for the observed differences. Differences in
role expectations and attitudes about providing care
and filial support may also account for variations in
caregiving experiences (Cox, 1993; Haley et al., 1996;
Lawton et al., 1992). Another possible explanation is
racial/ethnic differences in religiosity or coping
mechanisms that may moderate caregiving experi-
ences (e.g., Aranda & Knight, 1997; Picot et al.,
1997; Segall & Wykle, 1988).
In addition to contributing to the literature on Afri-
can American caregivers, the present study extends
previous research on race and caregiving in six im-
portant ways. First, the sample of caregivers is homo-
geneous with respect to kin relationship, composed
solely of daughters and daughters-in-law (both re-
ferred to hereafter as daughters). In prior research, a
variety of kin and even non-kin relationships have
been included. The present study chose to focus on a
homogeneous sample of caregivers because previous
research has shown that caregiving outcomes differ
between adult children and spouses (George &
Gwyther, 1986; Macera et al., 1992; Wood & Par-
ham, 1990), and such variation may confound differ-
ences between race and kin relationship. Second, the
research design also required that participants occupy
three roles (mother, wife, and employee) in addition
to that of primary caregiver. The relationship between
multiple roles and caregiving experiences has not
been taken into account in previous research on Afri-
can American and White caregivers. Third, the
present study focuses on positive aspects (rewards) as
well as negative aspects (stresses) of caregiving. Prior
research has devoted much more attention to stresses
than rewards. Fourth, the experience of caregivers
who provide care for an aging parent or parent-in-law
(both referred to hereafter as parent) with various
forms of impairment is examined. Most prior research
has focused on the experiences of African American
and White family members who provide care to rela-
tives with Alzheimer’s disease or some other form of
dementia (e.g., Haley et al., 1996; Hinrichsen &
Ramirez, 1992; Lawton et al., 1992). Therefore, the
results of such studies may not be generalizable to
family members with other forms of impairment (Ory,
Hoffman, Yee, Tennstedt, & Schulz, 1999). Fifth, the
present research focuses on a broader range of con-
Vol. 40, No. 6, 2000
719

structs than prior studies, including two constructs
(role-specific caregiving mastery and quality of the re-
lationship with the care recipient) that have received
particularly limited attention in previous literature on
race and caregiving. For relationship quality, both
positive and negative dimensions were assessed, us-
ing multiple indicators, whereas prior research (Mui,
1992) used a single bipolar indicator. Lastly, the
present analyses employed multivariate techniques
that take into account the interrelationships among
caregiving experiences. Prior research has typically
investigated each construct (e.g., stress) separately.
The present study hypothesized that African Ameri-
can women would report fewer symptoms of depres-
sion, less stress related to parent care, more rewards
from parent care, a greater sense of mastery in the
parent care role, and a more positive relationship
with their impaired parent than White women would
report. In order to further investigate experiences in
the parent care role, parent care centrality (i.e., the
importance of the caregiving role to the participant)
was also included. Role centrality reflects the degree
to which a role functions as a source of identity,
meaning, or behavioral guidance (Gurin, Veroff, &
Feld, 1960; Stryker & Serpe, 1994; Thoits, 1992). Al-
though centrality has received a fair amount of atten-
tion in the literature investigating such social roles as
employee and mother, the relationship between role
centrality and caregiving to an impaired parent is lim-
ited (Martire, Stephens, & Townsend, 1998). No a pri-
ori hypothesis was made about parent care centrality
because no known research has investigated the rela-
tionship between race and caregiver role centrality.
Additionally, participants in the study had to be
married and living with their spouse and had to have
at least one child who was 25 years of age or
younger (and no child over 25) living at home. The
upper age limit of 25 was selected in order to ac-
count for an increasing tendency for young adults to
remain at home for longer periods of time (U.S. Bu-
reau of the Census, 1992). Finally, participants had to
be employed either full time or part time.
Recruitment was conducted in northeastern Ohio
and in two Pennsylvania counties bordering Ohio,
through newspaper articles, radio and newspaper ad-
vertisements, brochures and posters, and notices pub-
lished in newsletters for businesses and social organi-
zations. Special efforts (e.g., contacts with African
American women’s organizations, announcements in
minority-oriented media) were made to recruit Afri-
can American women in proportion to their represen-
tation (13%) in the recruiting area (U.S. Bureau of the
Census, 1992). Potential participants volunteered for
the study by calling a toll-free number or returning a
postage-paid card to the project office. Women who
were eligible to participate in the study were inter-
viewed in-person in their home or other preferred lo-
cation. Interviews averaged one-and-a-half hours.
Sample
A total of 949 women were screened for eligibility.
Of those women, 317 (34%) met the eligibility crite-
ria and were willing to participate. Not providing
care to a parent (22%), not being the primary care-
giver (15%), and having children over the age of 25
or no children living at home (16%) were the three
most common reasons for ineligibility. Because infor-
mation about race was not obtained for women who
were ineligible to participate, no data are available to
compare the eligibility rates for African American
and White women. The final sample (N ϭ 317) con-
sisted of 261 White women (82%) and 56 African
American women (18%).
Methods
Design and Procedures
Data for the present study were obtained from the
first wave of a larger study on the mental health of
middle-generation caregiving women (Stephens &
Townsend, 1997). To be eligible for the study, women
were required to occupy four roles when the study
began: primary caregiver to an impaired parent or
parent-in-law; wife; mother to at least one child liv-
ing at home; and employee. In order to participate,
women must have occupied each of these four roles
for at least 2 months prior to the interview.
Primary caregiver was defined as the person who
spends the most time assisting the parent with at least
one personal activity of daily living (PADL; e.g., eat-
ing) or instrumental activity of daily living (IADL;
e.g., transportation) or with supervision. No restric-
tion was placed on the underlying cause of the par-
ent’s need for assistance. Each participant had to be
providing care to a parent who was living in the com-
munity, but who was not sharing a household with
the participant. Restriction of the sample to partici-
pants who initially did not share a residence with the
parent was based on research indicating differences
in well-being between adult-child caregivers who
live with the impaired parent and those who do not
(e.g., Deimling, Bass, Townsend, & Noelker, 1989).
Measures
Although the larger study collected data about
each of the four roles (parent care, mother, wife, and
employee), the present study will focus solely on the
parent care role. Descriptive data and bivariate cor-
relations for study measures are presented in Table 1.
Depression.—Depressive symptomatology was
measured using the Center for Epidemiologic Stud-
ies–Depression scale (CES–D; Radloff, 1977). This
scale consists of 20 items that assess how frequently
individuals experienced depressive symptoms or feel-
ings during the past week (␣ ϭ .90 for both African
American and White participants). Total scores can
range from 0 to 60, higher scores indicating greater
symptomatology, and scores of 16 or above indicate
a risk for clinically significant levels of depression.
The CES–D has been widely used in prior studies on
caregiver well-being (see review by Schulz, O’Brien,
Bookwala, & Fleissner, 1995). On average, the sam-
720
The Gerontologist

Table 1. Correlations and Descriptive Data for Total Sample (N ϭ 317)
Measures
1
2
3
4
5
6
7
1 CES–D
—
2 Parent care stress
3 Parent care rewards
4 Parent care mastery
5 Emotional support from parent
6 Emotional undermining from parent
7 Parent care centrality
M
.29*
—
–.15*
–.36*
–.16*
.14*
–.07*
10.50
10.11
–.26*
–.39*
–.35*
.43*
—
.35*
.48*
–.39*
.51*
3.21
0.72
—
.40*
—
–.67*
.27*
13.51
4.28
–.35*
.14*
24.25
4.13
—
–.06*
2.23
0.54
–.17*
10.38
4.17
—
8.42
1.98
SD
Note: CES–D ϭ Center for Epidemiologic Studies–Depression scale.
*p Յ .05.
ple reported a moderate level of depressive symp-
toms (Table 1). Approximately one third (30%) of the
total sample scored at or above 16.
(e.g., Bandura, 1977; Lawton, Kleban, Moss, Rovine,
& Glicksman, 1989; Pearlin, Lieberman, Menaghan,
& Mullan, 1981; Rotter, 1966). Participants evaluated
each item using response categories from 1 “strongly
disagree” to 5 “strongly agree.” The scores were
summed for a possible range from 7 to 35; higher
scores are indicative of greater mastery.
Parent Care Stress.—Fifteen items assessed stress in
the parent care role in the past two months (Stephens
& Townsend, 1997). These items were adapted from
previous caregiving stress research (Albert, 1991;
Kinney & Stephens, 1989a; Vitaliano, Russo, Young,
Becker, & Maiuro, 1991; Zarit & Zarit, 1983). Partici-
pants were asked to rate each item (e.g., “Dealing
with [your] parent’s memory problems”) using a re-
sponse scale ranging from 1 “not at all stressful” to 4
“very much stressful.” Participants were instructed to
indicate any item that was not applicable. Parent
care stress was operationalized by summing the rat-
ings across all 15 items and dividing by the number
of items that were endorsed as applicable (␣ ϭ .81
for African American participants and ␣ ϭ .80 for
White participants). The potential range for parent
care stress is 1 to 4, with higher scores indicating
greater stress.
Quality of the Relationship With the Impaired Par-
ent.—Both positive qualities (i.e., emotional support)
and negative qualities (i.e., emotional undermining)
were assessed (Martire et al., 1998). Five items mea-
sured the frequency with which emotional support
was received from the parent (e.g., “How often did
your [parent] listen carefully to your concerns?”) dur-
ing the past 2 months (␣ ϭ .77 for African American
participants and ␣ ϭ .85 for White participants).
These items were based on previous research on pos-
itive caregiving interactions (e.g., Kinney & Stephens,
1989a; Lawton et al., 1989). The response scale
ranged from 1 “rarely or none of the time” to 4 “most
of the time.” Scores were summed for a possible
range from 5 to 20; higher scores are indicative of
greater emotional support. Emotional undermining
from the parent was defined as negative evaluation,
negative affect, or conflict (Vinokur & Vinokur-Kaplan,
1990). Six items were used to assess the frequency
with which the parent reportedly engaged in emo-
tional undermining (e.g., “How often did your [par-
ent] criticize you?”) in the past 2 months (␣ ϭ .87 for
African American participants and ␣ ϭ .81 for White
participants). These items were based on previous re-
search on negative caregiving interactions (e.g., Kin-
ney & Stephens, 1989a; Lawton et al., 1989; Townsend
& Franks, 1995). Emotional undermining scores (pos-
sible range from 6 to 24) were calculated in the same
manner as emotional support scores, higher scores
indicating greater emotional undermining.
Parent Care Rewards.—A 6-item scale (␣ ϭ .84 for
African American participants and ␣ ϭ .81 for White
participants) was used to assess rewards in the parent
care role in the past 2 months (Stephens & Townsend,
1997). Participants evaluated each item (e.g., “How
rewarding has it been doing things to help your [par-
ent]”) using response categories from 1 “not at all” to
4 “very much.” Scores were calculated in the same
manner as parent care stress (i.e., summing the rat-
ings and dividing by the number of items that were
endorsed as applicable). The potential range for re-
wards also is 1 to 4, with higher scores indicating
greater rewards.
Parent Care Mastery.—A 7-item scale (␣ ϭ .62 for
African American participants and ␣ ϭ .68 for White
participants) was used to assess mastery in the parent
care role in the past 2 months (Christensen, Stephens,
& Townsend, 1998). These items (e.g., “To what ex-
tent do you agree or disagree that you are usually
certain about what to do in caring for your parent?”)
were developed based on previous research on glo-
bal and caregiving mastery, self-efficacy, or control
Parent Care Centrality.—This construct was as-
sessed by a single item (“How important to you is be-
ing a caregiver to your [parent]?”). Scores could range
from 1 “not at all important” to 10 “very important.”
Covariates.—Prior to examining the study hypoth-
eses, a total of 34 background characteristics were
Vol. 40, No. 6, 2000
721

considered as potential covariates, drawn from four
domains: caregivers’ optimism, demographic charac-
teristics of caregivers and their families, demographic
characteristics and impairment of the care recipient,
and characteristics of the caregiving situation. The
selection of potential covariates was based on previ-
ous research on predictors of caregiver well-being
and stress. Optimism was included in order to assess
women’s role experiences net of their generalized
expectancies for positive or negative outcomes in life
(Stephens & Townsend, 1997). Optimism was as-
sessed using the Life Orientation Test (LOT; Scheier
& Carver, 1992). Higher scores are indicative of greater
optimism.
Demographic characteristics of the participants in-
cluded age (in years), kin relationship (0 ϭ daugh-
ters-in-law, 1 ϭ daughters), physical health (a 3-item
index ranging from 3 ϭ poorer health to 15 ϭ better
health), years of education, years married, hours
worked per week, weeks worked in the past year, the
participant’s income (1 ϭ less than $5,000 to 10 ϭ
$80,000 and over), and household income (same
categories).
Children’s characteristics included the number of
children at home, age of the youngest child at home
(in years), and an index assessing the number of 11
tasks with which the child(ren) at home required
help. The husband’s characteristics included age (in
years), physical health (1 ϭ poor to 5 ϭ excellent),
and years of education.
The characteristics of the parent included age (in
years), gender (0 ϭ male, 1 ϭ female), number of
sons and daughters, and physical health (1 ϭ poor to
5 ϭ excellent). Measures of the parent’s impairment
included a rating of an index of the frequency of six
memory problems such as forgetting what day it is (6 ϭ
low to 24 ϭ high impairment), an additive index of
the frequency of five behavioral problems such as be-
coming angry or aggressive (5 ϭ low to 20 ϭ high
impairment), the amount of supervision needed (1 ϭ
none of the time to 5 ϭ most of the time), an index of
the amount of help required with eight IADLs, such
as transportation, meal preparation (8 ϭ low to 24 ϭ
high impairment), and an index of the amount of
help required with seven PADLs, such as bathing, toi-
leting (7 ϭ low to 21 ϭ high impairment).
Characteristics of the caregiving situation included
the number of years participants provided care to a
parent, hours the participants assisted the parent on a
typical weekday and on a typical weekend day,
whether supervision of the parent was provided or ar-
ranged by the participant (0 ϭ no, 1 ϭ yes), the num-
ber of eight IADLs with which the participant helps,
the number of seven PADLs with which the partici-
pant helps, and whether the participant receives help
with caregiving (0 ϭ no paid or unpaid help to 3 ϭ
help from family, friends/neighbors, paid agency).
Results
Background Characteristics
In this section, we present background information
about the sample as a whole, and the results of one-
way analyses of variance (ANOVAs) testing differ-
ences by race in background characteristics.
Table 2. Demographic Characteristics of Participants and Their Families by Participant’s Race
Total Sample African American Women
White Women
Measure
M
SD
M
SD
M
SD
Participant’s Characteristics
Age
43.40
0.86
6.37
1.36
2.36
1.84
8.23
12.02
7.55
2.27
1.36
38.28*
0.87
6.49
0.33
2.03
1.57
7.82
6.14
4.88
1.73
1.06
44.49*
0.86
5.79
0.35
2.43
1.89
8.38
12.65
8.12
2.35
1.41
Kin relationshipa
Physical healthb
Years of education
Years married
11.25
14.61
17.78
37.10
48.96
4.94
11.09
14.78
11.12*
41.36*
50.12
5.67*
8.18
11.29
14.58
19.21*
35.89*
48.63
4.79*
8.39
Hours worked per week
Weeks worked past year
Participant’s incomec
Total household incomec
Children’s Characteristics
Number of children at home
Age of youngest child at home
Children’s functional needsd
Husband’s Characteristics
Age
8.36
2.11
12.32
5.69
1.04
6.62
3.28
2.33
8.75*
7.02*
1.07
6.33
3.69
2.07
13.09*
5.41*
1.04
6.44
3.12
45.47
3.76
14.42
6.89
0.94
2.10
42.45*
3.70
14.11*
8.06
0.97
1.84
46.46*
3.78
14.54*
6.42
0.94
2.16
Physical healthe
Years of education
a0 ϭ daughter-in-law, 1 ϭ daughter.
bSum of three health items: 3 ϭ poorer health to 15 ϭ better health.
c01 ϭ less than $5,000, 02 ϭ $5,000–$9,999, 03 ϭ $10,000–$14,999, 04 ϭ $15,000–$19,999, 05 ϭ $20,000–$24,999, 06 ϭ
$25,000–$29,999, 07 ϭ $30,000–$39,999, 08 ϭ $40,000–$59,999, 09 ϭ $60,000–$79,999, 10 ϭ $80,000 and over.
dNumber of 11 tasks with which child(ren) at home require(s) help.
e1 ϭ poor to 5 ϭ excellent.
*p Յ .05, significant difference by race.
722
The Gerontologist

Optimism.—On average, the total sample reported
experiencing a moderate level of optimism (M ϭ
29.23, SD ϭ 4.92, range ϭ 14–40). There was no dif-
ference by race (for African American women, M ϭ
30.00, SD ϭ 4.70, range ϭ 16–40; for White women,
M ϭ 29.08, SD ϭ 4.96, range ϭ 14–40).
cantly from White women only in providing care for
parents who were approximately 7 years younger,
F(1,316) ϭ 44.82, p Ͻ .01. There were no differences
by race for parents’ impairment.
Characteristics of the Caregiving Situation.—Table 4
reports the characteristics of the caregiving situation
for the total sample along with significant differences
by race. A total of three significant differences by
race were found. African American women provided
almost one more hour of caregiving, on average, on a
typical weekend day, F(1,316) ϭ 11.93, p Ͻ .01;
they helped with slightly more IADLs, F(1,316) ϭ
4.45, p ϭ .03, and slightly more PADLs, F(1,316) ϭ
4.14, p ϭ .04, than White women did.
In summary, African American and White caregiv-
ers differed on 12 background characteristics: partici-
pants’ age, years of marriage, hours per week the par-
ticipants worked, participants’ individual income,
age of youngest child at home, children’s functional
needs, husbands’ age and years of education, care re-
cipients’ age, caregiving hours on weekends, and the
number of IADL and PADL activities with which the
participant helped. These variables were considered
for inclusion as covariates in multivariate analysis.
Demographic Characteristics of Participants and
Their Families.—Table 2 shows the demographic char-
acteristics of participants and their families for the to-
tal sample as well as significant differences by race.
A total of eight differences by race were significant
(Table 2). On average, the African American women
were 6 years younger, F(1,316) ϭ 50.69, p Ͻ .01,
and they were married about 8 years less, F(1,316) ϭ
43.97, p Ͻ .01. African American women worked
approximately five more hours per week, F(1,316) ϭ
11.08, p Ͻ .01, and the amount of income that they
contributed to their households was greater than the
amount contributed by White women, F(1,310) ϭ
6.75, p Ͻ .01. On average, African American women
also differed from White women in that their young-
est child at home was about 4 years younger, F(1,316) ϭ
21.04, p Ͻ .01, and their children at home required
help with approximately two more functional needs,
F(1,316) ϭ 11.42, p Ͻ .01. Finally, the African Amer-
ican women in the study had husbands who were 4
years younger, F(1,316) ϭ 16.36, p Ͻ .01, and
slightly less well educated, F(1,316) ϭ 4.37, p Ͻ .01
than husbands of the White women.
Multivariate Analysis
To investigate differences by race in depression,
stress, rewards, mastery, emotional support, emotional
undermining, and centrality, a one-way between-
subjects multivariate analysis of variance (MANOVA)
was performed (Tabachnick & Fidell, 1996). MANOVA
was chosen because study measures were moder-
ately intercorrelated (Table 1). The race of the care-
giver (African American or White) was the sole pre-
dictor. Prior to MANOVA, Box’s M was used to test
Demographic Characteristics and Impairment of
the Care Recipient.—Table 3 displays the demographic
characteristics and impairment of the care recipient
for the total sample along with significant differences
by race. African American women differed signifi-
Table 3. Demographic Characteristics and Impairment of the Parent by Participant’s Race
Total Sample African American Women
White Women
Measure
M
SD
M
SD
M
SD
Parent’s Characteristics
Age
75.78
0.77
1.42
1.96
7.56
0.42
1.41
1.36
69.93*
0.77
8.24
0.42
1.41
1.39
76.92*
0.77
6.82
0.43
1.41
1.36
Gendera
Number of sons
1.61
1.38
Number of daughters
Parent’s Impairment
Physical healthb
2.09
1.93
2.37
10.75
8.18
3.63
16.71
9.15
1.05
4.04
2.83
1.37
3.58
2.75
2.16
10.18
8.09
3.45
16.77
9.78
0.87
4.00
2.37
1.46
3.65
3.41
2.42
10.88
8.20
3.68
16.71
9.02
1.08
4.04
2.93
1.34
3.57
2.57
Memory impairmentc
Behavioral impairmentd
Supervision needed in typical weeke
IADL impairmentf
PADL impairmentg
Notes: IADL ϭ instrumental activity of daily living; PADL ϭ personal activity of daily living.
a0 ϭ male, 1 ϭ female.
b1 ϭ poor to 5 ϭ excellent.
c6 ϭ low to 24 ϭ high impairment.
d5 ϭ low to 20 ϭ high impairment.
e1 ϭ none of the time to 5 ϭ most of the time.
f8 ϭ low to 24 ϭ high impairment.
g7 ϭ low to 21 ϭ high impairment.
*p Յ .05, significant difference by race.
Vol. 40, No. 6, 2000
723

Table 4. Characteristics of the Caregiving Situation by Participant’s Race
Total Sample African American Women
White Women
Measure
M
SD
M
SD
M
SD
Years caregiving
6.20
5.90
5.86
5.25
6.23
6.21
Caregiving hours
Typical weekday
Typical weekend day
2.74
3.53
0.99
1.58
1.38
1.49
2.83
3.41
0.06
1.71
1.85
0.85
2.23
4.49*
1.00
6.09*
2.03*
1.39
1.90
4.20
0.00
1.59
2.14
0.78
2.85
3.23*
0.99
5.56*
1.48*
1.51
2.98
3.14
0.06
1.73
1.77
0.87
Participant provides supervisiona
Number of 8 IADLs that participant helps with
Number of 7 PADLs that participant helps with
Help from other caregiversb
Notes: IADLs ϭ instrumental activities of daily living; PADLs ϭ personal activities of daily living.
a0 ϭ no, 1 ϭ yes.
b0 ϭ no paid or unpaid help to 3 ϭ help from family, friends/neighbors and paid agency.
*p Յ .05, significant difference by race.
the assumption of homogeneity of variance–covari-
ance matrices (Tabachnick & Fidell, 1996). This test
revealed no significant difference between groups,
Box’s M ϭ 27.35, F(28,34501) ϭ 0.93, p ϭ .57. Om-
nibus MANOVA results indicated that the parent care
experiences of African American and White women
were significantly different, Wilks’s Lambda ϭ .94,
F(7,309) ϭ 3.02, p Ͻ .001, effect size ϭ .06.
being related to higher symptomatology) and the hus-
band’s education (r ϭ –.11, lower spouse education
being related to higher caregiver symptomatology).
Greater parent care stress was correlated with the
caregiver providing help with a greater number of
IADLs (r ϭ .14). Parent care rewards were not signifi-
cantly correlated with any of the background charac-
teristics. Greater parent care mastery was related to
the caregiver’s children needing less help (r ϭ –.13).
Emotional support from the parent was not signifi-
cantly related to any of the covariates. Emotional un-
dermining from the parent was correlated with the
parent’s age (r ϭ –.11, more undermining being asso-
ciated with younger parents) and the number of
hours spent providing care on a typical weekend day
(r ϭ –.12, more undermining being associated with
fewer hours). Greater parent care centrality was asso-
ciated with fewer years of marriage (r ϭ –.11) and
more hours spent providing care on a typical week-
end day (r ϭ .14).
In order to determine which outcomes contributed
significantly to the difference between groups, Roy-
Bargmann stepdown F tests were performed (Tabach-
nick & Fidell, 1996). To reduce the likelihood of
Type I error for the stepdown tests, the probability
level was adjusted using a correction procedure rec-
ommended by Stevens (1996): namely, a one-tailed
probability level (␣ ϭ .10), because of the study’s di-
rectional hypotheses, was divided by the number of
hypothesized differences (seven) to yield ␣ ϭ .014 as
the criterion for the statistical significance of race on
each of the stepdown tests. Stepdown F tests require
prioritizing the outcomes, then conducting a univari-
ate ANOVA to test between-group differences on the
first outcome, followed by an analysis of covariance
(ANCOVA) for each subsequent outcome (control-
ling for all prior outcomes). Thus, these stepdown
tests use the correlations between outcomes to con-
trol in later analyses for shared variance already ac-
counted for in earlier analyses (Tabachnick & Fidell,
1996). For the present study, outcomes were priori-
tized according to the number of previous studies
pertaining to differences by race, resulting in the fol-
lowing order: depression, stress, rewards, mastery,
emotional support, emotional undermining, and cen-
trality.
As shown in Table 5, the stepdown F tests re-
vealed a significant difference by race (p Յ .014) on
two of the seven study outcomes. As hypothesized,
African American women reported not only a lower
level of stress than White women did but also a
higher level of rewards in the parent care role. The
adjusted means (Table 5) and the effect sizes (Cohen,
2
1988) for stress (f ϭ .15, ␩ ϭ .02) and rewards (f ϭ
.17, ␩2 ϭ .03) indicated that these differences by race
were of small magnitude. Albeit small, these differ-
ences by race are notable, because they were statisti-
cally significant even after controlling for depression
and IADL assistance in the case of parent care stress
and even after controlling for depression and stress in
the case of parent care rewards.
To ensure that differences attributed to race were
not due to other characteristics, the stepdown tests
also included background covariates, chosen from
the 12 characteristics on which significant race dif-
ferences were found. A background characteristic
was included as a covariate in the stepdown analysis
for a particular outcome if it had a significant (p Յ
.05) bivariate correlation with that outcome. Depres-
sive symptomatology was correlated with the partici-
pant’s individual income (r ϭ –.16, lower income
Contrary to prior research, there was no significant
difference by race for depressive symptomatology.
Because of the discrepancy with prior results, a chi-
square analysis was conducted to further explore the
relationship between depressive symptomatology and
race of the caregiver. For this analysis, CES–D scores
were dichotomized using a score of 16 or greater to
represent risk of clinically significant depression (Rad-
loff, 1977). A quarter (25%) of the African American
724
The Gerontologist

Table 5. Results of Stepdown Tests
African American Women White Women
CI
Measure
M
M
CI
F
df
p
Depressiona
11.22
2.07
8.50–13.95
1.93–2.20
12.86
2.27
11.62–14.09
2.21–2.33
1.14
7.17
9.13
0.00
0.14
0.93
0.17
1,307
1,313
1,313
1,311
1,311
1,308
1,307
.286
.008
.003
.984
.712
.336
.682
Parent care stressb
Parent care rewardsc
Parent care masteryd
Emotional supporte
Emotional underminingf
Parent care centralityg
3.38
3.20–3.57
3.07
2.99–3.16
24.26
13.68
10.77
8.32
23.29–25.22
12.72–14.63
9.91–11.64
7.83–8.81
24.25
13.48
10.30
8.44
23.81–24.68
13.04–13.91
9.93–10.67
8.23–8.65
Notes: M ϭ adjusted means; CI ϭ confidence interval; F tests are shown for the main effect of race. Degrees of freedom (df ) vary due
to missing data.
aCovariates were caregiver’s individual income, F(1,307) ϭ 5.85, p ϭ 0.16, and husband’s education, F(1,307) ϭ 3.34, p ϭ .069.
bCovariates were depression, F(1,313) ϭ 25.34, p Ͻ .001, and number of IADL tasks with which caregiver assists, F(1,313) ϭ 6.42, p ϭ
.012.
cCovariates were depression, F(1,313) ϭ 1.59, p ϭ .208, and parent care stress, F(1,313) ϭ 14.20, p Ͻ .001.
dCovariates were depression, F(1,311) ϭ 24.02, p Ͻ .001, parent care stress, F(1,311) ϭ 23.28, p Ͻ .001, parent care rewards,
F(1,311) ϭ 22.45, p Ͻ .001, and children’s functional needs, F(1,311) ϭ 4.10, p ϭ .044.
eCovariates were depression, F(1,311) ϭ 0.24, p ϭ .627, parent care stress, F(1,311) ϭ 11.35, p Ͻ .001, parent care rewards, F(1,311) ϭ
50.33, p Ͻ .001, and parent care mastery, F(1,311) ϭ 15.05, p Ͻ .001.
fCovariates were depression, F(1,308) ϭ 0.31, p ϭ .578, parent care stress, F(1,308) ϭ 23.76, p Ͻ .001, parent care rewards, F(1,308) ϭ
1.89, p ϭ .170, parent care mastery, F(1,308) ϭ 0.22, p ϭ .640, emotional support, F(1,308) ϭ 131.53, p Ͻ .001, parent’s age, F(1,308) ϭ
1.56, p ϭ .213, and caregiving hours on weekends, F(1,308) ϭ 1.59, p ϭ .208.
gCovariates were depression, F(1,307) ϭ 0.22, p ϭ .641, parent care stress, F(1,307) ϭ 1.55, p ϭ .214, parent care rewards, F(1,307) ϭ
77.29, p Ͻ .001, parent care mastery, F(1,307) ϭ 0.35, p ϭ .551, emotional support, F(1,307) ϭ 2.24, p ϭ .135, emotional undermining,
F(1,307) ϭ 0.26, p ϭ .613, caregiving hours on weekends, F(1,307) ϭ 9.72, p ϭ .002, and years married, F(1,307) ϭ 1.76, p ϭ .186.
caregivers and a third (33%) of the White caregivers
scored above this cutoff. Although in the direction
predicted by prior research, the difference was not
statistically significant, ␹2(1) ϭ 1.74, p ϭ .19.
Similarly, no significant difference by race was
found for parent care mastery, emotional support, or
emotional undermining from the parent, or parent
care centrality. Each of these measures did show sig-
nificant relationships with one or more of the other
outcomes under consideration, however. Thus, care-
giving experiences were substantially interrelated,
but only stress and rewards varied by race.
care stress. These confounding experiences have not
been taken into account in previous research. It is
also important to note that the higher level of rewards
African American women reported receiving from care-
giving did not merely reflect a more positive outlook
on life generally, because there was no difference by
race on optimism.
Differences in values and beliefs about aging, role
expectations, and religiosity may account for why Af-
rican American caregivers report less parent care stress
and more parent care rewards than White caregivers
do. Dilworth-Anderson and Anderson (1994) theo-
rize that African American caregivers tend to “nor-
malize” the cognitive and/or behavioral problems of
older adults and interpret their symptoms of impair-
ment as an integral part of the aging process. African
Americans also express more positive attitudes to-
ward their elders, viewing aging as a source of dig-
nity and something to be respected (Haley et al.,
1996; Mutran, 1985).
Prior research has also shown that African Ameri-
can caregivers are more likely than White caregivers
to report that caregiving is an expected experience
and that caregiving responsibilities are less of an in-
trusion on their lives (Haley et al., 1996; Lawton et
al., 1992). African American families generally ac-
cept that family members should care for an aging
relative (Dilworth-Anderson & Anderson, 1994; Mc-
Adoo, 1993), and being able to fulfill the caregiving
role has been found to be a source of rewards for Af-
rican American caregivers (Picot, 1994). For African
American caregivers, prayer and faith may raise their
threshold for the stresses of caregiving (Segall &
Wykle, 1988; Wheaton, 1985), may act as a buffer
when caregiving stresses arise, and may be associ-
Discussion
As predicted, African American women reported a
lower level of parent care stress than White women
reported. This finding is consistent with most of the
previous research on family caregivers (Hinrichsen &
Ramirez, 1992; Lawton et al., 1992; Macera et al.,
1992; Miller et al., 1995; Mui, 1992). Both the present
study and these previous studies found significant dif-
ferences by race in caregiving stress even after con-
trolling for a variety of background characteristics.
Only three prior studies (Cox, 1993; Morycz et al.,
1987; Wood & Parham, 1990) have not found a dif-
ference by race. Thus, differences between African
American and White caregivers in caregiving stress
appear to be robust.
The present study also found that African Ameri-
can women reported a higher level of parent care re-
wards than White women did. This is consistent with
prior studies (Lawton et al., 1992; Picot et al., 1997).
However, in the present study, the analysis of parent
care rewards controlled for depression and parent
Vol. 40, No. 6, 2000
725

ated with perceptions of caregiving rewards, such as
being blessed by God for their caregiving efforts (Pi-
cot et al., 1997).
geneity of caregivers’ kin relationship (i.e., daughters
and daughters-in-law in the present study and spouses
in Miller et al.,1995). In contrast, Lawton and col-
leagues (1992) aggregated a variety of caregivers
(e.g., spouses, adult children, siblings), and the distri-
butions across kinship categories differed by race.
We found no significant difference by race for
emotional support or emotional undermining from
the parent when controlling for factors that were re-
lated to these two measures. The association be-
tween caregivers’ race and the quality of their rela-
tionship with care recipients has received limited
attention in previous research. The present study also
provided an exploratory analysis of the association
between race and parent care centrality. To our
knowledge, this is the first investigation of this topic.
Parent care rewards and the number of hours spent
caregiving on a typical weekend day predicted par-
ent care centrality, but race did not.
Contrary to expectation, there was no significant
difference by race for depressive symptomatology in
our study, despite African American women report-
ing less stress and more rewards from parent care
than White women reported. The present finding was
surprising. With one exception (Young & Kahana,
1995), prior research has consistently shown that Af-
rican American caregivers report significantly lower
depressive symptomatology than White caregivers
(Haley et al., 1995; Lawton et al., 1992; Miller et al.,
1995; Mintzer & Macera, 1992).
One possible explanation for the discrepancy be-
tween the present study and prior studies is the na-
ture of the samples. Most of the previous studies ag-
gregated a variety of family (and sometimes nonfamily)
caregivers, whereas the present study focused on a
more homogenous sample of caregivers (adult daugh-
ters and daughters-in-law). The one study that also
used a homogenous sample of caregivers but found a
significant difference in depression by race examined
spouse caregivers (Miller et al., 1995).
The care recipient’s impairment might also account
for the inconsistency in findings. All of the prior stud-
ies that found differences in depressive symptomatol-
ogy sampled caregivers who were assisting care re-
cipients suffering from some form of dementia (Haley,
1997; Haley et al., 1996; Hinrichsen & Ramirez,
1992; Lawton et al., 1992; Miller et al., 1995; Mint-
zer & Macera, 1992). In contrast, the present study
and Young and Kahana (1995), both of which did not
find a significant difference in depressive symptoma-
tology by race, used samples that were not limited to
dementia care recipients. The present study focused
on care recipients with a broad range of impairment,
physical and/or cognitive. Young and Kahana (1995)
focused on caregivers to persons with heart disease.
Thus, differences in caregivers’ depressive symptom-
atology by race may only occur with specific forms
of impairment such as dementia or when physical
and/or cognitive impairment reaches some critical
threshold.
Limitations of the Study
There are several noteworthy limitations to the
present study. The results may not generalize to other
ethnic groups or to women with different sociodemo-
graphic characteristics. The median income for White
families in the present study was considerably higher
than the national median for White married dual-
earner couples (Bennett, 1995), and both the African
American and White women in the present sample
were substantially better educated than women na-
tionally (Bennett, 1995). Also, the results of the present
study are cross-sectional. Therefore, we do not know
whether differences by race in parent care stress and
rewards will persist or whether other caregiving ex-
periences (i.e., depressive symptoms) will continue to
be unrelated to race over time. The longitudinal de-
sign of our study will enable us to address these ques-
tions in future analyses. Measures that could explain
why the observed differences by race occurred were
not available for the present study. In order to assess
caregivers’ role expectations, religiosity, and attitudes
about aging, it is recommended that future research
make an effort to include such measures.
Another possible explanation for the discrepancy
in findings may be that caregivers’ income was a sig-
nificant predictor of depressive symptomatology in
our analyses. This finding is consistent with other lit-
erature linking income and depressive symptomatol-
ogy (e.g., Eaton & Kessler, 1981). It may be that ear-
lier differences attributed to race may have been
confounded with socioeconomic factors.
The present study also found no significant differ-
ence by race in parent care mastery, which was con-
sistent with one prior study (Miller et al., 1995) and
inconsistent with another (Lawton et al., 1992). Dif-
ferences in operationalization of mastery and in care
recipients’ impairment are unlikely to account for the
inconsistency in findings, given that the studies by
Lawton and colleagues (1992) and Miller and associ-
ates (1995) closely resembled each other in those re-
gards. A more plausible explanation is greater homo-
Strengths of the Study
Despite these limitations, the present study con-
tributes to research on caregiving and race in several
important ways. Our study controlled for caregivers’
kin relationship (daughters and daughters-in-law) as a
means to limit potential confounds between race and
kin relationship. Although research has shown that
caregiving outcomes differ between adult child and
spouse caregivers (e.g., George & Gwyther, 1986),
prior studies investigating the caregiving experiences
of African Americans and Whites have aggregated a
variety of caregivers. Our research design also re-
quired participants to occupy three roles (mother,
wife, and employee) in addition to that of primary
caregiver. The relationship between multiple roles
and caregiving experiences has not been taken into
account in previous research on African American
726
The Gerontologist

Fiore, J., Becker, J., & Coppel, D. (1983). Social network interactions: A
buffer or a stress. American Journal of Community Psychology, 1, 67–82.
George, L. K., & Gwyther, L. (1986). Caregiver well-being: A multidimen-
sional examination of family caregivers of demented adults. The
Gerontologist, 26, 253–259.
Gurin, G., Veroff, J., & Feld, S. (1960). Americans view their mental health:
A nationwide interview survey. New York: Basic Books.
and White caregivers. The present study was able to
control for the effect that other roles may have on the
caregiving experience.
Many studies have investigated the experiences of
caregivers to relatives with Alzheimer’s disease or
other forms of dementia (e.g., Haley, Clair, & Sauls-
berry, 1992; Lawton et al., 1992). Results from de-
mentia samples may not be generalizable to individ-
uals who provide care to relatives without cognitive
impairment (Ory et al., 1999). The present study ex-
tends previous research by examining the experi-
ences of African American and White caregivers who
were providing care for aging parents with various
forms of impairment.
Haley, W. E. (1997). The family caregiver’s role in Alzheimer’s disease.
Neurology, 48, S25–S29.
Haley, W. E., Bartolucci, A. A., Levine, E. G., & Brown, S. L. (1987). Stress,
appraisal, coping, and social support as predictors of adaptational out-
come among dementia caregivers. Psychology and Aging, 2, 323–330.
Haley, W. E., Clair, J. M., & Saulsberry, K. (1992). Family caregiver satis-
faction with medical care of their demented relatives. The Gerontologist,
32, 219–226.
Haley, W. E., Levine, E. G., Brown, S. L., Berry, J. W., & Hughes, G. H.
(1987). Psychological, social, and health consequences of caring for a
relative with senile dementia. Journal of the American Geriatrics Soci-
ety, 35, 405–411.
Research on caregiving stress and race is limited,
but even fewer studies have examined the positive
aspects of caregiving by race. Similarly, research on
the quality of social relationships has focused more
often on social support than on social undermining
(Rook, 1991). A considerable strength of the present
study is that it examined potential differences by race
in both negative and positive aspects of the caregiv-
ing experience. Constructs such as parent care mas-
tery and parent care centrality have also received
limited attention in prior research investigating care-
giving experiences by race. Overall, this research
demonstrated the importance of examining a broad
range of caregiving experiences in order to detect
both similarities and differences between African
American and White caregivers.
Haley, W. E., Roth, D. L., Coleton, M. I., Ford, G. R., West, C. A., Collins,
R. P., & Isobe, T. L. (1996). Appraisal, coping, and social support as
mediators of well-being in Black and White family caregivers of pa-
tients with Alzheimer’s disease. Journal of Consulting and Clinical
Psychology, 64, 121–129.
Haley, W. E., West, C. A., Wadley, V. G., Ford, G. R., White, F. A., Barrett,
J. J., Harrell, L. E., & Roth, D. L. (1995). Psychological, social, and
health impacts of caregiving: A comparison of Black and White de-
mentia family caregivers and noncaregivers. Psychology and Aging,
10, 540–552.
Hinrichsen, G. A., & Ramirez, M. (1992). Black and White dementia care-
givers: A comparison of their adaptation, adjustment, and service
utilization. The Gerontologist, 32, 375–381.
Kinney, J. M., & Stephens, M. A. (1989). Hassles and uplifts of giving care
to a family member with dementia. The Gerontologist, 29, 402–408.
Lawton, M. P., Kleban, M. H., Moss, M., Rovine, M., & Glicksman, A.
(1989). Measuring caregiving appraisal. Journal of Gerontology: Psy-
chological Sciences, 44, P61–P71.
Lawton, M. P., Moss, M., Kleban, M. H., Glicksman, A., & Rovine, M.
(1991). A two-factor model of caregiving appraisal and psychological
well-being. Journal of Gerontology: Psychological Sciences, 46, P181–
P189.
Lawton, M. P., Rajagopal, D., Brody, E., & Kleban, M. H. (1992). The dy-
namics of caregiving for a demented elder among Black and White
families. Journal of Gerontology: Social Sciences, 47, S156–S164.
Macera, C. A., Eaker, E. D., Goslar, P. W., Deandrade, S. J., Williamson,
J. S., Cornman, C., & Jannarone, R. J. (1992). Ethnic differences in the
burden of caregiving. American Journal of Alzheimer’s Disease. Sep-
tember/October, 4–7.
Martire, L. M., Stephens, M. A., & Townsend, A. L. (1998). Emotional sup-
port and well-being in midlife women: Role specific mastery as a
mediational mechanism. Psychology and Aging, 13, 396–404.
McAdoo, H. P. (1993). Family ethnicity: Strength in diversity. Newbury
Park, CA: Sage Publications.
Miller, B., Campbell, R. T., Farran, C. J., Kaufman, J. E., & Davis, L. (1995).
Race, control, mastery, and caregiver distress. Journal of Gerontology:
Social Sciences, 50B, S374–S382.
References
Albert, S. M. (1991). Cognition of caregiving tasks: Multidimensional scal-
ing of the caregiver task. The Gerontologist, 31, 726–734.
Aranda, M. P., & Knight, B. G. (1997). The influence of ethnicity and cul-
ture on the caregiver stress and coping process: A socio-cultural review
and analysis. The Gerontologist, 37, 342–354.
Aneshensel, C. S., Pearlin, L. I., Mullan, J. T., Zarit, S. H., & Whitlatch, C. J.
(1995). Profiles of caregiving: The unexpected career. San Diego, CA:
Academic Press.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral
change. Psychological Review, 84, 191–215.
Bennett, C. E. (1995). The Black population in the United States, March
1994 and 1993. (U.S. Bureau of the Census, Current Population Re-
ports, P20-480). Washington, DC: U.S. Government Printing Office.
Christensen, K. A., Stephens, M. A., & Townsend, A. L. (1998). Mastery in
women’s multiple roles and well-being: Adult daughters providing care
to impaired parents. Health Psychology, 17, 163–171.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Erlbaum.
Coppel, D., Burton, C., Becker, J., & Fiore, J. (1985). Relationships of cog-
nitions associated with coping reactions to depression in spousal
caregivers of Alzheimer’s disease patients. Cognitive Therapy & Re-
search, 9, 253–266.
Mintzer, J. O., & Macera, C. A. (1992). Prevalence of depressive symptoms
among White and African American caregivers of demented patients.
American Journal of Psychiatry, 149, 575–576.
Morycz, R. K., Malloy, J., Bozich, M., & Martz, P. (1987). Racial differ-
ences in family burden: Clinical implications for social work.
Gerontological Social Work With Families, 10, 133–154.
Mui, A. C. (1992). Caregiver strain among Black and White daughter care-
givers: A role theory perspective. The Gerontologist, 32, 203–212.
Mutran, E. (1985). Intergenerational family support among Blacks and
Whites: Response to culture or to socioeconomic differences. Journal
of Gerontology, 40, 382–389.
Cox, C. (1993). Service needs and interests: A comparison of African
American and white caregivers seeking Alzheimer assistance. Ameri-
can Journal of Alzheimer’s Disease, May/June, 33–40.
Ory, M. G., Hoffman, R., III, Yee, J. L., Tennstedt, S., & Schulz, R. (1999).
Prevalence and impact of caregiving: A detailed comparison between
dementia and nondementia caregivers. The Gerontologist, 39, 177–
185.
Pagel, M., Becker, J., & Coppel, D. (1985). Loss of control, self-blame, and
depression: An investigation of spouse caregivers of Alzheimer’s dis-
ease patients. Journal of Abnormal Psychology, 62, 169–182.
Pearlin, L. I., Lieberman, M. A., Menaghan, E. G., & Mullan, J. T. (1981).
The stress process. Journal of Health and Social Behavior, 22, 337–356.
Pearlin, L. I., Mullan, J. T., Semple, S. J., & Skaff, M. M. (1990). Caregiving
and the stress process: An overview of the concepts and their mea-
sures. The Gerontologist, 30, 583–591.
Creasey, G., Myers, B., Epperson, M., & Taylor, J. (1990). Couples with an
elderly parent with Alzheimer’s disease: Perceptions of familial rela-
tionships. Psychiatry, 53, 44–51.
Deimling, G., Bass, D., Townsend, A. L., & Noelker, L. (1989). Care-related
stress: A comparison of spouse and adult-child caregivers in shared and
separate households. Journal of Aging and Health, 1, 67–82.
Dilworth-Anderson, P., & Anderson, N. (1994). Dementia caregiving in
Blacks: A contextual approach to research. In E. Light, G. Niederehe, &
B. D. Lebowitz (Eds.), Stress effects on family caregivers of Alzheimer’s
patients (pp. 385–409). New York: Springer.
Dura, J. R., Stukenberg, K. W., & Kiecolt-Glaser, J. K. (1991). A compari-
son of caregivers for elderly stroke and dementia victims. Psychology
and Aging, 6, 467–473.
Eaton, W., & Kessler, L. (1981). Rates of depression in a national sample.
American Journal of Epidemiology, 114, 528–538.
Picot, S. J. (1994). Choice and social exchange theory and the rewards of Af-
rican American caregivers. Journal of National Black Nurses, 7, 29–40.
Picot, S. J., Debanne, S. M., Namazi, K. H., & Wykle, M. L. (1997). Religi-
Vol. 40, No. 6, 2000
727

osity and perceived rewards of Black and White caregivers. The
Gerontologist, 37, 89–101.
Radloff, L. S. (1977). The CES–D scale: A self-report depression scale for
research in the general population. Applied Psychological Measure-
ment, 1, 385–401.
Rook, K. (1991). Detrimental aspects of social relationships: Taking stock
of an emerging literature. In H. Veiel & U. Baumann (Eds.), The mean-
ing and measurement of social support (pp. 157–169). New York:
Hemisphere.
Townsend, A. L., Noelker, L., Deimling, G., & Bass, D. (1989). Longitudi-
nal impact of interhousehold caregiving on adult children’s mental
health. Psychology and Aging, 4, 393–401.
U.S. Bureau of the Census. (1992). International population reports. Aging
world II (Publication No. P25, 92-3). Washington, DC: U.S. Govern-
ment Printing Office.
U.S. Bureau of the Census. (1996). Current population reports, special
studies. Sixty-five plus in the United States (Publication No. P-23–190).
Washington, DC: U.S. Government Printing Office.
Rotter, J. B. (1966). Generalized expectancies for internal versus external
control of reinforcement. Psychological Monographs: General & Ap-
plied, 80, 1–28.
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological
and physical well-being. Cognitive Therapy & Research, 16, 201–228.
Schulz, R., O’Brien, A. T., Bookwala, J., & Fleissner, K. (1995). Psychiatric
and physical morbidity effects of dementia caregiving: Prevalence, cor-
relates, and causes. The Gerontologist, 35, 771–791.
Schulz, R., Tompkins, C. A., & Rau, M. T. (1988). A longitudinal study of
the psychosocial impact of stroke on primary support person. Psychol-
ogy and Aging, 3, 131–141.
Segall, M., & Wykle, M. (1988). The Black family’s experience with de-
mentia. Journal of Applied Social Sciences, 13, 170–190.
Stephens, M. A., & Townsend, A. L. (1997). The stress of parent care: Posi-
tive and negative effects of women’s other roles. Psychology and
Aging, 12, 376–386.
Stevens, J. (1996). Applied multivariate statistics for the social sciences
(3rd ed.). Mahwah, NJ: Erlbaum.
Stryker, S., & Serpe, R. T. (1994). Identity salience and psychological cen-
trality: Equivalent, overlapping, or complementary concepts? Social
Psychology Quarterly, 57, 16–35.
Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd
ed.). New York: HarperCollins.
Vinokur, A., & Vinokur-Kaplan, D. (1990). “In sickness and in health”: Pat-
terns of social support and undermining in older married couples.
Journal of Aging and Health, 2, 215–241.
Vitaliano, P. P., Russo, J., Young, H., Becker, J., & Maiuro, R. (1991). The
screen for caregiver burden. The Gerontologist, 31, 76–83.
Wheaton, B. (1985). Models for the stress-buffering function of coping re-
sources. Journal of Health and Social Behavior, 26, 352–364.
Wood, J. B., & Parham, I. A. (1990). Coping with perceived burden: Ethnic
and cultural issues in Alzheimer’s family caregiving. Journal of Applied
Gerontology, 9, 325–339.
Wykle, M., & Kaskel, B. (1991). Increasing the longevity of minority older
adults through improved health status. In Minority elders: Longevity,
economics, and health (pp. 24–31). Washington, DC: The Gerontolog-
ical Society of America.
Young, R. F., & Kahana, E. (1995). The context of caregiving and well-
being outcomes among African American and Caucasian Americans.
The Gerontologist, 35, 225–232.
Zarit, S. H., Reever, K. E., & Bach-Peterson, J. (1980). Relatives of the im-
paired elderly: Correlates of feeling of burden. The Gerontologist, 20,
649–655.
Zarit, S. H., & Zarit, J. M. (1983). The burden interview. University Park,
PA: The Pennsylvania State University.
Thoits, P. A. (1992). Identity structures and psychological well-being: Gender
and marital comparisons. Social Psychology Quarterly, 55, 236–256.
Townsend, A. L., & Franks, M. M. (1995). Binding ties: Closeness and con-
flict in adult children’s caregiving relationships. Psychology and Aging,
10, 343–351.
Received September 3, 1999
Accepted May 23, 2000
Decision Editor: Vernon L. Greene, PhD
Please mention The Gerontologist when replying to advertisements.
728
The Gerontologist