Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Hispanic culture, acculturation, and distress among caregivers of dementia patients
(USC Thesis Other)
Hispanic culture, acculturation, and distress among caregivers of dementia patients
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
HISPANIC CULTURE, ACCULTURATION, AND DISTRESS AMONG
CAREGIVERS OF DEMENTIA PATIENTS
by
Gia Sophie Robinson
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)
December 2000
Copyright 2000 Gia Sophie Robinson
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1407922
___ ®
UMI
UMI Microform 1407922
Copyright 2002 by ProQuest information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
The Graduate School
University Park
LOS ANGELES, CALIFORNIA 90089-1695
This thesis, w ritten b y
6 lf t SD?W\& ROfcuviSoiv]
U nder th e direction o f h.f.L.. Thesis
Com m ittee, and approved b y a ll its m embers,
has been p resen ted to an d accepted b y The
Graduate School, in p a rtia l fulfillm en t o f
requirem ents fo r th e degree o f
Dean o f Graduate Studies
D a te December 1 8 , .2000
THESIS COMMITTEE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Acknowledgments
The author was supported in part by a National Institute on Aging
multidisciplinary training grant (V. Bengtson, director). The members of the thesis
committee, Bob Knight, Margaret Gatz, and Jo Ann Farver are acknowledged for
their useful comments in earlier drafts of this manuscript.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ill
Table of Contents
Acknowledgments.................................................................................................................ii
List of Tables........................................................................................................................ iv
Abstract................................................................................................................................... v
Introduction............................................................................................................................. 1
M ethod...................................................................................................................................15
Results................................................................................................................................... 24
Discussion.............................................................................................................................45
References............................................................................................................................. 54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV
List of Tables
Table 1: Descriptive Statistics for the Sample............................................................... 26
Table 2: Descriptive Statistics for the Variables........................................................... 27
Table 3: Intercorrelations between Variables.................................................................30
Table 4: Summary of Hierarchical Regression Analysis Predicting CES-D
Scores when Controlling for Acculturation..................................................................... 32
Table 5: Summary of Hierarchical Regression Analysis Predicting Log of
CES-D Scores when Controlling for Acculturation.......................................................32
Table 6: Summary of Hierarchical Regression Analysis Predicting Burden
when Controlling for Acculturation.....................................................................
Table 7: Summary of Hierarchical Regression Analysis Predicting CES-D
Scores when Controlling for Generation..........................................................................34
Table 8: Summary of Hierarchical Regression Analysis Predicting Log of
CES-D scores when Controlling for Generation............................................................ 34
Table 9: Summary of Hierarchical Regression Analysis Predicting Burden
when Controlling for Generation.......................................................................................35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
Abstract
There is little research on the influence of ethnicity and culture in explaining
caregiver distress. This study examined which elements of Hispanic culture were
associated with increasing acculturation, and which cultural variables, if any,
predicted depressed mood and burden among Hispanic caregivers of dementia
patients. Forty-eight caregivers were interviewed using a survey questionnaire that
assessed acculturation, familism, marianism, fatalism, dissatisfaction with family
support, and distress. Results indicated that only familism was significantly
associated with Anglo and Hispanic cultural identifications. No cultural values were
significantly correlated with depressive symptoms, however, familism was
significantly associated with lesser burden. Dissatisfaction with family support
received was associated with higher depression and burden scores. Research efforts
should focus on developing interventions for Hispanic caregivers that help them
reconnect with their familism values and increase family support.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
Introduction
Early multivariate models of the stress and coping process have neglected the
influence of ethnicity and culture in explaining caregiver distress. Researchers have
suggested sociocultural factors that might influence the stress and coping process of
Latino caregivers, yet little empirical investigation has been undertaken.
Consequently, there is a need to examine which sociocultural factors play a
significant role in the stress and coping process of Latino caregivers.
In general, stress and coping models of caregiver distress encompass the
following variables: a) background and contextual variables such as age, gender,
socioeconomic status, relationship of the caregiver to the patient, etc.; b) primary
and secondary stressors as a direct effect of caring for the impaired person and
consequently spilling over into other domains of the caregiver’s life such as work
and family; c) the caregiver’s appraisal of demands as satisfying or stressful; d) the
mediators of burden such as coping and social support; and e) the consequences of
those demands such as quality of life, and physical and mental health (Lawton,
Moss, Kleban, Glicksman, & Rovine, 1991; Pearlin, Mullan, Semple, & Skaff, 1990;
Poulshock & Deimling, 1984; Schulz, Tompkins, & Rau, 1988; Zarit, 1994).
Although these multivariate models seem comprehensive, they fail to include the
influence of ethnicity and culture in the stress and coping process of ethnic minority
caregivers. Aranda and Knight (1997) indicated that several factors inherent in the
ethnicity and culture of Latino caregivers such as higher care recipient frailty,
emphasis on familism, and distinctive social support networks may play a significant
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
role in the stress and coping process of Latino caregivers. Thus, sociocultural factors
should be incorporated in these theoretical models in order to understand the
experience of caregiving among ethnic caregivers.
Several sociocultural factors such as acculturation, familism, fatalism, and
marianism have been examined in the Hispanic literature. These factors have been
defined as follows. Acculturation has been described as choosing a mode of
adaptation when dealing with issues of whether one’s cultural identity is of value and
should be retained, and whether positive relations with the larger society should be
sought (Berry, 1980). Thus, acculturation has been viewed as multidimensional
whereby individuals can be independently involved in both Hispanic and Anglo
cultures. Consistent with this multidimensional view of acculturation, Berry (1980)
proposes four modes of adaptation that an individual can choose: assimilation,
integration, separation, and marginalization. Assimilation involves relinquishing
one’s cultural identity and accepting the one held by the host country. Integration is
maintaining one’s identity and becoming part of the larger societal framework.
Separation means having no desire to relate to the host society but wanting to
maintain one’s traditions and culture. Marginalization involves losing contact with
both the dominant and the traditional group.
Now that the four modes of acculturation have been explained, the definitions
of the other sociocultural factors follows. Familism refers to the family representing
the central core of life or “a strong identification and attachment of individuals with
their families (nuclear and extended), and strong feelings of loyalty, reciprocity and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
solidarity among members of the same family” (Sabogal, Marin, Otero-Sabogal,
Marin & Perez-Stable, 1987). Three separate dimensions of familism have been
identified: 1) family obligations (the individual’s perceived obligation to provide
material and emotional support to the family), 2) perceived support from the family
(the perceptions of family members as reliable providers of help and support to solve
problems), and 3) the family as a referent behaviorally and attitudinally (Sabogal et
al., 1987). Fatalism is the belief that individuals cannot control or prevent adverse
events if they are the will of God or powerful external agents (Rothman, Gant &
Hnat, 1985; Vega, 1990). Marianism comprises specific sex role expectations
whereby women are expected to be submissive, docile and sentimental, to take care
of children at home, and to devote their time to activities for the benefit of their
children and husband while the men are expected to be dominant and strong
(Paniagua, 1994).
Now that the sociocultural factors have been defined, the literature on these
factors will be discussed. In terms of research on familism among Hispanics,
Sabogal et al. (1987) in their comparative study of non-elderly Latinos and non-
Latino Whites found perceived support from the family changed the least with
acculturation as measured by the Marin scale in comparison to the other two
dimensions of familism, family obligations and family as a referent among Latinos.
Thus, values such as family obligations and family as a referent decreased in
importance with increased acculturation and exposure to U.S. culture, while
perceived support from the family remained important among Latinos. Several
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cn
4
reasons have been given to explain why perceived support from the family remained
strong despite acculturation. One reason could have been that “the knowledge that
problems could be jointly solved with family members and that support could always
be found among relatives was a much valued worldview that individuals would want
to hold on to even as they acculturate” (Sabogal et al., 1987). Another reason was
that the more integrated Hispanic family network system might have been more
supportive than the networks found among white non-Hispanics (Sabogal et al.,
1987). Thus, perceived support from the family was an important value that
Hispanics would want to maintain despite of acculturation.
Research also showed that fatalism and marianism were salient in Hispanic
culture in addition to familism values. Castro, Furth and Karlow (1984) noted that
less acculturated women of Mexican origin reported a lower sense of responsibility
and control over their own health and were characterized by a greater acceptance of
powerful external agents, God, chance, and the doctor as determinants of health
outcomes. Therefore, fatalism predominated in Hispanic culture and among those
with lesser acculturation.
Furthermore, there existed gender differences in fatalism, which could be
accounted for by Hispanic culture’s emphasis on traditional sex roles. Older
Mexican American women were more likely to feel they had little control over their
health, whereas older men were more likely to feel they had a great deal of control
(Black, Markides & Miller, 1998). These findings tied in to marianism as it was
expected of men to be strong, and thus, it was implied that strong men were also in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
control of their own health. Research also showed that marianism varied by
acculturation level. Being second generation and better educated was associated
with lesser sex-role traditionalism, and more assertiveness among Puerto Rican
women (Soto, 1983). Leaper and Valin’s (1996) work added to this finding as
Mexican American fathers and mothers were significantly more likely to have
gender egalitarian attitudes if they had higher levels of education, and endorsed
communal values such as concern for the group rather than individualistic values
such as self-reliance. In sum, acculturation influenced family obligations, family as
a reference, fatalism, and marianism, however, perceived family support did not
change with acculturation.
Although familism, fatalism, and marianism were salient values among
Hispanics and were negatively associated with acculturation, the relationship
between these values and psychological distress has not been investigated.
Nevertheless, since familism involves strong feelings of loyalty, reciprocity, and
solidarity among members of the same family, caregivers with higher value for the
family will be more willing to provide to the family, and consequently, they will be
less burdened and depressed in a caregiving situation. Similarly, caregivers
endorsing fatalism values will believe it is the will of God or powerful external
agents that their loved one has dementia and that this is an illness beyond their
control. Consequently, they will feel less burdened and depressed because they will
be more accepting of an uncontrollable stressor, caring of a loved one with dementia,
and an incurable illness, dementia. Similarly, since it is expected of women to take
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
care of their children and husband, it is quite possible that women who endorse these
marianism values will believe taking care of a loved one with dementia is an
extension of these gender expectations. Consequently, women who endorse
marianism values will feel less burdened and depressed because they will be more
accepting of their caregiving duties. In sum, the definitions of familism, fatalism,
and marianism suggested that in a caregiving situation, these values would be
negatively associated with distress.
Although the relationship between these cultural values and distress has not
been examined, degree of acculturation to U.S. mainstream society has often been
explored as it relates to psychological functioning among Hispanics, yet the research
findings are inconsistent. In these studies, acculturation was measured on a
continuum whereby individuals adopted the ways of the majority culture and moved
from traditionalism to biculturalism to assimilation. Some authors found higher
levels of acculturation were associated with lower depressive symptoms (Black et al.,
1998; Gallagher-Thompson, Talamantes, Ramirez & Valverde, 1997; Golding &
Burnam, 1990; Golding, Burnam, Benjamin & Wells, 1992; Griffin, 1983; Kaplan &
Marks, 1990; Krause & Goldenhar, 1992; Masten, Penland & Nayani, 1994; Ortiz &
Arce, 1984; Tran, Fitzpatrick, Berg & Wright, 1996; Vega, Warheit, Buhl-Auth &
Meinhardt, 1984; Zamanian, et al., 1992). Other researchers argued that socio
economic variables such as poverty status and unemployment had a much stronger
effect on depressive symptoms than acculturation (Canabal & Quiles, 1995; Negy &
Woods, 1992), and yet others found no association between mental health and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
acculturation (Lang, Munoz, Bernal & Sorensen, 1982). Despite the abundance of
research examining the impact of acculturation on psychological distress among
Hispanics, the findings were inconclusive because there have been pitfalls in the
measurement of acculturation.
First, 62% of the variance in acculturation scales was accounted by factors
assessing language use, language skills and direct contact with the country of origin
(Burnam, Telles, Hough & Escobar, 1987; Cuellar, Harris & Jasso, 1980). Inclusion
of social activities, beliefs, and values might have accounted for the remaining
variance explaining acculturation but this has not been confirmed with the inclusion
of social scales (Burnam, Telles et al., 1987; Marin & Gamba, 1996), and inclusion
of beliefs and values has added only 13% of the variance in acculturation (Cuellar,
Arnold & Gonzalez, 1995).
Second, early acculturation scales assumed that involvement in one culture
necessarily diminished the level of involvement in the other culture since
respondents could not choose an answer portraying independent involvement in both
cultures (Cortes, Rogler & Malgady, 1994). Consequently, bicultural scales were
developed to measure the degree of involvement in both cultures. Results challenged
the assumption that biculturalism represented a halfway point in the acculturation
continuum and suggested that it was a unique pattern. Specifically, when comparing
bicultural individuals to those who had acculturated primarily by shifting their
identification toward the English speaking culture, there were no significant
differences in levels of education (12.5 vs. 12 years) and percentage of lifetime spent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
in the U.S. (86.8% vs. 78.8%) (Magana et al., 1996). Such a finding suggested
biculturalism was not a transitional state, but a differing endpoint. Furthermore,
Cortes et al. (1994) found two general factors when measuring acculturation,
involvement in American culture and involvement in Puerto Rican culture, which
clashed with the assumption that acculturative change was comprised of mutual
cultural exclusivity. Moreover, bicultural and shifting individuals had different
patterns of language use and social affiliations. Those individuals who were
bicultural spoke predominantly Spanish within the family but maintained social
affiliations with individuals from both cultural groups, whereas individuals whose
acculturation style was shifting tended to speak English within the family but
maintained social affiliations primarily with other Hispanic individuals (Magana et
al., 1996). Therefore, research suggested acculturation was not a continuum
whereby one progressed from one end to the other by giving up one’s culture for
another but rather multidimensional whereby one could be independently involved in
both cultures.
Berry’s (1980) four modes of adaptation (assimilation, integration,
separation, and marginalization) described earlier in this paper supported this
multidimensional view of acculturation because individuals could be independently
involved in both cultures. Cuellar, Arnold, and Maldonado (1995) sampled college
students using the Acculturation Rating Scale for Mexican Americans to obtain a
distribution of these four acculturation types. In addition to providing these four
acculturation types, this scale measured independently identification with Hispanic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
and Anglo cultural orientations. Consequently, measuring acculturation as
multidimensional by using this scale could help clarify the relationship between
acculturation and depressive symptoms.
Thus far, research indicated familism, fatalism, and marianism were salient
values among Hispanics, and these values decreased with acculturation.
Furthermore, no empirical investigation of the relationship between these cultural
values and psychological distress has been undertaken. However, the definitions of
these values suggested they would be associated with lesser distress among
caregivers. In addition, the relationship between acculturation and psychological
distress has not been elucidated in research studies because acculturation was
measured on a continuum, not as multidimensional. Consequently, using Berry’s
(1980) four acculturation types as well as measuring independently identification
with Hispanic and Anglo cultural orientations could help clarify the relationship
between acculturation and psychological distress.
In addition to cultural values, acculturation, and psychological distress,
dissatisfaction with family support received and individual expectations of who
should be the person to assume caregiving responsibility are important variables to
consider among caregivers. It is important to distinguish dissatisfaction with family
support received and the familism dimension of perceived support from the family
described earlier. Whereas the first one assesses the extent to which family support
needed is received in the caregiving situation, the other taps into the caregiver’s
general worldview that family members should or should not be providers of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
support. Consequently, one is a measure of dissatisfaction with family support
received and the other, a general worldview of what family support should be like.
Now that this distinction has been made, a description of why family dissatisfaction
is important follows.
In a study of Hispanic caregivers, dissatisfaction with family support received
was significantly correlated with depression among more bicultural caregivers (or
integrated, using Berry’s (1980) term) but not among more traditional (or separated)
ones (Polich & Gallagher-Thompson, 1997). As Polich and Gallagher-Thompson
(1997) suggested, exploration of caregivers’ expectations of family involvement with
caregiving duties could help explain this finding. Hispanic culture dictated caring of
an older person in the family was passed along blood and gender lines so caregiving
responsibility fell first upon blood female relatives and lastly, blood male relatives
(Gallagher-Thompson et al., 1997). For instance, if the caregiver was a wife and she
could no longer take care of her husband, it would be culturally expected that first,
her daughters, second, her grand-daughters, third, her daughters-in-law, fourth, her
sons, fifth, her grand-sons, and so forth assume caregiving responsibility. On the
other hand, if the caregiver was a husband and he could no longer take care of his
wife, if would be culturally expected that first, his daughters, second, his grand
daughters, third, his daughters-in-law, and so forth assume caregiving responsibility.
Following such blood and gender lines is called traditional caregiving assignment or
following traditional sex and kinship lines in this study. The influence of
acculturation on traditional caregiving assignment could help explain the findings in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
Polich and Gallagher-Thompson’s (1997) study since it was possible that
dissatisfaction with family support received and depression among Hispanic
caregivers were due to a disparity between cultural and individual expectations of
who should be the persons assuming caregiving responsibility, which could be
accounted for by acculturation level.
Such a possibility would translate as follows. Individuals with an Anglo
orientation (or who were not first generation) could expect more family involvement
with caregiving whereas caregivers with a Hispanic orientation (or who were first
generation) would assume sole responsibility of caregiving duties because it is
expected of younger generations to take care of older parents or relatives. Since
caregivers with a Hispanic orientation (or who were first generation) would view
cultural expectations of taking care of older parents as their sole responsibility, these
caregivers would name themselves first when asked to list the persons who should
assume caregiving responsibility, and thus, follow traditional caregiving assignment
according to sex and kinship relation. Furthermore, since caregivers with a Hispanic
orientation (or who were first generation) would name themselves first, they would
not expect the family to share this responsibility, and thus, they would not feel
deprived of family support, depressed, and burdened. On the other hand, caregivers
with an Anglo orientation (or who were not first generation) would disagree with
these cultural expectations because of the influence of American values and thus,
would expect more family involvement. Consequently, caregivers with an Anglo
orientation (or who were not first generation) would name first another relative (i.e. a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
sibling) who ranks at the same level in the family hierarchy when asked to list the
persons who should assume caregiving responsibility. By naming another relative as
their first choice of who should assume caregiving responsibility, caregivers with an
Anglo orientation (or who were not first generation) would expect to share
caregiving responsibility with other relatives who ranked at the same level and
would anticipate more family involvement, and thus, be more dissatisfied with
family support received, and feel more depressed and burdened. Finally, caregivers
who were dissatisfied with family help received would be more depressed and
burdened than those who were satisfied with family help.
In summary, early multivariate models of the stress and coping process failed
to include the influence of ethnicity and culture in explaining caregiver distress. The
sociocultural factors of acculturation, familism, fatalism, and marianism were
examined in the Hispanic psychology literature. Acculturation influenced family
obligations, family as a reference, fatalism, and marianism, however, perceived
family support did not change with acculturation. Furthermore, the definitions of
familism, fatalism, and marianism suggested they could be negatively associated
with perceived burden and distress. Also, acculturation could affect depressive
symptomatology, nevertheless, the evidence was mixed because there were pitfalls in
the measurement and conceptualization of acculturation. These pitfalls could be
attenuated by viewing acculturation as multidimensional; individuals could be
independently involved in both cultures. Finally, acculturation appeared to moderate
the relationship between dissatisfaction with family support received and depression
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
among Hispanic caregivers by playing a role in caregivers’ individual expectations
of who should be the persons to assume or share caregiving responsibility.
Consequently, the purposes of this study were to examine which elements of
Hispanic culture changed with increasing acculturation, and which cultural variables,
if any, predicted depressed mood and burden among Hispanic caregivers.
Furthermore, this study sought to determine whether dissatisfaction with family
support received and depression among Hispanic caregivers were due to a disparity
between cultural and individual expectations of who should be the persons to assume
caregiving responsibility, which could be accounted for by acculturation level.
The hypotheses of this study were as follows:
1. Familism, fatalism, and marianism would be positively associated with
Hispanic orientation. These cultural values would be negatively associated
with Anglo orientation.
2. Familism, fatalism, and marianism would be associated with less depression
and less burden.
3. In regression equations, Anglo orientation would be associated with higher
burden scores; the three cultural values would be negatively associated with
burden and would contribute additional explained variance in the association
between Anglo orientation and burden. Hispanic orientation would be
associated with lower burden; the three cultural values would be negatively
associated with burden and would contribute additional explained variance in
the inverse relationship between Hispanic orientation and burden scores.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Similarly, Anglo orientation would be associated with higher depression
scores; the three cultural values would be negatively related to depression and
would contribute additional explained variance in the association between
Anglo orientation and depression. Hispanic orientation would be associated
with lesser depression; the three cultural values would be negatively
associated with depression and would contribute additional explained
variance in the inverse relationship between Hispanic orientation and
depression. These regression analyses would be repeated using generation
status in place of the orientation measures.
4. Cuellar, Arnold and Maldonado’s (1995) measure of acculturation would
provide four acculturation types: assimilated, integrated, separated, and
marginalized.
5. Hispanic orientation and generation status would predict traditional
caregiving assignment according to sex and kinship relation whereas Anglo
orientation and generation status would predict sharing caregiving
responsibility with other relatives who ranked at the same level. Caregivers
who followed traditional sex and kinship lines when listing persons who
should assume caregiving responsibility would be less depressed, less
burdened, and less dissatisfied with family support received than caregivers
who did not follow traditional sex and kinship lines in their caregiving
assignment list. Caregivers who named first a relative of the same rank in the
family in their caregiving assignment list (and who thus felt they wanted to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
share caregiving responsibility) would be more depressed, more burdened,
and more dissatisfied with family support received than caregivers who did
not. Caregivers who were dissatisfied with family help received would be
more depressed and more burdened than those who were satisfied with family
help received.
6. Concerning the three dimensions of familism, Hispanic orientation and Anglo
orientation would not be associated with beliefs about the family being a
reliable provider of help and support for the caregiver. There would be no
significant differences between first and second generation caregivers
concerning their beliefs about the family being a reliable provider of help and
support. Familial obligations and the perception of the family as a referent
would be endorsed more by caregivers with a strong Hispanic orientation and
who were first generation. Familial obligations and familial reference would
be negatively associated with Anglo orientation.
Method
Participants
The sample in this study consisted of Hispanic caregivers who had contacted
community agencies in Los Angeles for services. Community agencies provided
names and telephone numbers of potential subjects to the investigator so that she
could contact them to determine eligibility for the study, inform them of the study,
and obtain consent to participate in the study. To qualify for the study, subjects had
to be the person in charge of providing care to a family member with dementia,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
identify themselves as Hispanic, and speak either English or Spanish. Once subjects
met inclusion criteria for the study and consented to participate, telephone interviews
were done by the investigator and a trained research assistant within one week of
consent. Subjects were not financially compensated for their participation in the
study due to lack of funds. The absence of financial remuneration did not affect
participants’ willingness to participate in the study because those who refused to
participate stated they were declining because of lack of time, or no interest, not
because of the absence of financial compensation.
Recruitment
Recruitment sources were community agencies that provided services to
Hispanic caregivers. Some agencies allowed the investigator to directly contact
caregivers by telephone whereas others required mailing letters to support group
leaders to inform them about the study so that they could contact the investigator for
more information. The Los Angeles Caregiver Resource Center (LACRC) had 175
Hispanic persons who had contacted the agency between 1994 and 1999. Of these
175 persons, 85 were ineligible for the study because 33 care recipients had died, 18
had been placed in a nursing or convalescent home, 14 had an inappropriate type of
dementia (multiple sclerosis, epilepsy, Creutzfeld-Jacob), 16 persons were not the
primary caregiver when they had called the agency, and 4 caregivers were no longer
providing care because they had hired full-time care or the care recipient had moved
to Mexico. In addition to these 85 ineligible persons, 41 caregivers were unable to
be located because their phone numbers were disconnected or no longer in service,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
or they had moved. Thus, a total of 126 persons out of 175 from the LACRC were
ineligible or unable to be located; out of the 49 eligible caregivers, 42 participated in
the study and 7 refused.
In addition to the LACRC, the Alzheimer’s Association was contacted to
recruit caregivers. The Alzheimer’s Association sent out letters to support group
leaders to inform them of this study and its importance, and to ask them to contact
the investigator to provide referrals for the study. Two support group leaders from
the Alzheimer’s Association referred 4 caregivers; 1 was ineligible for the study
because she was a hired caregiver, 2 refused, and 1 participated in the study.
In addition to the LACRC and the Alzheimer’s Association, a support group
leader who conducted groups with Hispanic clients in a hospital was contacted to ask
permission to give a talk on stress management and to recruit subjects for the study.
A total of 10 caregivers attended this talk, 2 left before the investigator could tell the
group about the study, 2 were ineligible because the care recipients had died, 1
refused, and 5 participated in the study.
Other sources of recruitment that failed to yield referrals were a health fair
sponsored by USC/LA Times and caregiver workshops sponsored by the
LACRC/Alzheimer’s Association.
In sum, a total of 48 caregivers participated in the study and 10 refused. The
refusal rate of this study was low (17.24%), which is a strength of this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
Measures
Variables in this study included the following: demographics, traditional
caregiving assignment, sharing caregiving assignment, dissatisfaction with family
support, Hispanic orientation, Anglo orientation, other measures of acculturation
such as generation, familism, familial reference, familial obligations, familial
support, fatalism, marianism, depressive symptoms, and burden.
Demographics. Demographic variables included caregiver age, gender,
marital status, relationship to the care recipient, birthplace, place they grew up as a
child, education, income, number of years lived in the U.S., preferred language for
the interview, care recipient’s diagnosis and date of diagnosis, and the number of
months caregivers had been caring for their relative. Education was determined by
asking respondents what was the highest level of education they completed.
Answers could include: less than high school, some high school, high school
graduate, some college work, college graduate, and post-graduate degree. Income
was determined by asking respondents for their annual household income. Answers
included the following categories: 1) under $8,000, 2) $8,000-11,999, 3) $12,000-
15,999, 4) $16,000-19,999, 5) $20,000-25,999, 6) $26,000-29,999, 7) $30,000-
$35,999, 8) $36,000-39,999, 9) $40,000 or above, and 10) caregiver did not answer
or missing data. Thus, two measures of socio-economic status were education and
income.
Traditional Caregiving Assignment. Caregiver assignment was measured by
asking respondents to think of all the people they knew and indicate who should be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
the first, second, and third person to assume caregiving responsibility of the care
recipient. In interviewing caregivers, two issues arose regarding this question:
whether all the people they knew could include the caregiver as well as all the
persons available in the family network. Consequently, when asking this question,
interviewers stated orally that all the persons the caregivers knew included
themselves. Furthermore, after giving the first three choices, caregivers were asked
orally if these choices represented all the persons that were available in the family
network. The first choice of caregiving assignment given by caregivers was coded 1
if it followed traditional sex and kinship lines, and 0 if it did not. An example of
traditional caregiving would be if the caregiver was a wife and her first choice was
herself, her daughter, or her daughter-in-law; if her first choice was her son, then that
was considered non-traditional because the caregiver did not follow traditional
gender lines. If the caregiver was a daughter and her first choice was herself, her
sister, her daughter, or her sister-in-law, caregiving assignment was coded as
traditional; if her first choice was her husband or uncle, caregiving assigment was
coded as non-traditional because it did not follow traditional gender lines. If the
caregiver was a husband and his first choice was his daughter or daughter-in-law,
caregiving assignment was coded as traditional; if his first choice was himself,
caregiving assignment was coded as non-traditional.
Sharing Caregiving Responsibility. Sharing caregiving responsibility was
measured by using the same question as above (asking respondents to think of all the
people they knew and indicate who should be the first, second, and third person to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
assume caregiving responsibility of the care recipient) but coding it differently. If a
caregiver mentioned a relative who ranked at the same level in the family hierarchy
as first choice, it was assumed that the caregiver wanted to share the responsibility.
The first choice of caregiving assignment given by caregivers was coded 1 if it was a
relative who ranked at the same level in the family hierarchy, and 0 if it was not. An
example of sharing caregiving responsibility would be if the caregiver was a
daughter and her first choice was her sister, her brother, or her sister-in-law or
brother-in-law (all relatives who ranked at the same level in the family hierarchy
because they were siblings or in-laws married to the siblings) as opposed to her
mother or her daughter; if her first choice was herself, then that was considered not
wanting to share caregiving responsibility.
Dissatisfaction with family support received. Dissatisfaction with family
support received was assessed by asking respondents to think of all the help they got
from family and friends in looking after their relative and rate whether that help was:
not received (-4), far less than needed (-3), somewhat less than needed (-2), about
what was needed (-1), not needed (0), and more than needed (1). This question came
from the intake questionnaire used by the Los Angeles Caregiver Resource Center.
Since caregivers had the choice of answering whether help was not needed, it was
assumed that an answer of not receiving help was a reflection of extreme
dissatisfaction because the help was needed but not received. The answers were
coded on a continuum from getting no help to getting more help than needed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
Acculturation Rating Scale for Mexican Americans. Acculturation was
measured using the two subscales of the Acculturation Rating Scale for Mexican
Americans (Cuellar, Arnold & Maldonado, 1995), Anglo orientation and Mexican
Orientation. Since participants were from different Latino countries, the Mexican
Orientation subscale was relabeled Hispanic Orientation. Participants who were not
from Mexico answered questions regarding their country of origin. The Anglo and
Hispanic orientation subscales were composed of 13 and 17 items each. These items
asked respondents to rate on a Likert scale (not at all, very little, moderately, much
or extremely) their frequency of language use (English and Spanish), ethnic identity
(Mexican or other Latino group, American, Mexican-American or other Latino-
American group), and ethnic interaction (association with Mexicans or other Latino
group, Anglos, Mexican-Americans or other Latino-American group). Items in each
subscale are scored from 1 to 5, with higher scores indicating more identification
with a cultural orientation. Items 1, 3, 5, 6, 8, 11, 12, 14, 17, 18, 20, 21, 22, 24, 26,
28, and 29 pertain to the Hispanic Orientation subscale. Items 2, 4, 7, 9, 10, 13, 15,
16, 19, 23, 25, 27, and 30 are part of the Anglo Orientation subscale. The two
cultural orientation subscales have been found to have high internal reliabilities, a =
.86 and a = .89 respectively, and good construct validity (Cuellar, Arnold et al,
1995). In this study, reliabilities of the two cultural orientation subscales were high
(a = .91 and a = .89 respectively). Scale 2 of the Acculturation Rating Scale for
Mexican-Americans which measures the extent to which a person associates with
Mexicans, Mexican-Americans and Americans was not used in this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Acculturation was also assessed by asking respondents where their parents,
grand-parents, and themselves were born in order to determine the respondent’s
generation status in the U.S. This categorical variable was considered separately
from the Acculturation Rating Scale for Mexican Americans.
Familism. Familism was assessed by the Sabogal et al. (1987) familism
scale, which consisted of items found in the Bardis (1959) and Triandis et al. (1982)
familism scales. Sabogal et al.’s (1987) familism scale consisted of 14 Likert items
measuring perceived familial obligations, support from the family, and familial
reference. Items 1 through 6 tapped into familial obligations (i.e. make great
sacrifices to guarantee a good education for children, share home with relatives if
they are in need). Items 7 through 9 tapped into familial support (i.e. one can count
on help from relatives to solve most problems). Items 10 through 14 tapped into
familial reference (i.e. family should consult close relatives regarding important
decisions, one should be embarrassed about bad things done by brothers and sisters).
The three dimensions (family obligations, family support, and family reference) as
well as the total score obtained in the familism scale were considered separate
variables. Answers to items could range from very much in disagreement (1) to very
much in agreement (5); higher scores indicated stronger endorsement of familism
values. Reliability and validity of the Sabogal et al. (1987) familism scale has been
established among Flispanics (a = .70). In this study, reliability of this scale was
good (a = .75).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
Fatalism. Fatalism was assessed using Pearlin and Schooler’s (1978)
Mastery scale. This scale tapped into the extent to which an individual had control
over events in his or her life. The scale had 7 Likert items ranging from strongly
agree (4) to strongly disagree (1); item scoring was reversed, where necessary, so
that a high score indicated high fatalism. The reliability (a = .71 ) and validity of
this scale has been established (Wrinkle, Anzaldua, & Reed-Sanders, 1988) in a
Flispanic sample. In this study, reliability of this scale was good (a = .75).
Marianism. Sex role traditionalism was assessed with the short form of the
Attitudes Toward Women Scale (Spence, Flelmreich, & Stapp, 1973). The short
form of the Attitudes Toward Women Scale (ATWS) consisted of 15 declarative
statements tapping into the rights and roles of women in such areas as vocational,
educational, and intellectual activities; dating behavior and etiquette; sexual
behavior; and marital relationships. Answers to items could range from strongly
agreeing (3) to strongly disagreeing (0); item scoring was reversed so that 3
represented the most traditional and 0, the most contemporary, pro-feminist
response. The reliability and validity of the ATWS has been well established
(Spence et al, 1973; Kranau, Green, & Valencia-Weber, 1982). In this study,
reliability of this scale was good (a = .75).
Burden Interview. Burden was measured using a modified version of the
Burden Interview scale developed by Zarit, Reever, and Bach-Peterson (1980). This
modified version (Knight, Fox & Chou, in press) consisted of 14 items tapping into
how caring for a relative had affected the caregiver’s emotional and social life.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Items were scored from 0 to 4 with higher scores indicative of greater caregiver
distress. Reliability of this scale has been established (a = .92) (Knight et al, in
press). In this study, reliability of this scale was good (a = .70).
Center for Epidemiological Studies-Depression. Depression among
caregivers was measured using a modified version of the CES-D scale developed by
the Center for Epidemiological Studies (Radloff, 1977). The modified CES-D
(Kohout, Berkman, Evans, & Huntley, 1983) consisted of 11 items that tapped into
depressive symptoms in the respondent over the past week. This modified CES-D
used 11 of the original 20 items (items 2, 6, 7, 11, 12, 14, 15, 16, 18, 19, and 20).
Answers to items included four response categories: rarely or none of the time, some
of the time, occasionally, and most of the time. The instrument has high internal
consistency and reliability (Kohout et al, 1983; O’Hara, Kohout, & Wallace, 1985)
and is correlated (r = .95 ) with the 20 item version. In this study, reliability of this
scale was good (a = .75).
Results
The results are organized based on initial analyses and the order of the
hypotheses. An alpha level of .05 was used for all analyses.
Initial Analyses
To check for normality of the data, histograms and normal probability plots
were examined. The demographic variables (age, years in U.S., years caring for
relative), cultural values, Hispanic orientation and burden scores met assumptions of
normality. However, CES-D scores showed a curvilinear pattern in the normal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
probability plot and positive skewness (most caregivers scored in the low (non
depressed end) in the histogram, thereby requiring a natural log transformation. This
natural log transformation corrected most deviations from normality in the
probability plot. Anglo orientation scores did not meet assumptions of normality
because they had negative skewness; most caregivers scored in the high (Anglo) end.
Reflecting and square root transformations of these scores decreased skewness but
increased kurtosis, thereby not improving fit to assumptions of normality. Thus,
Anglo orientation scores were not transformed. Dissatisfaction with family support
received showed negative skewness; few caregivers were satisfied with the help
received. Reflecting and square root transformations of dissatisfaction with family
support received decreased skewness but increased kurtosis, thereby not helping to
meet assumptions of normality. Consequently, categories of dissatisfaction with
family support received were collapsed into a dichotomous variable; getting no help,
receiving far less or somewhat less than needed were considered a reflection of
dissatisfaction whereas getting about what was needed, more than needed, and not
needing help were considered satisfaction with help. Dissatisfaction was coded 1
and satisfaction was coded 0.
Descriptive Analyses
The sample (N = 48) had a mean age of 58 (SD=13.32, range 24-85), was
predominantly female (79.2%), and consisted of daughters (37.5%), wives (29.2%),
husbands (14.6%), daughters-in-law (6.3%), and other relatives or friends (12.4%) of
care recipients. The majority of care recipients had dementia of the Alzheimer’s
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6
type (54.2%); 31.3 % had vascular dementia, 4.2% Parkinson’s disease, 2.1% a
combination of Parkinson’s and Alzheimer’s dementia, and 8.3% could not specify
the type of dementia. Approximately two thirds of caregivers were married, had an
annual income less than $20,000, had been caring for the care recipient for 1 to 5
years, and were interviewed in English. Fifty-six percent of caregivers had a high
school education or less; 72.9% finished school in the U.S. Forty-six percent of
caregivers were born and raised outside the U.S.; 29% of the sample was born in
Mexico, 6.3% in Puerto Rico, 4.2% in Central America (El Salvador and
Guatemala), 4.2% in Cuba, and 2.1% in Argentina. Caregivers born abroad had been
living in the U.S. an average of 34 years (SD=17, range=5-75). Table 1 shows
more detailed descriptive statistics for the demographic variables of the sample.
Table 2 shows descriptive statistics for the other variables in the sample.
Table 1
Descriptive Statistics for the Sample (N = 48)
Demographics________________________
First language learned (% )
English 31.3
Spanish 64.6
Both 4.2
Interview language (% )
English 66.7
Spanish 33.3
Education (% )
< than high school 2 0 .8
some high school 6.3
high school graduate 29.2
some college 27.1
college graduate 16.7
post-graduate degree 0 .0
Country finished school (% )
U.S.A 72.9
Mexico 2 0 .8
Cuba 4.2
Guatemala 2.1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
1 (continued)
il Status (% )
Married 64.6
Separated 4.2
Divorced 14.6
Widowed 4.2
Living w/ partner 2.1
Never married 10.4
il household income (% )
< $8 ,0 0 0 10.4
$8,000-$l 1,999 16.7
$12,000-$15,999 12.5
$16,000-$19,999 6.3
$20,000-$25,999 12.5
$26,000-$29,999 2.1
$30,000-$35,999 6.3
$36,000-$39,999 2.1
> $40,000 14.6
missing 16.7
caring for relative (% )
< 1 10.4
1-2 22.9
3-5 29.2
6 -1 0 27.1
> 11 10.4
Table 2
Variable Mean Median SD Range Scale.
HOS 56.58 58.50 14.14 25-83 17-85
AOS 45.17 47.50 11.72 18-61 13-65
Familism 49.14 49.00 7.41 29-63 14-70
Family Reference 14.94 15.00 3.79 5-22 5-25
Family Support 10.15 10 .0 0 2.24 5-15 3-15
Family Obligations 24.04 24.50 3.41 15-30 6-30
Fatalism 16.29 17.00 3.79 8-27 7-28
Marianism 13.54 14.00 6.62 2-26 0-45
Burden 20.69 2 0 .0 0 8.04 0-37 0-56
CES-D 1 1 9.56 7.00 6.78 0-25 0-33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Table 2 (continued)
Note. HOS: Hispanic Orientation Scale, AOS: Anglo Orientation Scale, CESD-11 : Center for
Epidemiological Studies-Depression 11 item version.
Using Cuellar, Arnold and Maldonado’s (1995) definition of generations
yielded 22 subjects being classified as first generation, 18 as second, 3 as third, 3 as
fourth, and 2 as fifth. Since approximately equal numbers were obtained for first and
second generations, these two generations were used for comparisons.
In terms of traditional caregiving assignment, thirty-three caregivers followed
traditional blood and gender lines in their first choice of caregiving assignment and
fourteen did not. Concerning sharing caregiving responsibility, seven caregivers
named first a relative who ranked at the same level in the family hierarchy (for
example, if the caregiver was a daughter, she named first a sibling, or brother or
sister-in-law) and 40 did not (for example, if the caregiver was a daughter, she
named herself first).
In terms of perception of family help received, twenty caregivers thought
help from family and friends was not received, 8 thought help was far less than
needed, 8 somewhat less than needed, 11 about what was needed, 1 more than
needed, and 0 did not need help. As described in the initial analyses section, getting
no help, receiving far less or somewhat less than needed were collapsed into one
category, dissatisfaction with family help received, and getting about what was
needed, more than needed, and not needing help were collapsed into another
category, satisfaction with help received because the original categories were not
normally distributed and did not fit assumptions of normality when using reflecting
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
and square root transformations. Using these two categories, thirty-six caregivers
were classified as dissatisfied and 12 as satisfied with help received. Dissatisfaction
was coded 1 and satisfaction, 0.
Acculturation
There was a significant negative association between Anglo Orientation and
Hispanic orientation scores (r = -.69, p<.001). Such a high correlation suggested
these two scores tapped the same construct in this sample of caregivers.
Furthermore, this result provided support for a linear continuum of acculturation in
this sample whereby lower Anglo orientation scores were associated with higher
Hispanic orientation scores. Even though this correlation supported the claim that
acculturation was linear in this sample, the two scores were used separately as
indices of acculturation in the analyses because the acculturation literature indicated
that acculturation should be viewed as bidimensional; individuals could be
independently involved in both cultures.
Cultural values and acculturation (Hypothesis 1)
To determine the associations between cultural values and acculturation,
Pearson’s correlations were calculated (see Table 3). As expected, familism was
positively associated with Hispanic orientation, and negatively associated with
Anglo orientation. Fatalism and marianism were not significantly correlated with
either Hispanic or Anglo orientation, contrary to the hypothesis. The cultural values
did correlate among themselves.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
Cultural values, depression, and burden (Hypothesis 2)
In order to test for negative associations between cultural values, depression,
and burden, Pearson’s correlations were obtained (see Table 3). Familism was
significantly associated with lesser burden, as expected. Fatalism and marianism
were not significantly correlated with depression or burden, contrary to the
hypotheses.
Table 3
Intercorrelations between Variables (N =48)
Variable 1 2 3 4 5 6 7 8 9 10 ii
1. HOS -- -.69*** .28* .3 3 *** .16 .12 .16 .16 .02 -.08 -. 10
2. AOS . 4 3 *** -47***-28 -.23 -.1 2 -.14 .18 .01 .10
3. Familism
go*** 7 3 ***
go* * *
.07
2 4 * * * _ 2 9 **
-.15 -.21
4. Family Reference
27*** 27* * *
.27*
5 7 * * *
-.18 .04 -.01
5. Family Support -
51***
-.04 .13 -.09 -.16 -.17
6 . Family Obligations - -.13 .02 - 38*** -.28** .34’
7. Fatalism -
_ 4 1 * * *
.04 .19 .11
8 . Marianism - -.19 .24 .17
9. Burden -
49 * * * .54’
10. CES-D 11 - .94"
11. Log CES-D 1 1
- -
Note. HOS: Hispanic Orientation Subscale, AOS: Anglo Orientation Subscale,
*** p< .01 (one-tailed), ** p<025 (one-tailed), * p<05
Predictors of depression and burden (Hypothesis 3)
To determine the effects of income and education on depression and burden
scores, one-way ANOVA’s were conducted using the 10 levels of income and 5
levels of education. There were no significant differences in CES-D (F (9, 32) =
1.16, MSE = 54.83, p = .36; F (4, 43) = .82, MSE = 38.27, p = .52), log of CESD (F
(9, 32) = 1.16, MSE = .14, p = .36; F (4, 42) = .82, MSE = .09, p = .52) and burden
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
(F (9, 32) = .52, MSE = 34.53, p = .85; F (4, 43) = 1.07, MSE = 68.79, p = .38)
scores by income or education levels respectively. In order to increase the power of
these analyses, degrees of freedom were reduced by collapsing levels of income and
education into 2 levels of income (<=$20,000 and >$20,000) and 2 levels of
education (<= high school education and > high school education). T-tests were
calculated using these 2 levels of income and education. There were no significant
differences in CES-D (t (38) = -.11, p = .91; t (46) = -.26, p = .79), log of CES-D (t
(37) = .23, p = .81; t (45) = -.58, p = .57), and burden (t (38) = .87, p = .39; t (46) = -
.93, p = .36) scores between caregivers with high and low levels of income and
education respectively. In sum, there were no significant differences in depression
and burden by low and high levels of education and income, consequently there was
no need to control for education and income in the regression analyses.
To determine whether cultural values predicted depression and burden when
controlling for acculturation, multiple hierarchical regression analyses were
performed; acculturation was entered as the first step, and all 3 cultural values, as
the second. Once acculturation was controlled for, and all 3 cultural values were
considered simultaneously, none of the cultural values (including familism) were
significant predictors of depression (see Tables 4 and 5) or burden (see Table 6).
The change in variance in step 2 was not significant, thus, the cultural values did not
significantly explain additional variance in the relationship between acculturation
and depression (see Tables 4 and 5) or acculturation and burden (see Table 6).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4
Summary of Hierarchical Regression Analysis for Variables Predicting CES-D Scores when
32
Controlling for Acculturation (N = 48)
Variable B SEB
e
Step 1
Hispanic Orientation -.04 .07 -.08
Step 2
Hispanic Orientation -.03 .07 -.07
Familism -.2 2 .14 -.24
Marianism .30 .17 .29
Fatalism .17 .28 .10
Note. R2= .01 for Step 1 (p = .61); A R " := .12 for Step 2 (p = .13).
Step 1
Anglo Orientation .007 .09 -.01
Step 2
Anglo Orientation -.03 .09 -.06
Familism -.26 .15 -.28
Marianism .30 .17 .29
Fatalism .14 .28 .08
Note. R " = .00 for Step 1 (p = .93); AR“ - = .12 for Step 2 (p = . 12).
Table 5
Summary of Hierarchical Regression Analysis for Variables Predicting Log of CES-D Scores when
Controlling for Acculturation (N = 48)
Variable B SEB
Step 1
Hispanic Orientation -.0 0 .00 -.1 0
Step 2
Hispanic Orientation -.0 0 .00 -.08
Familism -.01 .01 -.29
Marianism .01 .01 .27
Fatalism no m m
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5 (continued)
Note. R2= .01 for Step 1 (p=.50); AR2 = .11 for Step 2 (p = .18).
Step 1 .003 .00 .10
Anglo Orientation
Step 2
Anglo Orientation .00005 .01 .01
Familism -.01 .01 -.30
Marianism .01 .01 .27
Fatalism .00 .01 .03
Note. R2= .01 for Step 1 (p=.51); AR2 = = .10 for Step 2 (p = .20).
Table 6
Summary of Hierarchical Regression Analysis for Variables Predicting Burden when Controlling for
Acculturation (N = 48)
Variable B SEB
Step 1
Hispanic Orientation .009 .08 .02
Step 2
Hispanic Orientation .06 .09 .10
Familism -.29 .17 -.27
Marianism -.2 0 .20 -.16
Fatalism .23 .34 .11
Note. R2= .00 for Step 1 (p=.91); AR2 = .12 for Step 2 (p = .15).
Step 1
Anglo Orientation .12 .10 .18
Step 2
Anglo Orientation .06 .11 .08
Familism -.23 .18 -.21
Marianism -.2 0 .20 -.17
Fatalism .28 .34 .13
Note. R2 = .03 for Step 1 (p=.23); AR2= .08 for Step 2 (p = .28).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
To determine whether cultural values predicted depression and burden when
controlling for generation status (first and second), multiple hierarchical regression
analyses were performed; generation status was entered as a dummy variable in the
first step, and all 3 cultural values, as the second step. When controlling for
generation status, none of the cultural values were significant predictors of
depression (see Tables 7 and 8) or burden (see Table 9). The change in variance in
step 2 was not significant, thus, the cultural values did not significantly explain
additional variance in the relationship between generation status and depression (see
Tables 7 and 8) or generation status and depression (see Table 9).
Table 7
Summary of Hierarchical Regression Analysis for Variables Predicting CES-D Scores when
Controlling for Generation (N = 40)
Variable B SEB
fi
Step 1
Generation -.58 1.09 -.08
Step 2
Generation -.34 1.20 -.05
Familism -.25 .15 -.27
Marianism .29 .17 .28
Fatalism .14 .29 .08
Note. R/= .01 for Step 1 (p = .60); AR2= .12 for Step 2 (p = .14).
Table 8
Summary of Hierarchical Regression Analysis for Variables Predicting Log of CES-D Scores when
Controlling for Generation (N = 40)
Variable B SE B { 3
Step 1
Generation .02 .06 .05
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 8 (continued)
Step 2
Generation .01 .06
j
.04
Familism -.01 .01 -.30
Marianism .01 .01 .27
Fatalism
i _
.00
. - 'J _ _ _
.02 .03
Note. R " = .00 for Step 1 (p=.76); AR2 =.11 for Step 2 (p = . 17).
Table 9
Summary of Hierarchical Regression Analysis for Variables Predicting Burden when Controlling for
Generation (N = 40)
Variable B SE B
£
Step 1
Generation 1.96 1.26 .22
Step 2
Generation 1.39 1.42 .16
Familism -.22 .17 -.20
Marianism -.16 .21 -.13
Fatalism .34 .34 .16
Note. R7= .05 for Step 1 (p=.13); AR^= .08 for Step 2 (p = .30).
Acculturation types (Hypothesis 4)
Using Cuellar, Arnold and Maldonado’s (1995) cut-off scores to partition the
sample into one of four acculturation types resulted in 13 cases being classified as
segregated (separated), 13 as marginal (low bicultural), 2 as high bicultural, 6 as
assimilated, and 14 as unclassifiable. Descriptive statistics of the Anglo and
Hispanic orientation scores used to obtain these four acculturation types can be
found in Table 2. An examination of the scatter plot of the Mexican Orientation and
Anglo Orientation mean scores showed that Cuellar’s cut-off scores had non
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
contiguous regions, thereby considering subjects who fell in these non-contiguous
regions to be unclassifiable. An example of an unclassifiable subject would be
someone whose Anglo orientation mean score was greater than 3.7 but less than 4.11
and whose Hispanic orientation mean score was greater than 2.44 but less than 3.59,
thereby falling in the region between the bicultural and assimilated groups. The
limitations of Cuellar’s classification system are further delineated in the discussion.
Since small cell sizes were obtained for each acculturation type and the range of
acculturation types was restricted to segregated and marginalized caregivers, using
acculturation types in hypotheses 1 through 3 was not pursued. Other ways of
partitioning the sample into acculturation types such as using median splits were not
pursued because this study had low power as a result of the small sample.
Consequently, using median splits would have resulted in a further loss of statistical
power and increased the likelihood of spurious effects. The significance of finding
predominantly segregated and marginalized caregivers is pursued in the discussion.
Acculturation, generation, traditional caregiving, sharing caregiving, distress, and
family dissatisfaction (Hypothesis 5)
The first choice of caregiving assignment given by caregivers was coded 1 if
it followed traditional sex and kinship lines, and 0 if it did not. Thirty-three
caregivers were traditional in their first choice of caregiving assignment and fourteen
were not. To determine whether Hispanic orientation and generational status
predicted traditional caregiving assignment, logistic regression was performed.
Neither Hispanic orientation (1 3 = .01, SE = .02, p = .59) nor generational status (OR
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
= .63, p = .63) were significant predictors of traditional caregiving assignment,
contrary to the hypothesis.
To distinguish caregivers who endorsed sharing caregiving responsibility
with a relative ranking at the same level in the family, the first choice of caregiving
assignment given by caregivers was coded 1 if it was a relative who ranked at the
same level in the family and 0 if it was not. Seven caregivers endorsed sharing
caregiving responsibility, forty did not. To determine whether Anglo orientation and
generational status predicted sharing caregiving responsibility, logistic regression
was performed. Neither Anglo orientation (B = .08,SE = .05,p = .14) nor
generational status (OR = 2.5, p = .47) were significant predictors of sharing
caregiving assignment, contrary to the hypothesis.
To determine whether caregivers who followed traditional sex and kinship
lines in their caregiving assignment were less depressed and burdened than
caregivers who did not, t-tests were calculated. There were no significant
differences in depression (t (45) = -.38, p = .71) or burden (t (45) = .47, p = .64)
between caregivers who followed traditional sex and kinship lines and those who did
not. To determine whether caregivers who followed traditional sex and kinship lines
were less dissatisfied with family help, a chi-square was calculated. There was no
significant relationship between following traditional sex and kinship lines, and
family dissatisfaction (% 2 (1, N = 47) = .10, p = .76). Thus, caregivers who followed
traditional sex and kinship lines were not significantly less depressed, burdened, or
dissatisfied with family help.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To determine whether caregivers who named first a relative of the same rank
in the family as their first choice of caregiving assignment (and thus felt they wanted
to share caregiving responsibility) were more depressed and burdened than
caregivers who did not, t-tests were calculated. There were no significant
differences in depression (t (45) = 1.61, p = .12) or burden (t (45) = .16, p = .88)
between caregivers who named a relative of the same rank in the family and those
who did not. To determine whether caregivers who named a relative of the same
rank in the family were less dissatisfied with family help, a chi-square was
calculated. There was no significant relationship between naming a relative of the
same rank in the family and family dissatisfaction (%2 (1, N = 47) = 1.30, p = .26).
Thus, caregivers who named a relative of the same rank in the family (and who thus
felt they wanted to share caregiving responsibility) were not significantly less
depressed, burdened, or dissatisfied with family help.
To determine whether caregivers who were dissatisfied with family help
received were more depressed and burdened than those who were satisfied with help,
t-tests were calculated. There was a significant difference in depression (t (46) =
2.83, p < .01) and burden (t (46) = 2.56, p = .01) scores between dissatisfied and
satisfied caregivers with dissatisfied caregivers having higher depression and burden
scores.
Acculturation orientation, generation type, and familism dimensions (Hypothesis 6)
To test for differential associations between Hispanic orientation, Anglo
orientation, and familism dimensions, Pearson’s correlations were calculated (see
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
Table 3). There were no significant associations between Hispanic orientation,
Anglo orientation, and familial support, although the size of the correlation (r = -.28,
p = .06) between familial support and Anglo orientation suggested a small to
moderate effect which would not be significant due to small sample size and
resulting low power. These findings partially confirmed the hypothesis of no
relationship between acculturation and familial support. To test for a lack of
differences in perceived familial support between first and second generation
caregivers, a t-test was performed. There was no significant difference in familial
support, (t (38) =.82, p=.42) between groups, confirming the hypothesis that
Hispanic caregivers endorsed beliefs of familial support regardless of generation
type. In sum, the relationship between familial support and Anglo acculturation
scores was inconclusive but it did not differ by Hispanic orientation or between
generation types.
To test for a positive relationship between Hispanic orientation and the other
two dimensions of familism, familial obligations and the perception of the family as
referents, Pearson’s correlations were calculated (see Table 3). There was no
significant association between Hispanic orientation and familial obligations, which
was the individual’s perceived obligation to provide material and emotional support
to the family, thereby disconfirming the hypothesis of a positive relationship
between Hispanic orientation and familial obligations. However, there was a
significant association between Hispanic orientation and familial reference,
perceiving the family as a referent behaviorally and attitudinally, thereby supporting
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
the hypothesis of a positive relationship between Hispanic orientation and familial
reference. To test for a negative relationship between Anglo orientation, familial
obligations, and familial reference, Pearson’s correlations were also calculated (see
Table 3). The association between Anglo orientation and familial obligations did not
reach significance. However, power was low and the correlation (r = -.23, p = .06)
could indicate a small effect. This finding disconfirmed the hypothesis of a negative
relationship between Anglo orientation and familial obligations. As expected, there
was a significant association between Anglo orientation and familial reference; this
confirmed the hypothesis of a negative relationship between Anglo orientation and
familial reference. In sum, familial reference was lower among those with higher
Anglo acculturation scores and familial obligations was not, although the negative
relationship between familial obligations and Anglo orientation cannot be ruled out
due to low power.
To test for differences in perceived familial obligations and reference
between first and second generation caregivers, t-tests were performed. There was a
significant difference in familial reference (t (38) = 2.84, p = .01) between groups
with first generation caregivers having higher familial reference scores than second
generation caregivers. This finding confirmed the hypothesis that familial reference
was lower in the second generation. There were no significant differences in familial
obligations (t (38) = 1.46, p = .15) between groups, lending no support to the
hypothesis that familial obligations was lower in the second generation. In sum,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
familial reference was correlated with Hispanic orientation and was lower among
second generation caregivers while familial obligations was not.
Additional analyses
Since familism was the only cultural value found to be significantly
associated with both acculturation orientations and burden, the possible mediating
role of familism was considered by testing first whether either acculturation
orientation predicted burden. Regression analyses indicated that neither Hispanic (B
= .009, SE = .08, p = .91) nor Anglo (B= .12, SE = .10, p = .23) orientations
significantly predicted burden. Consequently, further testing of the mediating role
of familism was not pursued.
Regression analyses indicated that the three cultural values did not
significantly explain additional variance in burden when controlling for
acculturation. However, of these three cultural values, only familism was correlated
with burden. Consequently, regression analyses were performed to examine
whether familism alone significantly contributed to additional variance in burden
when controlling for acculturation. Results indicated that familism significantly
accounted for additional variance in burden when controlling for Hispanic
orientation (AR =9.5 %, p = .035) but not when controlling for Anglo orientation
2 t
(AR = .06, p = .10). Thus, when controlling for Hispanic orientation, familism was
a significant predictor of burden (B = -.35,SE = .16,p = .035)but not when
controlling for Anglo orientation (B = -.27, SE = .17, p = .10).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
Since the caregiving literature has found that burden and depression are
positively correlated, the relationship between burden and depression was examined
by regression analysis. Burden significantly predicted CES-D scores (B=.42,
S E = .ll, p<.001) and the log of CES-D scores (B = .02, SE = .00, p < .001). The
indirect effect of familism on depression through burden was considered.
Regression analyses indicated that familism (B = -.01, SE = .00, p = .15) did not
significantly predict depression. Consequently, further testing of the mediating role
of burden was not pursued.
The indirect effect of burden on depression through family dissatisfaction
was also considered. Logistic regression indicated that burden (B=. 11, SE=.05,
p=.02) was a significant predictor of family dissatisfaction. Regression analyses
also showed that family dissatisfaction, entered as a dummy variable (B=5.97,
SE=2.11, p=.007) was a significant predictor of CES-D scores. From previous
analyses, burden (B=.42, SE=.l 1, p<.001) was a significant predictor of depression.
When controlling for family dissatisfaction, burden (B=.34, S E -.l 1, p=.004) was a
significant predictor of CES-D scores, and its standardized beta weight was lower
than the standardized beta weight obtained when regressing depression on burden.
This finding suggested that family dissatisfaction mediates the relationship between
burden and depression. In order to confirm that family dissatisfaction was the
mediator, an alternative model was tested in which burden was the mediator of the
relationship between family dissatisfaction and depression. Regression analysis
indicated that family dissatisfaction (B=6.47, SE=2.53, p=.01) was a significant
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
predictor of burden. When controlling for burden, family dissatisfaction (B=3.75,
SE=2.08, p=.08) was not a significant predictor of CES-D scores. This finding
confirmed that burden mediated the relationship between family dissatisfaction and
depression. In sum, regression analyses suggested two contradictory findings:
family dissatisfaction mediated the relationship between burden and depression, and
burden mediated the relationship between family dissatisfaction and depression.
Since this study consisted of cross-sectional data, it is difficult to disentangle which
variable, family dissatisfaction or burden, is a mediator.
Since caregiving research has shown that dissatisfaction with family support
was significantly correlated with depression among more bicultural caregivers but
not among more traditional ones (Polich & Gallagher-Thompson, 1997), the
relationship between acculturation and family dissatisfaction was examined.
Logistic regression analyses indicated that neither Anglo orientation (B = -.01,
SE=.Q3, p=.63) nor Hispanic orientation (B=.03, SE=.Q3, p=.17) scores were
significant predictors of family dissatisfaction. Similarly, generation status was not
a significant predictor of family dissatisfaction (OR = 1.31, p = .71).
To determine whether familism values confounded the results obtained
concerning traditional caregiving assignment, sharing caregiving responsility, and
family dissatisfaction, t-tests were performed. There were no significant differences
in familism values between caregivers who endorsed traditional caregiving
assignment and those who did not (t (45) =1.03, p=.31). There were no significant
differences in familism scores between those caregivers who wanted to share
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
caregiving responsibility and those who did not (t (45)=.29, p=.77). Finally, there
were no significant differences in familism scores between caregivers satisfied with
family help and those who were dissatisfied (t (46) =-.08, p = .94).
Since this sample included some males (n = 10), it was quite possible that
males with marianism values would feel more depressed and burdened than females
endorsing marianism values because males would perceive caregiving as the duty of
women, not the responsibility of men. Consequently, the relationship between
marianism, acculturation, and distress in this sample could have been attenuated by
the answers given by males. Therefore, one-tail correlational analyses examining
the relationship between marianism, acculturation, and distress were explored for
females only (n = 38). Results indicated that marianism was not significantly
associated with Anglo and Hispanic orientations (r = -.23, p = .08, and r = .23, p =
.08), with burden (r = -.17, p = .15), or with the log of CES-D scores (r = .20, p =
.11).
Finally, since the negative correlation (r = -.68) between Anglo and Hispanic
orientation scores suggested that acculturation could be measured on a linear
continuum in this sample, a linear measure of acculturation was calculated and used
in correlational analyses. This linear measure of acculturation was called
acculturation score and was obtained by subtracting the mean of the Hispanic
orientation score from the mean of the Anglo orientation score (AOS mean - HOS
mean); positive scores represented individuals who were more Anglo oriented and
negative scores, those who were more Hispanic oriented. Descriptive statistics
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
indicated the mean acculturation score was .15 (M = .38, SD = 1.59, range -2.73 to
2.91). A histogram showed that acculturation scores were not normally distributed
because of kurtosis and negative skewness. Reflecting and taking the square root of
acculturation scores increased kurtosis, thereby not helping meet assumptions of
normality. Thus, acculturation score was not transformed. Pearson’s correlations
indicated that the acculturation score was not significantly associated with CES-D (r
= .05, p = .38), log of CES-D (r = .05, p = .38) and burden scores (r = .09, p = .27).
Furthermore, the acculturation score was not significantly correlated with fatalism (r
= .15, p = .16) and marianism (r = .16, p = .14). However, the acculturation score
was significantly associated with familism (r = -.39, p = .003), suggesting that
higher Anglo orientation scores were associated with lower familism scores. Since
this was the only significant correlation found, further analyses were not conducted.
Discussion
In summary, the results of this study indicated that there were significant
associations between acculturation, familism, burden, dissatisfaction with family
support, and depression. A path diagram of these associations would be:
Acculturation -> Familism -> Burden -> Dissatisfaction with family support ->
Depression.
Correlational results from the current study showed that each individual connection
in the path diagram was significant. However, as noted in the analyses section, it
was impossible to determine whether burden or dissatisfaction with family support
was a mediator of depression, consequently, the location of these two variables can
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
be interchanged in this path diagram. The products of the standardized regression
coefficients on the paths between two variables of interest yield the indirect effects
of one variable on another. In this model, the indirect effects of acculturation and
familism on depression (and of acculturation on burden) were very small or virtually
zero.
Consequently, the results of this study suggested that culture, as measured by
acculturation does not directly affect outcomes such as depression and burden in the
stress and coping model of Hispanic caregivers, but rather, it has small to negligible
indirect effects on caregiving outcomes, thereby indicating limitations to the
influence of culture on the caregiving outcomes of Hispanic caregivers. The fact that
acculturation had small to near zero indirect effects on caregiving outcomes does not
rule out the importance of culture in the stress and coping model of Hispanic
caregivers because acculturation affected precursor variables such as familism, and
familism (a cultural value) predicted burden when controlling for the effects of
acculturation. Thus, the findings of this study suggested that culture plays a
significant role in the process of caregiving, but not the emotional distress outcomes
of Hispanic caregivers.
This absence of cultural influences on caregiving outcomes has not been
found in Knight, Silverstein, McCallum, and Fox’s (2000) study of African-
American and non-African American caregivers. In their study, African-American
caregivers appraised caregiving as less burdensome than Whites and had equal levels
of depression and anxiety than Whites; however, African-Americans used more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
emotion-focused coping and were younger, and in poorer health than Whites, thereby
increasing their burden and emotional distress. Thus, the inverse effect of race on
burden was counterbalanced by the effect of race on coping styles, health, and age,
with a net effect of equal distress outcomes between African-American and White
caregivers. In this study, coping styles and caregiver health were not assessed, thus,
it is not possible to ascertain the potential influence of culture on the coping styles
and health of Hispanic caregivers. However, Knight et al.’s (2000) results suggest
that culture could affect the coping styles used by Hispanic caregivers as well as
their health. Furthermore, other cultural influences such as beliefs about dementia,
and traditional ways of seeking help for dementia by means of curanderos,
espiritistas, or herbalistas might play a role in the stress and coping process of
Hispanic caregivers. Thus, future research studies need to examine the influence of
culture on factors that affect the process of caregiving such as coping styles, health,
beliefs about dementia, and traditional ways of seeking help.
Compared to other studies of Hispanic caregivers, the lack of an association
between acculturation and depression corroborated Polich and Gallagher-
Thompson’s (1997) findings when using a subset of six questions from the Cuellar et
al. (1980) scale of acculturation and years spent in the U.S.
The association between Hispanic orientation and familism tied in with Cox
and Monk’s (1990, 1993) findings in which Hispanic caregivers believed that parents
were justified in expecting assistance from their children and it was the children’
duty to provide it; thus, being Hispanic or associating with Hispanic cultural values
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
was related to endorsing familism values and filial support. However, Cox and
M onk’s studies did not examine the influence of acculturation on filial beliefs; thus,
this study contributed to the literature by demonstrating that Anglo acculturation is
associated with lesser familism.
Since this study did not include caregivers of different ethnicities, it is
difficult to ascertain whether this sample of Hispanic caregivers had higher or similar
depression levels than Black or White caregivers in comparative studies found in the
literature. Nevertheless, when converting the CES-D 11 item mean score in this
sample to obtain a CES-D 20 item mean score by using an algebraic expression
(mean CES-D 11 item/11 = mean CES-D 20 item/20), similar levels of depression
were obtained in this sample compared to other Hispanic and White caregiver
samples. In this sample, the converted mean CES-D score was 17.38 which was
similar to the mean depression scores (19.68, 18.5, and 16.41 respectively) obtained
in Cox and Monk’s (1990, 1993) Hispanic samples, Haley et al.‘s (1996) White
caregiver sample, and Mintzer et al.’s (1992) comparative study. Compared to Black
caregiver samples, this Hispanic sample had higher levels of depression since the
mean depression score in Cox and Monk (1990) and Haley et al.’s (1996) Black
samples was 9.74 and 11.88. In sum, this Hispanic sample had similar levels of
depression than other Hispanic and White caregiver samples, and higher depression
scores than some Black caregiver samples.
When converting the Burden Interview 14 item mean score in this sample to
obtain a Burden Interview 22 item mean score (mean Burden 14 item/14= mean
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
Burden 22 item/22), the mean burden score obtained in this sample was in between
the ones found in other Hispanic samples. In this sample, the converted mean
Burden score was 32.51, and the mean burden score in Cox and Monk’s (1990,
1993) Hispanic samples was 26.3 and 36.6. Compared to White samples, this
sample had lower burden scores since the mean burden score in Hinrichsen and
Ramirez’s (1992) White sample was 38.15. Finally, in comparison to Black
samples, this sample had similar burden scores since the mean burden score in Cox
and Monk (1990) and Hinrichsen and Ramirez’s (1992) Black samples was 24.8 and
29.47. In sum, this Hispanic sample had similar burden scores than Black samples
and lower burden scores than White caregiver samples in the literature.
In terms of familism values, lower familism values among those with higher
acculturation scores has been found in a study of Korean caregivers using a different
scale than the one in this study, in which Koreans had higher levels of familism than
White American caregivers, with Korean Americans between the two groups (Youn,
Knight, Jeong, & Benton, 1999). However, the negative association between
familism and burden found in this sample of Hispanic caregivers was not consistent
with Youn et al.’s (1999) study as both Koreans and Korean Americans had high
levels of burden. Consequently, it appears that familism is a protective factor against
burden for certain cultural groups such as Hispanic caregivers but not for other
cultural groups such as Korean caregivers.
The findings of this study also indicated that perceiving the family as a
referent behaviorally and attitudinally was the familism component most influenced
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
by acculturation. Lesser familial reference with acculturation could have important
implications for future waves of acculturated Hispanic caregivers because they might
not define their selves in relationship to the family, and consequently, view
caregiving as an intrusion to their lives, not a natural process of family life.
Appraising caregiving as intrusive could thus result in an increase in the utilization
of formal caregiving services by more acculturated Hispanic caregivers.
Furthermore, the results indicated that familial reference was the one component of
familism that contributed to the significant relationships between familism, burden,
and acculturation. Thus, the extent to which an individual identified with the family
of origin was inversely related to the degree to which an individual acculturated and
felt burdened. This finding has important clinical implications for acculturated
caregivers because these caregivers might benefit more from interventions focusing
on reducing their cognition of caregiving as an intrusion to their personal life by
preserving familial reference.
The results of this study also demonstrated that acculturation was
unidimensional across the range of traditional and marginal caregivers obtained in
this sample of caregivers since the two acculturation orientations were highly
correlated. Had this study included a wider range of acculturation types, Cuellar’s
acculturation scale might have tapped two truly orthogonal dimensions as it was
meant to do. Thus, although Cuellar’s acculturation scale was supposed to show a
two dimensional model of acculturation, the results indicated that this scale tapped
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
one dimension of acculturation across the range of traditional and marginal
caregivers in this sample.
This study revealed severe pitfalls in Cuellar’s classification of acculturation
types which was based on the means and standard deviations obtained in his sample
of university students. As mentioned in the results section, 14 caregivers were
unclassifiable because the cut-off scores used to obtain orthogonal acculturation
types did not include all participants in the study. Consequently, caregivers who
were in undefined regions were excluded from one of the four orthogonal
acculturation types. The percentage of unclassifiable, traditional and marginalized
caregivers in this sample (27%, 27%, and 29%) was higher than the percentages
found in Cuellar’s study (18%, 10.5%, and 10.3%), most likely as a result of the
scale being normed on a college population. Whereas this sample consisted of lay
caregivers in the community with a diverse range of education, Cuellar’s sample was
based on university students with 1 to 2 years of college education. Regardless of
the differences in percentages of unclassifiable persons in these two samples, an
orthogonal classification system is incomplete if it has a group of unclassifiable
persons. Furthermore, an orthogonal classification system is inadequate if the two
dimensions of acculturation, Hispanic and Anglo orientations, are highly correlated
because these correlated dimensions indicate a unidimensional construct of
acculturation. In sum, Cuellar’s classification system, although well-intentioned in
attempting to measure acculturation in a multidimensional manner, was inadequate
for obtaining acculturation types and an orthogonal measurement of acculturation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
The limitations of the study included the small sample size, the small to
moderate effect sizes obtained under conditions of low to moderate power, the
narrow range of acculturation obtained, and the traditional caregiving assignment
measure used. The question used in the traditional caregiving assignment measure
was ambiguous about how to include oneself and lacked information about the
nature of the family network. Ideally, one would want to know everyone in the
family (including fictive kin), whether the caregiver thought he or she should be the
primary caregiver, whether someone else should be the primary caregiver (and if so,
who), and whether the preference would be to share (if so, among whom). Exclusion
of caregivers who had moved or had phones disconnected could have resulted in
selective bias as these excluded caregivers might have been migrant laborers or
persons of low socio-economic status. As in other studies, this sample consisted of
caregivers who had sought help from community agencies and so the findings of this
study can only be generalized to Hispanic caregivers who sought help from
community agencies. Although this sample consisted of a help-seeking group
recruited through mainstream caregiving agencies, the breakdown of acculturation
types in this sample revealed a major strength of this study, a group of caregivers
that was not highly acculturated and that did not represent assimilated middle class
Hispanics, but rather, a group of traditional and marginalized Hispanic caregivers
that was quite fluent in the English language. Another strength of this study was the
low refusal rates among eligible caregivers. Consequently, although exclusion of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
caregivers who had moved or had phones disconnected could have resulted in
selective bias, the sample obtained was a major strength of this study.
In terms of recommendations for future research and theory development,
there is a need to replicate the absence of a cultural influence on the outcomes of
Hispanic caregivers as this is the first study to examine cultural influences in the
stress and coping model of Hispanic caregivers. Furthermore, the small sample size
and the over-representation of caregivers of Mexican origin did not permit
comparing the stress and coping process of caregivers representing different
Hispanic groups. The reasons for immigration to the U.S. (financial, war, family
reunification) could have greater influence on the experience of caregiving for some
Hispanic groups and not others. Consequently, there is a need to determine whether
the role of culture on caregiving is moderated by the specific Hispanic group one is
studying. The results of this study also point to the need to investigate whether
other factors in the process of caregiving (i.e. coping styles, appraisal of caregiving
as intrusive, beliefs about dementia) are influenced by culture. Finally, there is a
need to evaluate how the stress and coping model differs for those who are taking
care of a demented relative from those who are not; thus, future studies need to
include caregivers and non-caregivers.
In conclusion, the findings partially confirmed Aranda and Knight’s (1997)
argument for the incorporation of sociocultural factors in the stress and coping
process of ethnic caregivers. Familism, marianism, and fatalism were hypothesized
to be three cultural values salient in Hispanic culture; familism was the only cultural
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m
54
value that was associated with both acculturation orientations and lesser burden, and
that predicted burden when controlling for Hispanic orientation. Consequently, the
findings suggested that culture as measured by acculturation had little influence on
the outcomes of Hispanic caregivers but played a role in the process of caregiving by
affecting cultural values such as familism which had an effect on burden. Future
studies need to replicate the relative lack of a cultural influence on caregiving
outcomes and to determine whether there exist other cultural influences on the
process of caregiving.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
References
Aranda, M. P., & Knight, B. G. (1997). The influence of ethnicity and
culture on the caregiver stress and coping process: A sociocultural review and
analysis. The Gerontologist, 37, 342-354.
Bardis, P. D. (1959). A Familism scale. Marriage and Family Living, 21,
340-341.
Berry, J. W. (1980). Acculturation as varieties of adaptation. In A. M.
Padilla (Ed.), Acculturation: Theory, models, and some new findings (pp 9-25).
Boulder, CO: Westview.
Black, S. A., Markides, K. S., & Miller, T. Q. (1998). Correlates of
depressive symptomatology among older community-dwelling Mexican Americans:
The Hispanic EPESE. Journal of Gerontology: Social Sciences, 53B, S198-S208.
Bumam, M. A., Telles, C. A., Hough, R. L., & Escobar, J. I. (1987).
Measurement of acculturation in a community population of Mexican Americans.
Hispanic Journal of Behavioral Sciences, 9, 105-130.
Canabal, M. E., & Quiles, J. A. (1995). Acculturation and socioeconomic
factors as determinants of depression among Puerto Ricans in the United States.
Behavior and Personality, 23, 235-248.
Castro, F. G., Furth, P., & Karlow, H. (1984). The health beliefs of
Mexican, Mexican American and Anglo American women. Hispanic Journal of
Behavioral Sciences, 6, 365-383.
Cortes, D. E., Rogler, L. H., & Malgady, R. G. (1994). Biculturality among
Puerto Rican adults in the United States. American Journal of Community
Psychology, 22, 707-721.
Cox, C., & Monk, A. (1990). Minority caregivers of dementia victims: A
comparison of Black and Hispanic families. The Journal of Applied Gerontology, 9,
340-354.
Cox, C., & Monk, A. (1993). Hispanic culture and family care of
Alzheimer’s patients. Health and Social Work, 18, 92-99.
Cuellar, I., Arnold, B., & Gonzalez, G. (1995). Cognitive referents of
acculturation: Assessment of cultural constructs in Mexican Americans. Journal of
Community Psychology, 23, 339-356.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
Cuellar, I., Arnold, B., & Maldonado, R. (1995). Acculturation Rating Scale
for Mexican-Americans-II: A revision of the original ARSMA scale. Hispanic
Journal of Behavioral Sciences, 17, 275-304.
Cuellar, I., Harris, L. C., & Jasso, R. (1980). An acculturation scale for
Mexican American normal and clinical populations. Hispanic Journal of Behavioral
Sciences, 2, 199-217.
Gallagher-Thompson, D., Talamantes, M., Ramirez, R., & Valverde, I.
(1997). Service delivery issues and recommendations for working with Mexican-
American family caregivers. In G. Yeo and D. Gallagher-Thompson (Eds.),
Ethnicity and the dementias, (pp. 137-152). Bristol, PA: Taylor & Francis.
Golding, J. M., & Burnam, M. A. (1990). Immigration, stress, and
depressive symptoms in a Mexican-American community. Journal of Nervous and
Mental Disorders, 178, 161-171.
Golding, J. M., Burnam, M., Benjamin, B., & Wells, K. B. (1992). Risk
factors for secondary depression among Mexican-Americans and non-Hispanic
Whites. The Journal of Nervous and Mental Disease, 181, 166-175.
Griffin, J. (1983). Relationship between acculturation and psychological
impairment in adult Mexican Americans. Hispanic Journal of Behavioral Sciences, 5,
431-459.
Haley, W. E., Roth, D. L., Ford, G. R., West, C. A. C., Collins, R. P., &
Isobe, T. L. (1996). Appraisal, coping, and social support as mediators of well
being in Black and White family caregivers of patients with Alzheimer’s disease.
Journal of Consulting and Clinical Psychology, 64, 121-129.
Hinrichsen, G. A., & Ramirez, M. (1992). Black and White dementia
caregivers: A comparison of their adaptation, adjustment, and service utilization.
Gerontologist, 32, 375-381.
Kaplan, M. S., & Marks, G. (1990). Adverse effects of acculturation and
psychological distress among Mexican American young adults. Social Sciences and
Medicine, 31, 1313-1319.
Knight, B. G., Fox, L. S., & Chou, C. P. (in press). Factor structure of the
Burden Interview. Journal of Clinical Geropsychology.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
Knight, B. G., Silverstein, M., McCallum, T. J., & Fox, L. S. (2000). A
sociocultural stress and coping model for mental health outcomes among African-
American caregivers in Southern California. Journal of Gerontology: Psychological
Sciences, 55B, 3, P142-P150.
Kohout, F., Berkman, L., Evans, D., & Huntley, J. C. (1983, month?).
Psychosocial assessment devices adopted for population surveys of the elderly: Two
new forms of the CES-D. Paper presented at the annual meeting of the American
Public Health Association, Dallas TX.
Kranau, E. J., Green, V., & Valencia-Weber, G. (1982). Acculturation and
the Hispanic Woman: Attitudes toward women, sex-role attribution, sex-role
behavior, and demographics. Hispanic Journal of Behavioral Sciences, 4, 21-40.
Krause, N., & Goldenhar, L. M. (1992). Acculturation and psychological
distress in three groups of elderly Hispanics. Journal of Gerontology: Social
Sciences, 47, S279-S288.
Lang, J. G., Munoz, R. F., Bernal, G., & Sorensen, J. L. (1982). Quality of
life and psychological well-being in a bicultural Latino community. Hispanic Journal
of Behavioral Sciences, 4, 433-450.
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.
Leaper, C., & Valin, D. (1996). Predictors of Mexican American mothers’
and fathers’ attitudes toward gender equality. Hispanic Journal of Behavioral
Sciences, 18, 343-355.
Magana, J. R., Rocha, O. de la, Amsel, J., Magana, H. A., Fernandez, M. I.,
& Rulnick, S. (1996). Revisiting the dimensions of acculturation: Cultural theory
and psychometric practice. Hispanic Journal of Behavioral Sciences, 18, 444-468.
Marin, G., & Gamba, R. J. (1996). A new measurement of acculturation for
Hispanics: The bidimensional acculturation scale for Hispanics (BAS). Hispanic
Journal of Behavioral Sciences, 18, 297-316.
Masten, W. G., Penland, E. A., & Nayani, E. J. (1994). Depression and
acculturation in Mexican American women. Pychological Reports, 75, 1499-1503.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
Mintzer, J. E., Rubert, M. P., Loewenstein, D., Gamez, E., Millor, A.,
Quinteros, R., Flores, L., Miller, M., Rainerman, A., & Eisdorfer, C. (1992).
Daughters caregiving for Hispanic and non-Elispanic Alzheimer’s patients: Does
ethnicity make a difference? Community Mental Health Journal, 28, 293-302.
Negy, C., & Woods, D. J. (1992). A note on the relationship between
acculturation and socioeconomic status. Ehspanic Journal of Behavioral Sciences, 14,
248-251.
O’Hara, M. W., Kohout, F. J., & Wallace, R. B. (1985). Depression among
the rural elderly: A study of prevalence and correlates. Journal of Nervous and
Mental Disease, 173, 582-589.
Ortiz, V., & Arce, C. H. (1984). Language orientation and mental health
status among persons of Mexican descent. Hispanic Journal of Behavioral Sciences,
6, 127-143.
Paniagua, F. A. (1994). Assessing and treating culturally diverse clients: A
practical guide. Thousand Oaks, CA: Sage.
Pearlin, L. I., Mullan, J. T., Semple, S. J., & Skaff, M. M. (1990).
Caregiving and the stress process: An overview of concepts and their measures. The
Gerontologist, 30, 583-594.
Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal of
Health and Social Behavior, 19, 2-21.
Polich, T. M., & Gallagher-Thompson, D. (1997). Preliminary study
investigating psychological distress among female Hispanic caregivers. Journal of
Clinical Geropsychology, 3, 1-15.
Poulschok, S. W., & Deimling, G. T. (1984). Families caring for elders in
residence: Issues in the measurement of burden. Journal of Gerontology, 39, 230-
239.
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for
research in the general public. Applied Psychological Measurement, 1, 385-401.
Rothman, J., Gant, L. M., & Hnat, S. A. (1985). Mexican-American family
culture. Social Service Review, June, 197-215.
Sabogal, F., Marin, G., Otero-Sabogal, R., Marin, B. V., & Perez-Stable, E.
J. (1987). Hispanic familism and acculturation: What changes and what doesn’t?
Hispanic Journal of Behavioral Sciences, 9, 397-412.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
Schulz, R., Tompkins, C.A., & Rau, M. T. (1988). A longitudinal study of
psychosocial impact of stroke in primary support persons. Psychology and Aging, 3,
131-141.
Soto, E. (1983). Sex-role traditionalism and assertiveness in Puerto Rican
women living in the United States. Journal of Community Psychology, 11, 346-354.
Spence, J. T., Helmreich, R., & Stapp, J. (1973). A short version of the
Attitudes toward Women Scale (AWS). Bulletin of the Psychonomic Society, 2,
219-220.
Tran, T. V., Fitzpatrick, T., Berg, W. R., & Wright, R. JR. (1996).
Acculturation, health, stress, and psychological distress among elderly Flispanics.
Journal of Cross-cultural Gerontology, 11, 149-165.
Triandis, H. C., Marin, G., Betancourt, H., Lisansky, J., & Chang, B. (1982).
Dimensions of familism among Hispanic mainstream Navy recruits, Technical
Report No. 14, Department of Psychology, University of Illinois, Champaign.
Vega, W. (1990). Hispanic families in the 1980s: A decade of research.
Journal of Marriage and the Family, 52, 1015-1024.
Vega, W., Warheit, R., Buhl-Auth, J., Meinhardt, K. (1984). The prevalence
of depressive symptoms among Mexican Americans and Anglos. American Journal
of Epidemiology, 120, 592-607.
Wrinkle, R., Anzaldua, H., & Reed-Sanders, D. (1988). Analyses of scales
measuring self-esteem, life satisfaction, and mastery for Hispanic Elderly
Populations. In M. Sotomayor and H. Curiel (Ed.), Hispanic Elderly: A cultural
signature (pp. 185-202). Edinburg: Pan American University Press.
Youn, G., Knight, B. G., Jeong, H. S., & Benton, D. (1999). Differences in
familism values and caregiving outcomes among Korean, Korean American, and
White American Dementia caregivers. Psychology and Aging, 14, 1-9.
Zamanian, K., Thackrey, M., Starrett, R. A., Brown, L. G., Lassman, D. K.,
& Blanchard, A. (1992). Acculturation and depression in Mexican-American
elderly. Hispanic Aged Mental Health, 109-121.
Zarit, S. H. (1994). Research perspectives on family caregiving. In M. H.
Cantor (Ed.), Family caregiving: Agenda for the future (pp. 9-24). San Francisco:
American Society on Aging.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Zarit, S. H., Reever, K., & Bach-Peterson, J. (1980). Relatives of impaired
elderly: Correlates of feelings of burden. Gerontologist, 20, 373-377.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Ethnic identity in Parsee teenagers
PDF
Ethnic identity, acculturation, self-esteem and perceived discrimination: A comparison study of Asian American adolescents
PDF
Depression as a risk factor for dementia
PDF
The influence of personality, ethnicity, and culture on social support and emotional distress in dementia caregivers: A transactional stress and social support model
PDF
Dementia caregiving and ethnicity: African American caregivers and the sociocultural stress and coping model
PDF
Experimental evaluation of a new intervention designed to promote dementia caregivers' acceptance and empathic responding towards care -recipients
PDF
Depression and suicidality in Latino adolescents: A study of acculturation and gender role beliefs
PDF
Frequency and risk factors of poststroke dementia
PDF
Impact of language and culture on a neuropsychological screening battery for Hispanics
PDF
History of depression, antidepressant treatment, and other psychiatric illness as risk factors for Alzheimer's disease in a twin sample
PDF
Emergent literacy differences in Latino and African American children: Culture or poverty?
PDF
Identifying aggressive victims in Chinese children's peer groups
PDF
A stress model of acculturation, depressive symptoms, and self-esteem among Mexican-American undergraduates
PDF
Effects of parenting style and ethnic identity on European American and Asian Indian adolescents' academic competence and self esteem
PDF
Effects of cultural values and behaviors in sociocultural stress and coping among Korean-American and Korean caregivers
PDF
Considering the impact of culture on the therapeutic relationship: A look at Latino school-age children in a special education setting
PDF
Predictors of pretend play in Korean-American and Anglo-American preschool children
PDF
Head injury and dementia: A co-twin control study of Swedish twins
PDF
An integrative analysis of racism and quality of life: A comparison of multidimensional moderators in an ethnically diverse sample
PDF
Austronesian voyaging from Taiwan: Cultivating Amis folk songs on the international stage
Asset Metadata
Creator
Robinson, Gia Sophie
(author)
Core Title
Hispanic culture, acculturation, and distress among caregivers of dementia patients
School
Graduate School
Degree
Master of Arts
Degree Program
Psychology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
anthropology, cultural,OAI-PMH Harvest,Psychology, clinical,sociology, ethnic and racial studies
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Knight, Bob G. (
committee chair
), Farver, JoAnn (
committee member
), Gatz, Margaret (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-40186
Unique identifier
UC11341101
Identifier
1407922.pdf (filename),usctheses-c16-40186 (legacy record id)
Legacy Identifier
1407922.pdf
Dmrecord
40186
Document Type
Thesis
Rights
Robinson, Gia Sophie
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
anthropology, cultural
sociology, ethnic and racial studies