Acculturation, perceived social support, and health-related quality of life as factors in health-care-seeking behavior a...

MISSING IMAGE

Material Information

Title:
Acculturation, perceived social support, and health-related quality of life as factors in health-care-seeking behavior among African-American women
Physical Description:
vii, 127 leaves : ill. ; 29 cm.
Language:
English
Creator:
Reid, Alaycia DeNeen
Publication Date:

Subjects

Subjects / Keywords:
Psychology thesis, Ph.D   ( lcsh )
Dissertations, Academic -- Psychology -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2000.
Bibliography:
Includes bibliographical references (leaves 116-126).
Statement of Responsibility:
by Alaycia D. Reid.
General Note:
Printout.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 023003510
oclc - 45073978
System ID:
AA00013607:00001


This item is only available as the following downloads:


Full Text











ACCULTURATION, PERCEIVED SOCIAL SUPPORT, AND HEALTH-RELATED
QUALITY OF LIFE AS FACTORS IN HEALTH-CARE-SEEKING BEHAVIOR
AMONG AFRICAN-AMERICAN WOMEN

















By

ALAYCIA D. REID


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2000






















For my father, Ulysses Nathaniel Reid, and my mother,

Margaret Lovings Reid, whose unconditional love and support

have made this achievement possible.















ACKNOWLEDGMENTS

I am forever grateful to my advisor and friend, Dr. Carolyn

Tucker, for her tireless efforts and unwavering faith in

assisting me with this project from beginning to completion.

Through her guidance, support, encouragement, and, most of

all, patience, she has shown me the true meaning of the word

"excellence." Thanks are also in order for my committee

members Drs. Mary Fukuyama, Dorothy Nevill, Kim Walsh-

Childers, Franz Epting, and Robert Ziller whose thought-

provoking questions aided me in the development of this

project. A special thanks goes to John Dixon and Don Segal

for their invaluable assistance with the statistical

analyses. Last, but certainly not least, my deepest

appreciation goes to my family and friends, whose support

and encouragement over the years have enabled me to make my

dream become a reality.
















TABLE OF CONTENTS
page

ACKNOWLEDGMENTS............ .............................. iii

ABSTRACT ............ ......................... ........... vi

CHAPTERS

1 INTRODUCTION. ........... ................. ... ......... 1

Statement of the Problem ..................... ......... 2
Need for the Study......................... .. ........... 4
Purpose of the Study ....................... ........... 5

2 REVIEW OF LITERATURE ........................ .......... 7

Health-Care-Seeking Behaviors ............. ........... 7
Health Behavior Definitions................. ............ 7
Acculturation. .......... ................. .. .......... 19
Acculturation and Health................... .......... 21
Perceived Social Support................... .......... 27
Health-Related Quality of Life......................... 34

3 METHODOLOGY. ......... .................. .. ........... 38

Operational Definitions of the Variables............... 38
Hypotheses and Research Questions ....... .............. 40
Participants .......... ..................... ........... 42
Instruments .............. ............... .. ......... 44
Procedure................................. ............ 54
Phase 1: Training of Research Assistants............... 54
Phase 2: Recruitment of Participants .................. 56
Phase 3: Data Collection................... .......... 60

4 RESULTS ............. ................... .. .......... 63

Descriptive Data on all Major Variables ............... 63
Results from the Preliminary Pearson Correlation
Analysis ........................................... 63
Results from the Analysis to Test Hypotheses
1, 2 and 3 ......................................... 65
Results from the Analyses to Examine Research
Questions 1, 2, 3 and 4 ............................ .. 68









5 DISCUSSION ............... ................... .......... 73

Summary and Interpretation of the Results............... 74
Limitations of the Study.............................. .. 87
Implications for Practice and Future Research.......... 90
Conclusion ............. .................. .. ......... 91

APPENDICES

A Informed Consent Form .................................. 93

B Demographic Questionnaire................ ........... 95

C African-American Acculturation Scale ................. 96

D Multi-Dimensional Support Scale....................... 102

E RAND 36-Item Health Survey (Version 1.0) ............. 104

F Health-Care-Seeking Behavior Questionnaire............ 111

G Marlowe-Crowne Social Desirability Scale.............. 113

H Debriefing Form. ........... .............. ........... 115

REFERENCES ................. ............. .......... 116

BIOGRAPHICAL SKETCH .................. ..... .......... 127















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ACCULTURATION, PERCEIVED SOCIAL SUPPORT, AND HEALTH-RELATED
QUALITY OF LIFE AS FACTORS IN HEALTH-CARE-SEEKING BEHAVIOR
AMONG AFRICAN-AMERICAN WOMEN

By

Alaycia D. Reid

May 2000


Chairman: Carolyn M. Tucker
Major Department: Psychology

This study examined acculturation, perceived social

support and overall health-related quality of life (as well

as physical health and psychological well-being) as factors

in health-care-seeking behaviors among African-American

women. An assessment battery consisting of the following

instruments, was administered to 124 African-American women

in the southeastern part of the United States: (1) The

African-American Acculturation Scale (AAAS); (2) The family

and friends subscale of The Multi-Dimensional Support Scale

(M-DSS); (3) The RAND 36-Item Health Survey, version 1.0

(RAND); (4) The Health-Care-Seeking Behavior Questionnaire

(HCSBQ); (5) The Marlowe-Crowne Social Desirability Scale,









Short Form [M-C SDS (20)]; and (6) A Demographic

Questionnaire. Participants were recruited from churches,

colleges, businesses and organizations within various

African-American communities.

Controlling for the influence of the social

desirability variable with partial correlations, no

significant associations were found between any of the

independent variables (acculturation, perceived social

support, and health-related quality of life) and health-

care-seeking behaviors. In addition, results indicated that

there were no significant age or income-related differences

in levels of acculturation, perceived social support, or

overall health-related quality of life (as well as physical

health and psychological well-being). Finally, results of a

multiple regression analysis suggested that neither

acculturation, perceived social support, nor health-related

quality of life were significant predictors of heath-care-

seeking behaviors in African-American women. Analyses

indicated, however, that low but significant correlations

were found between (1)level of acculturation and

psychological well-being and (2) level of social support and

psychological well-being.















CHAPTER 1
INTRODUCTION


Health behavior research is indeed an emerging area of

interdisciplinary study. As such, promising opportunities

for the inclusion of counseling psychologists in health care

have been well documented over the past several years

(Alcorn, 1991; Klippel & DeJoy, 1984; May, 1977; Alcorn &

McPhearson, 1997). Although counseling psychologists have

demonstrated a growing interest in health care, some have

been hesitant to embrace opportunities in health-care

settings due to the more traditional role in educational

settings and from reservations regarding the fact that the

traditional medical model embraces a disease orientation

(Alcorn & McPhearson, 1997).

Over the years, there has been an on-going shift from a

more traditional medical (disease) model to a more

preventative model of health care. Thus, individuals are

encouraged to engage in proactive health behaviors and adopt

personal habits that would facilitate better health. As

such, interventions made available by counseling

psychologists have become increasingly salient to the health











enterprise (Alcorn & McPhearson, 1997). These interventions

have primarily focused on (1) identifying barriers to health

care and/or utilization of health-care services and (2)

finding ways in which to facilitate better health care

through improved health-care practices and/or increased

utilization of health care services.

Statement of the Problem

In recent years, several lines of research have focused

on the important role of individual health behavior in the

prevention of illness and premature mortality. Studies have

concluded that failure to seek preventive health care or

engage in preventive health-care behavior ultimately leads

to diminished physical health and/or higher mortality rates

(Belloc, 1973; Belloc & Breslow, 1972; Broman, 1993;

Ferraro, 1993; Mutchler & Burr, 1991;). Both

epidemiological and service use studies have shown

significant racial differences in the health status of adult

Americans (Ferraro, 1993). On most measures of health,

including life expectancy, morbidity, and mortality,

African-Americans evidence lower levels of health than their

White counterparts.

According to Cummings and Dehart (1995), a report

published by the Task Force on Minority Health in 1985

stated that a number of disparities have been identified











between the health status of ethnic groups and Whites in the

United States. Of all major causes of death combined -

including heart disease, diabetes, cancer and accidents -

African Americans showed a 1.5 percent greater likelihood of

dying in comparison to Whites. More importantly, African-

American females showed more than a 4 to 1 ratio over White

females for such deaths.

While health status among African-Americans has

improved over the past twenty years or so, the health

differential between Whites and African Americans remains

quite notable (Cummings & Dehart, 1995). Ironically, the

lower health status of African Americans as compared to

Whites has not resulted in increased research to understand

the health status of African Americans; there has especially

been a paucity of such research with African-American women

(Ahijevych & Bernhard, 1994).

Some of the limited research that has included African-

American women has focused on breast cancer a disease that

is of major concern to both White women and African-American

women. This research has shown that while African-American

women have lower incidence rates of breast cancer than do

White women, African-American women have a more advanced

stage at diagnosis and higher mortality in comparison to

White women (Lauver, 1994). These differences in stages at











diagnosis and mortality rates have been linked directly to

racial differences in health-care-seeking behavior. Other

health-related research involving African-American women has

included AIDS-related risky sexual behavior (Hines, Snowden,

& Graves, 1998; Klonoff & Landrine, 1997), smoking (Landrine

& Klonoff, 1997; Landrine & Klonoff, 1996), and hypertension

(Landrine & Klonoff, 1996)

Research that focuses on a particular segment or on a

particular population is vital in that it can enable

researchers to draw conclusions that are appropriate to the

specific cultural group under study (Sue, Chun & Gee, 1995).

For instance, research is needed to identify those variables

that may serve as barriers to health-care-seeking behavior

of African-American women. Thus, this study was exploratory

in nature and was designed to examine both physical and

psychological variables that influence the health-care-

seeking behavior of African-American women. Specifically,

this project was designed to assess acculturation, perceived

social support, and health-related quality of life as

factors that influence health-care-seeking behavior among

African-American women.

Need for the Study

The impetus for this study came from the literature

indicating a need for more in-depth research involving











African-American women and their health-care practices. The

current study, therefore, assessed potential psychological

factors (i.e., acculturation, perceived social support and

psychological well-being) and physical factors (self-

reported physical health) in the health-care-seeking

behavior of African-American women. The benefits that may

be gained from this research are: (1) identification of

areas for health psychologists and counseling psychologists

to target interventions to facilitate health-care-seeking

behavior among African-American women; (2) identification of

training for health-care professionals that will improve the

success of their health-care outreach programs in African-

American communities; and (3) provision of an impetus for

more culturally sensitive health-related research with the

potential of promoting health and preventing illness among

African-American women.

Purpose of the Study

The purpose of this study was to investigate potential

psychological and physical factors that are associated with

health-care-seeking behaviors among African-American women.

Specifically, this study examined (1) the relationship

between acculturation and health-care seeking behavior, (2)

the relationship between perceived social support and

health-care-seeking behavior, (3) the relationship between

health-related quality of life and health-care seeking











behavior, (4) the differences in levels of acculturation,

perceived social support, and health-related quality of life

in association with age and income, and (5) whether

acculturation, perceived social support, and health-related

quality of life are significant predictors of health-care-

seeking behavior.















CHAPTER 2
LITERATURE REVIEW

This literature review focuses on the literature

relevant to the variables that were investigated in this

study. The variables investigated were health-care-seeking

behavior, acculturation, perceived social support, and

health-related quality of life.

Health-Care-Seeking Behavior

The literature currently abounds with studies that have

focused on health behaviors and factors that influence them.

Several of these studies have focused on behaviors such as

cancer screening (Kang & Bloom, 1993; Kang, Bloom, & Romano,

1994), smoking cessation (Murry et al., 1995), and exercise

(Burk & Kimiecik, 1994). A common thread in most of these

studies is that they focus on specific behaviors that, when

practiced, result in or facilitate better general health.

What follows is a synopsis of some of the ways in which

health behavior has been viewed or studied in the past.

Health Behavior Definitions

Because there are no specific or rigid boundaries for

the term "health behavior," a specific definition for the

term has yet to emerge. However, based on earlier research,









8

a working definition of health behavior has been established

as

those personal attributes such as beliefs,
expectations, motives, values, perceptions, and other
cognitive elements; personality characteristics,
including affective and emotional states and traits;
and overt behavior patterns, actions and habits that
relate to health maintenance, to health restoration and
to health improvement. (Gochman, 1982, p. 169)

According to Gochman (1988), behavior denotes something

that one does or refrains from doing. This action is not

always conscious or voluntary. Furthermore, health behavior

does not encompass clinical improvement or physiological

recovery; rather, it includes analyses of specific behaviors

which, in turn, have an impact on improvement or recovery.

Finally, according to Gochman, this broad definition of

health behavior includes not only directly observable, overt

actions, but also those mental events and feeling states

that can be measured indirectly. A definition such as this

recognizes "that these personal attributes are influenced

by, and otherwise reflect family structure and processes,

peer group and social factors, and societal, institutional,

and cultural determinants" (Gochman, 1982, p. 169).

Many theories and models have been developed to explain

why some individuals engage in preventive measures and/or

other health-seeking behaviors. According to Nemcek (1990),

adoption of preventive behavior appears to be influenced by











the degree to which an individual values health, as well as

the degree to which that individual expects that the

behavior will influence health outcomes.

According to Kasl and Cobb (1966), health behavior can

be delineated into three major categories: (1) preventive

and protective behavior, (2) illness behavior, and (3) sick-

role behavior. Preventive and protective behaviors are

further defined as being primary or secondary. Primary

preventive and/or protective behaviors are those such as

eating well, managing weight, exercising, wearing seat

belts, obeying traffic laws, and following safety

regulations at work. Secondary preventive behaviors do not

prevent conditions from occurring; rather, they minimize the

impact of a condition by facilitating its early detection.

Health behaviors that fall into this category are those that

center around regularly scheduled examinations such as pap

smears, breast examinations, or dental exams.

The two remaining health behaviors as defined by Kasl

and Cobb (1966) illness behavior and sick-role behavior -

are somewhat similar in nature. Illness behavior comprises

those actions taken by individuals who are experiencing

signs and symptoms of illness whereas sick-role behavior

denotes those actions taken by individuals who already have

been classified by themselves or by others as being ill. In











the former case, behaviors can range from actively seeking

information and advice from those deemed as having health

expertise to passively waiting to see if the symptoms go

away. In the latter case, behaviors include actions related

to recovery and wellness such as adherence to and/or

acceptance of a medically prescribed regimen.

For the purpose of this study, health behavior is more

specifically defined as health-care-seeking behavior. While

health-care-seeking behavior encompasses facets of all three

types of health behaviors outlined by Kasl and Cobb (1966),

it primarily focuses on how likely an individual is to seek

medical attention to prevent illness or to address and/or to

receive follow-up care. As such, the goal of this study is

to see what types of variables are associated with African-

American women's health-care-seeking behaviors behaviors

that, when practiced regularly and consistently, may

facilitate good health.

Throughout the literature, there have been several

documentation of determinants of health services

utilization as well as other health-related behaviors. In

1991, it was noted that 21% of African-Americans reported

having no usual source of medical care. Instead, many

African-Americans choose to seek medical care only when

their medical symptoms became acute. This source of care is











usually that of an outpatient hospital setting such as a

hospital clinic or an emergency room. In these settings,

the lack of money or insurance presents less of an obstacle

than when seeking services in a physician's office. The

downside, however, is that this source of usual care does

not allow for continuity of care due to the sporacity of the

visits (National Center for Health Statistics, 1993). As

such, this type of medical care is seen as a "Band-Aid:"

when the symptoms are alleviated, many African Americans

cease to seek continued care (Flack, Amaro, Jenkins, Kunitz,

Levy, Mixon & Yu, 1995).

According to Flack, et al., (1995), cultural barriers

also significantly affect the rate of utilization of health

services among African Americans. For instance, there is

often a degree of discomfort or distrust by African

Americans of non-minority health-care providers. Part of

this discomfort stems from the lack of cultural awareness

and/or understanding on the part of the health-care

professional. Thus, due to the lack of practicing health-

care providers of the same ethnic background, African

Americans typically shy away from the medical attention they

need. If more efforts were structured around understanding

their culture and their lifestyle, African Americans might

be more receptive to efforts to improve their health status.











Throughout the literature, several studies have

examined health-care utilization patterns in particular.

Specifically, studies have examined the relationship between

health-care utilization and the following variables:

hypertension (Pavlik, Hyman, Vallbona, Toronjo, & Louis,

1997); race (Fichtenbaum, & Gyi-nr- T--c -r., 1997);

caregiving for others (Burton, Newsom, Schulz, Hirsch, &

German, 1997); alcohol use/frequency of use (Kunz, 1997);

and preventive health behaviors (Takemura, Hashimoto, &

Gunji, 1997). While most of these studies focus on health-

care utilization in terms of number of doctor's visits, or

compliance with a medical regime, other studies focus on a

broader sense of utilization of service in terms of

information-seeking as a major component of a personal

health practice repertoire.

In an effort to promote health and prevent disease,

individuals must search out new information, as having

adequate information is a significant contributor to

learning and adopting new health practices. In a study by

Rakowski, Assaf, Lefebvre, Lasater, Niknian, and Carleton

(1990), 281 adults ranging in age from 18 to 75 were

interviewed about a variety of personal health practices.

Results of this study indicated that women were more likely

than men to seek out information pertaining to health.











Furthermore, an association was found between information

seeking and favorable responses to several other health-

related practices. Interesting, however, is the fact that

using formal health services was the only type of health

practice that was not associated with information seeking.

According to documentation by the US Department of

Health and Human Services (1992), other factors that can be

attributed to differences in health-care utilization between

Blacks and Whites in particular are access to care,

dissatisfaction with past services and symptom perception.

In most cases, studies such as these have resulted in

inconsistent findings. However, few studies have actually

examined factors associated with health-care-seeking

behaviors (i.e., factors that determine whether one is

likely to engage in routine visits to a primary care

physician or whether one is likely to maintain follow-up

visits or be compliant with a prescribed medical regimen

once under a physician's care) among non-White populations

(Keith & Jones, 1990). The lack of research in this area is

especially true for African-American women.

According to Nemcek (1990), several studies have

investigated the correlation between the health-belief model

and the practice of breast self examination. While the

results have been encouraging, several gaps remain in the











findings due to inconsistencies. In addition, subjects in

these studies were limited primarily to Caucasian, middle-

income, high-school-educated women. According to Nemcek,

African-American women were typically not sampled, and

studies pertaining to the breast self examination and health

beliefs among this particular group were not located.

In their study on preventive health behavior among

adults, Rundall and Wheeler (1979) attempted to compare

three models financial constraints, health beliefs, and

system barriers for predicting preventive medical visits.

Data for this study were collected by means of a

retrospective cross-sectional survey of adults. These

adults were questioned about their beliefs regarding

susceptibility and severity of four diseases as well as the

benefits of professional intervention for preventing these

diseases. Findings indicated that susceptibility was the

only significant bivariate correlate of preventive visits.

Findings also indicated that "having a usual source of care"

was the single best predictor of preventive visits. This

"usual source of care" might be seen as an indirect measure

of barriers to care. In other words, individuals might be

less likely to engage in preventive measures if they did not

have a health-care facility or a primary care physician who

was readily available to them.











An exploration of the existing literature seems to

indicate that the majority of previous research on health

behaviors, beliefs, and attitudes was conducted using what

is known as a "deficit model" approach (Landrine & Klonoff,

1994). Research using a deficit model compares the

performance of African Americans and other non-majority

groups to Whites without giving consideration to the

determinants of this performance (i.e., factors directly

associated with the concept in question). In addition, the

performance of Whites is often used as the standard or basis

for such comparisons. What typically results is a case in

which African Americans and other non-majority groups

perform lower than their White counterparts. This lower

performance is then deemed as being inferior or deficit

performance.

The "difference model" approach to research (Oyemade &

Rosser, 1980) differs from the deficit model approach in

that it is sensitive to and respectful of cultural

differences between various ethnic groups. As such, it

separately examines factors or variables in target behaviors

of each cultural group. Thus, research using a difference

model approach focuses on identifying or isolating

determinants of differences among African Americans rather

than focusing on their deficits relative to Whites.











Because studies in the past that have examined racial

differences in health and health-care utilization between

Whites and African-Americans have typically done so using a

deficit model, the findings usually depict African-Americans

as having a significantly lower health-status and a lower

level of health-care utilization in comparison to Whites

(Blendon, Aiken, Freeman & Corey, 1989; Mutchler & Burr,

1991). Some factors associated with these differences

between African Americans and Whites are socioeconomic

status, education level, occupation, and access to a

physician. In addition, when compared to Whites, African

Americans are more likely to come up deficient in areas such

as economics (Manuel & Reid, 1982), general health (Travino

& Moss, 1984), and life satisfaction (Linn, Hunter, & Perry,

1979).

Because of the aforementioned disadvantages, it is

possible to assume that factors such as socioeconomic

status, health insurance coverage, perceived health status,

and psychological well-being may be more influential in

explaining utilization patterns for African Americans than

for Whites (Keith and Jones,1990). According to Fichtenbaum

and Gyimah-Brempong (1997) however, even when factors such

as these are equalized across race, significant differences

in the utilization of health-care services still remain.











When it came to health behaviors such as exercise, nutrition

practices, and stress management however, Felton, Parsons,

Misener, and Oldaker (1997) found that there were more

commonalities than differences between White and Black women

once socioeconomic status was taken into consideration.

Keith and Jones (1990) examined determinants of health

services utilization among Black and White elderly. Data

for this study were taken from the 1975 National Survey of

the Aged. This survey, one of the few to include a

representative number of African Americans for separate

analysis, attempted to describe the social support,

employment, assets and general living conditions of 2,143

older Americans. Of this number, only 413 were African

American.

Results of the Keith and Jones (1990) study indicated

the following: (1) African Americans were more likely to be

female, widowed, and have lower educational attainment; (2)

Whites were more likely to have participated in the labor

force; and (3) Whites had a definite advantage in terms of

income and health insurance. It also was determined that

racial differences in need for medical care were also

evident. Specifically, African Americans did not score as

high as Whites on the index of general morale and were more

likely to be lonely. In addition, African Americans seemed











to have a more negative outlook on life. They were more

likely to rate their health as being poor rather than good

or excellent and showed greater evidence of disability in

comparison to Whites (31 percent of African Americans versus

13 percent of Whites). Finally, racial differences in

utilization behaviors were found as well. It was determined

that African Americans were more likely to be hospitalized

and to have longer hospital stays than Whites. This was due

in part to the fact that African Americans, due to lack of

resources and other factors, tend to delay treatment until

illness is at a more advanced stage. According to Keith and

Jones, there is a need for research that separately examines

determinants of health-care utilization for African

Americans and Whites (i.e., a difference model approach).

Unlike most studies, one by Lauver (1994) did not find

differences in care-seeking behavior based on race. Lauver

proposed that a possible explanation for this finding is

that the African-American and Caucasian women in this study

were similar in terms of education, income, occupation,

resources for care, and psychosocial variables.

In a study by Pappas, Queen, Hadden, and Fisher (1993),

it was determined that when levels of SES are similar among

Blacks and Whites, racial differences in mortality rates are

still observed. It was noted that at similar levels of











educational attainment, mortality rates were higher among

Black men and women in comparison to their White

counterparts. These racial differences were especially

noted at the lower end of the SES hierarchy. Thus findings

from this research suggest that although ethnic differences

in SES are important, other factors are influential in

accounting for ethnic differences in morbidity and mortality

rates.

Due to (1) the lack of research that has examined

health behavior and/or health care-seeking behavior in a

cultural context, and (2) the number of discrepancies in

research findings that have examined health-related

variables across cultures, further studies are needed that

give weight to this important factor. This type of cultural

sensitivity was the goal of this research.

Acculturation

Acculturation, the extent to which ethnic-cultural

minorities participate in the cultural traditions, values,

beliefs, assumptions, and practices of the dominant White

society, has emerged as a promising, nonracist model for

explaining and understanding differences among various

ethnic groups (Landrine & Klonoff, 1995). According to the

concept of acculturation, one can be identified as

acculturated (adopting the practices and beliefs of the











dominant White society), traditional (remaining immersed in

one's own cultural traditions), or bicultural (participating

in the traditions of one's own culture and those of the

dominant White society. Research has shown that while

highly traditional ethnic minorities differ significantly

from Whites on a number of scales and behaviors, highly

acculturated minorities typically do not (Dana, 1993). As

such, many ethnic minorities can be understood in terms of

degree of acculturation without resorting to a deficit model

explanation. Thus, the concept of acculturation has the

potential to decrease racist beliefs about ethnic

differences and increase the understanding of such

differences as a manifestation of culture (Landrine &

Klonoff, 1995).

Because acculturation is seen as an essential precursor

to the understanding of cultural diversity in human

behavior, acculturation scales have been developed for a

variety of racial/ethnic groups including, but not limited

to, Latino Americans (Cuellar, Harris, & Jasso, 1980) and

Native Americans (Hoffman, Dana, & Bolton, 1985). While

these scales have existed for various ethnic groups such as

those mentioned, an acculturation scale did not exist for

African Americans until 1994 (Landrine & Klonoff, 1996).

According to Landrine and Klonoff, one reason for this lack









21

of development of an African-American acculturation scale is

that traditionally, psychology as a discipline tended to

regard African Americans as a racial group rather than as

both a racial group and an ethnic group. As such, African

Americans were viewed by traditional psychology and society

as persons without a culture. Many researchers in the area

of Black Psychology, however, have provided empirical

evidence for a distinct African-American culture, which led

to the development of the African-American Acculturation

Scale (see Landrine & Klonoff, 1994).

Acculturation and Health

Research has found strong relationships between

acculturation and stress (Berry & Annis, 1974; Berry,

Uichol, & Mok, 1987), hypertension (Dressler, Mata, Chavez,

& Viteri, 1987), neuropsychological test performance in

normal and HIV-positive individuals (Manly, Miller, Heaton,

Byrd, Reilly, Velasquez, Saccuzzo & Grant, 1998), sexual

(HIV) risk-taking (Newcomb, Wyatt, Romero, Gloria, Tucker,

Wayment, Carmona, Solis & Mitchell-Kernan, 1998) and

psychiatric disorders (Burnam, Hough, Karno, Escobar, &

Telles, 1987). Such research indicates the importance of

studying the cultural context of health behavior. As such,

it provides clear information as to how to tailor











interventions to be more effective across ethnic groups

(Landrine & Klonoff, 1994).

While there are no known studies to date that have

examined the relationship between acculturation and health-

care-seeking behaviors as defined by this study, studies do

exist that have examined acculturation and general health

behavior/health status. Maxwell, Bastani, and Warda (1998)

examined the relationship between mammography utilization

and acculturation among Korean-American women. In this

study, a convenience sample of 229 predominately low-income

Korean women were used. Face-to-face interviews were

conducted with this sample of women, who ranged in age from

50 to 75 years old. Of this sample, 49% had never had a

mammogram, while 24% had received a mammogram within the

past twelve months, and 36% had received a mammogram within

the past two years. Results indicated that variables

positively associated with receiving a mammogram were as

follows: having health insurance, having an income greater

than $25,000, having received a recommendation to have a

mammogram by a physician, holding positive group norms and

greater acculturation. Findings also revealed that fear of

finding cancer, time factors, transportation difficulties,

embarrassment, and discomfort with the physician were











variables that were negatively associated with receiving a

mammogram.

Several studies have examined the relationship between

health behavior/health status and acculturation among

African Americans in particular. In a study by Hines,

Snowden, and Graves (1998), the relationship between

acculturation, alcohol consumption, and AIDS-related risky

sexual behavior was examined. The participants in this

study consisted of 470 African-American women from a

national probability sample. Results revealed that there

was a significant negative relationship between

acculturation and drinking. Specifically, the heaviest

drinkers were the least acculturated. Results of the study

also indicated that there was a direct relationship between

risky sexual behavior and acculturation. In other words,

those individuals who were high in acculturation were also

high in risky sexual behavior.

Two well-documented studies that have examined

acculturation and health-behavior/health status among

African Americans have focused on smoking the single most

preventable cause of death (U.S. Department of Health,

Education, and Welfare, 1979) and hypertension one of the

most serious health problems among African Americans

(Hildreth & Saunders, 1992; Polednak, 1989).









24

Smoking-related health problems such as cancer, stroke,

hypertension, and cardiovascular disease are foremost among

the major health problems of African-Americans (Centers for

Disease Control, 1989). In addition, these smoking-related

health problems are more prevalent among African-Americans

than Whites in general (Hildreth & Saunders, 1992) and

African-American women in particular (Klonoff, Landrine, &

Scott, 1995). As such, a study was conducted by Landrine

and Klonoff (1996) that examined the role of African-

American acculturation in cigarette smoking among adults.

Four hundred and forty-four African-American adults

participated in this study.

The sample completed a questionnaire consisting of the

African-American Acculturation Scale (AAAS), demographic

questions, and a question assessing whether or not the

participant smoked. Results indicated that African-American

smokers were more traditional (less acculturated) than

African-American nonsmokers. Furthermore, it was found that

acculturation was a better predictor of smoking than status

variables such as income and education level. Results

revealed that the prevalence of smoking among traditional

African Americans was 33.6% in comparison to the prevalence

of smoking among acculturated African Americans, which was

15.3%. According to Klonoff and Landrine, these findings











suggest the need for further exploration of the role of

acculturation in African-American health and health-related

behavior.

A related study conducted by Landrine and Klonoff

(1997) also examined the role of African-American

acculturation in cigarette smoking among adults. Five

hundred and twenty African-American adults participated in

this study. The sample completed the African-American

Acculturation Scale along with questions on smoking obtained

from the National Health Interview Survey. Results again

indicated that smokers scored higher than nonsmokers on the

AAAS total score, indicating a more traditional cultural

orientation. Specifically, results revealed that smokers

had more traditional health beliefs, more strongly endorsed

cultural superstitions, had more traditional interracial

attitudes (greater distrust of Whites), and were more likely

than nonsmokers to prepare and consume traditionally

African-American foods.

In a similar study, Landrine and Klonoff (1996)

examined the relationship between hypertension and

acculturation among African-American adults. Participants

consisted of 153 African-American adults who completed a

questionnaire consisting of the AAAS, demographic questions,

and a question that read "Has your doctor ever told you that









26

you have high blood pressure?" which was answered yes or no.

Results indicated that hypertensives (those who responded

yes to the blood pressure question) scored higher than

normotensives (those who responded no to the blood pressure

question) on the total AAAS in general and two subscales

(Traditional African American Childhood Socialization and

Foods) in particular. Thus, hypertensives tended to be more

traditional and normotensives tended to be more

acculturated.

Although several studies have focused on acculturation

and health behaviors or health status in general, few, if

any have examined acculturation and its relationship to

health-care-seeking behaviors as defined in this study. Due

to the lack of research in this area, the goal of the

present study was to examine acculturation as a factor

associated with health-care-seeking behavior. In doing so,

this study provided a means for investigating and

interpreting the health-care-seeking behavior of African-

American women in a culturally sensitive, rather than a

racist manner. In addition,

the concept of acculturation provides a rudimentary but
parsimonious theoretical framework for predicting the
nature and the direction of ethnic differences; without
such predictions, empirical findings are interesting,
but nonetheless serendipitous and their practical
applications subsequently limited. (Landrine & Klonoff,
1994, p. 105)











Perceived Social Support

Social support has been demonstrated to affect a number

of different health-related behaviors. There are a variety

of different types of social support, which range from more

artificial systems (those that are time-limited and not

normally a part of one's environment) to natural systems

(those such as family, friends, and co-workers who are part

of one's typical environment) (see Murray, Johnston, Dolce,

Lee, & O'hara, 1995). Because of the varied ways of

defining social support and the various types of support,

the literature on social supports in health demonstrates a

lack of clarity about what the term social support means.

Traditionally, it has been used to indicate the different

ways in which one is linked to others. Problems occur,

however, when comparing conclusions from studies that have

used social support in different ways.

Social support has been conceptualized along several

dimensions, including existence (designation of the presence

or absence of a supportive relationship), network structure

(where the support comes from), and type (Depner,

Wethington, & Ingersoll-Dayton, 1984). In addition, some

studies consider social support in terms of demographic or

family-related factors such as marriage/partnership or

living arrangements (Eaton, 1978). Finally, according to











Hirsch (1981) and Weiss (1974), it is imperative to know

information pertaining to the source of the support as well

as the type of support. Information such as this would be

helpful in determining whether or not a particular type of

social support is indeed significant.

Despite the lack of conceptual clarity, a number of

studies in the literature have investigated the relationship

between social support and health behaviors. Social support

has been positively related to never smoking, exercising,

desirable weight, and adequate sleep (Gottlieb & Green,

1984). In addition, it has been associated with smoking

cessation (Coppatelli & Orleans, 1985; Murray, Johnston,

Dolce, Lee, & O'hara, 1995), decreased alcohol use (Kline,

Canter, & Robin, 1987), readiness to make positive dietary

changes (Sorensen, Stoddard, & Macario, 1998), exercise

(Steptoe, Wardle, Fuller, Holte, Justo, Sanderman, &

Wichstrom, 1997), quality of life (Ota & Tanaka, 1997),

diabetes mellitus (Ford, Tilley, & McDonald, 1998) and

decreased stress (Smith, 1985).

In a study by Milligan, Burke, Beilin, Richards,

Dunbar, Spencer, Balde, and Gracey (1997), psychosocial

variables associated with various health-related behaviors

were examined. Participants in this study included 301

Australian men and 282 Australian women (age 18). Results











indicated that barriers for engaging in desirable levels of

physical activity included lack of social support. Lack of

family support in particular was perceived as a barrier to

smoking cessation.

Schaffer and Lia-Hoagberg (1997) examined the effects

of social support on prenatal care and other pregnancy-

related health behaviors of low-income women. Data for this

study were collected in five metropolitan prenatal clinics

serving low-income women. Participants included 101

ethnically diverse, primarily single, low-income pregnant

women who were between 28 and 40 weeks of pregnancy.

Participants completed the Norbeck Social Support

Questionnaire, the Prenatal Health Questionnaire, and the

Demographic/Pregnancy Questionnaire. Results indicated that

social support provided by the partner correlated positively

with adequacy of prenatal care (e.g., visits to the health-

care professional), whereas social support from others

correlated positively with prenatal health behaviors (e.g.,

exercise and eating properly).

In a study examining the relationship between social

support and breast self-examination (Wagle, Komorita, & Lu,

1997), twenty-two women age 55 and older completed the

Norbeck Social Support Questionnaire and two other measures

that were used to assess the accuracy and frequency of











breast self-examination. Results indicated that social

support was indeed significantly related to the frequency of

breast self-examination. However, social support was not

significantly related to accuracy of breast self-

examination. These findings suggest that women's health may

indeed benefit from having supportive relationships.

The role of social support also has been studied in

African Americans. Specifically, studies have examined

social support in African-American adolescents (Coates,

1985; Cauce, 1986), African-American elderly (Marcus-

Bernstein, 1986), low-income African Americans (Ball, 1983),

and African Americans who are physically challenged

(Belgrave & Walker, 1991). These studies typically reveal a

positive relationship between social support and mental and

physical health outcomes in African Americans (Belgrave &

Lewis, 1994).

In a study by Belgrave and Lewis (1994), the role of

social support in compliance and other health-related

behaviors was investigated. Subjects consisted of 49

African Americans with sickle cell disease and 78 African

Americans with diabetes. Subjects were administered a

battery of questionnaires that assessed appointment-keeping

behaviors, adherence to health activities, and social

support. Results indicate that social support was











significantly associated with appointment-keeping behavior

and adherence to health activities in both medical

populations.

In research examining the role of social support among

African-Americans in general, women were found to have

significantly larger and more extended informal social

support networks than men (Barker, Morrow, & Mitteness,

1998). Little research has been done, however, to determine

the role that social support plays for African-American

women related to their health-care behaviors. In response

to this lack of research, Kang, Bloom, and Romano (1994)

examined the role of social support and social ties in

cancer screening among African-American women. Four

different measures of social ties were included: marital

status, number of relatives and friends, church

participation, and participation in other organizations.

Results indicated that women with more social ties were more

likely to have had a routine mammogram than those with fewer

social ties. This was true even after variables such as

health status, age, health insurance, and having a primary

source of care were controlled for.

Although significant associations were found between

social ties and cancer screening, no associations were found

between social support and the use of cancer screening.











According to Kang, Bloom and Romano (1994), the social

support measures included a small number of items; therefore

they may not have captured the qualitative features of

supportive social interactions. The authors suggest that

functional aspects of social support may be less important

than the social ties themselves. Despite the limitations in

this study, the findings suggest that African-American women

with larger social networks are more likely to receive a

mammogram than those with smaller social networks. Thus

interventions that make use of social ties may indeed play a

significant role in promoting early cancer detection.

According to some researchers, studies have

consistently shown that Whites have access to more social

support than other ethnic groups such as African Americans

and Asians (Jay and D'Augelli, 1991; Koniak, Griffin,

Lominska and Brecht, 1993; Williams, 1993). However, as

with many other variables, the literature pertaining to

perceived social support among African Americans is

contradictory. A study by Williams (1993) concluded that

African Americans perceive their support as being more

adequate than do Whites. On the other hand, Koniak,

Griffin, Lominska and Brecht (1993) reported that African

Americans, in comparison to Whites, perceived less

availability of social support. A third study by











Silverstein and Waite (1993) revealed no significant

differences in African Americans' or Whites perceptions of

support. As is common, however, research focused on

African-American women in particular is lacking in this

area.

A search of the literature has shown that there are

indeed conflicting results on how social support and/or

social ties are associated with health behaviors and the

utilization of health-care services. Specifically, social

support or social ties has been positively associated

(Zapka, Stoddard, Costanza, & Greene, 1989), negatively

associated (Romano, Bloom, & Syme, 1991), or not associated

at all (Rutledge, 1987) with health behaviors or health-care

services utilization. Because of the disparities in the role

of social support for African Americans in general and the

lack of research on social support and African-American

women in particular, further research is needed in this area

- especially in the area of social support and health-care-

seeking behaviors among African-American women.











Health-Related Quality of Life

Several studies throughout the literature have examined

health-related quality of life and its relationship to other

health-related variables. Specifically, health-related

quality of life has been studied in association with

personality type (Yamaoka, Shigehisa, Ogoshi, Haruyama,

Watanabe, Hayashi, & Hayashi, 1998), pain and discomfort

(Skevington, 1998), aftermath of stroke (Williams, 1998),

obesity (Barofsky, Fontaine, & Cheskin, 1997), human

immunodeficiency virus (HIV) (Chan & Revicki, 1998), and

depression (Leidy, Palmer, Murray, Robb, & Revicki, 1998).

As with health behavior and social support, however, there

are many discrepancies in terms of how health-related

quality of life is defined and/or measured.

It has been noted in the literature that there is often

confusing and unclear terminology in regard to the concept

of health-related quality of life (Leplege & Hunt, 1998).

One problem that exists is that of confounding health-

related quality of life with quality of life in general.

According to Albert (1998), health-related quality of life

measures are typically more highly correlated with health

status. In addition, health-related quality of life

measures are more sensitive to changes in health than more

general quality of life measures. Health-related quality of











life, for purposes of this study, refers to one's overall

health-related quality of life, as well as one's physical

health and one's psychological well-being.

In a study by Skevington (1998), the impact of pain on

quality of life was investigated in 320 healthy participants

and participants experiencing a range of various illnesses.

The mean age of the participants was 44.9 years. In this

study, quality of life was assessed using a

multidimensional, multilingual generic profile designed for

cross-cultural use in health care. Results indicated that

pain and discomfort significantly impacted perceptions of

general quality of life as it relates to health.

In a related study, Barofsky, Fontaine, and Cheskin

(1997) examined the impact of pain on health-related quality

of life in medically obese individuals. As reported by the

authors, obesity is a major health concern as it is often

associated with increased health risks, chronic pain,

decreased functional health status and overall well-being.

In this study, 312 individuals seeking medically supervised

treatment for weight loss completed a sociodemographic

questionnaire and the Medical Outcomes Study Short-Form

Health Survey (SF-36). In addition, these individuals

underwent a series of clinical evaluations. Participants

reported experiencing at least moderate pain in the four











weeks prior to treatment. After controlling for

sociodemographic factors, body-mass index and depression,

obese participants reporting pain scored significantly lower

on all SF-36 domains than those not reporting pain. These

findings suggest that pain itself is independently

associated with impaired health-related quality of life in

almost half of the individuals seeking weight-loss

treatment.

According to Congress and Lyons (1992), minority

status and/or ethnicity exerts an influence on one's quality

of health and perceptions of well-being. In addition,

Halpern (1993) suggests that minority status might be a

plausible risk factor for poor physical health due to

variables typically associated with minority status (e.g.,

low income and lack of access to medical care).

Furthermore, Gove (1984) asserts that the causes of women's

morbidity rates are multiple and complex and center around

the various role demands women assume. These morbidity

rates might be especially high among non-White women

considering the fact that minority status alone brings with

it increased stressors in the forms of discrimination and

prejudice (Halpern, 1993).

Because of the noted differences in health-related

quality of life between White women and African-American









37

women, this study examined whether health-care-seeking

behavior is associated with health-related quality of life

among African-American women. Specifically, this study

examined overall health-related quality of life and how it

relates to health-care-seeking behavior. In addition, this

study examined two specific components of health-related

quality of life (physical health and psychological well-

being) and their relationship to health-care-seeking

behavior among African-American women.















CHAPTER 3
METHODOLOGY


This chapter presents the following information:

operational definitions of the variables in the study,

hypotheses and research questions, subjects, instruments,

and procedure.

Operational Definitions of the Variables

Acculturation is defined in the literature as the

degree to which ethnic and/or cultural minorities

participate in the cultural traditions, values, beliefs, and

practices of their own culture versus those of the dominant

or White society (Landrine and Klonoff, 1996). In the

present study, acculturation for African-American women is

operationally defined as the score on the African-American

Acculturation Scale (AAAS) (Appendix C). The African-

American Acculturation Scale is a continuous scale with

lower scores indicating more acculturation and higher scores

indicating more traditional beliefs and behaviors.

According to the authors, it is not possible to know what

scores in the midrange indicate.











Social Support is defined in the present study as the

degree to which one feels support from significant people in

her life. Social support is operationally defined as the

score on the Family and Friends subscale on the Multi-

Dimensional Support Scale (M-DSS) (Appendix D). The M-DSS

measures perceived availability of social support and

satisfaction with social support from three potential

sources family and friends, peer group, and supervisors.

Health-related quality of life is defined in the

present study as one's overall perception of the quality of

her life. In addition, particular attention is given to

one's perception of her physical health and her

psychological well-being two components of overall health-

related quality of life. As such, health-related quality of

life is operationally defined in the following manners:

(1)Health-related quality of life is operationally defined

as the total score on the RAND 36-Item Health Survey

(Version 1.0); (2) health-related quality of life is

operationally defined as the score on the Physical

Functioning (physical health) subscale of the RAND 36-Item

Health Survey (Version 1.0); and (3) health-related quality

of life is operationally defined as the score on the

Emotional Well-Being (psychological well-being) subscale of

the RAND 36-Item Health Survey (Version 1.0) (Appendix E).











Health-care-seeking behavior is defined in the present

study as the degree to which one actively engages in health-

care behavior, actively seeks out health-care services

and/or maintains follow-up care once under a health-

provider's care. Health-care-seeking behavior is

operationally defined as the score on the Health-care-

seeking Behavior Questionnaire (HCSBQ) (Appendix F).

Hypotheses and Research Questions

This study tested the following hypotheses:

(1) Degree of acculturation (AC) is significantly associated

with health-care-seeking behaviors (HCSB) such that the

greater the degree of acculturation, the greater the level

of engagement in health-care-seeking behaviors.

(2) Perceived social support (SS) is significantly

associated with health-care-seeking behaviors(HCSB) such

that the greater the level of perceived social support, the

greater the level of engagement in health-care-seeking

behaviors.

(3) Overall health-related quality of life (HRQL), as well

as physical health (PHY) and psychological well-being

(PSYCH) two components of health-related quality of life -

are significantly associated with health-care-seeking

behaviors (HCSB) such that the greater the level of overall

health-related quality of life, as well as physical health











and psychological well-being, the greater the level of

engagement in health-care-seeking behaviors.

The possible effects of age (AGE) and total household

income (INCOME) on the major variables of study were

addressed in the following research questions:

(1) Is there a significant difference in level of

acculturation (AC) in association with age (AGE) or total

household income (INCOME)?

(2) Is there a significant difference in level of perceived

social support (SS) and level of overall health-related

quality of life (HRQL) (as well as physical health (PHYS)and

psychological well-being (PSYCH)) in association with age

(AGE) or total household income (INCOME)?

(3) Is there a significant difference in level of health-

care-seeking behaviors (HCSB)in association with age (AGE)

or total household income (INCOME)?

The possible association between level of

acculturation, level of social support, and level of health-

related quality of life (overall health-related quality of

life, as well as physical health and psychological well-

being) and health-care-seeking behavior was examined in the

following research question:

(4) Do acculturation (AC), perceived social support (SS), or

health-related quality of life (overall (HRQL), as well as











physical health (PHYS) and psychological well-being (PSYCH))

predict health-care-seeking behaviors (HCSB)?

Participants

Participants in this study consisted of 124 African-

American women (age 21 and older) recruited from various

groups and organizations (e.g., churches, community

organizations, and community health centers) in two cities

in Georgia that were easily accessible to the researcher

involved. Both of these cities have high concentrations of

African-American women of varying socioeconomic status.

African-American women from the community were used for this

project, as opposed to college students, to increase the

likelihood that the results would be generalizable to the

majority of African-American women. Participants in this

study were paid volunteers who received $10 each for their

participation.

The demographic characteristics of the participants in

this study are shown in table 3.1. Twenty-three

participants were between the ages of 21 and 30 (18.5%), 30

were between the ages of 31 and 40 (24.3%), 37 were between

the ages of 41 and 50 (29.8%), 16 were between the ages of

51 and 60 (12.9%), and 18 were age 61 or older (14.5%).

Approximately seventy-four percent (74.2%) of the

participants reported full-time paid employment (n = 92),

11.3% reported part-time paid employment (n = 14), and 14.5%











Table 3.1

Summary of Demographic Data



Variable Frequency Percentage



Age of Participant

21 30 23 18.5
31 40 30 24.3
41 50 37 29.8
51 60 16 12.9
> 61 18 14.5

Paid Employment Status

Full Time 92 74.2
Part Time 14 11.3
None 18 14.5

Relationship Status

Single 26 21.1
Partnered 7 5.6
Married 54 43.5
Legally Separated 6 4.8
Legally Divorced 17 13.7
Widowed 14 11.3

Current Total Household Income

< $15,000 27 21.8
$15,001 $30,000 31 25.0
$30,001 $45,000 28 22.6
> $45,000 38 30.6



reported no paid employment (n = 18). Twenty-six (26) of

the participants were single (21.1%), 7 were partnered

(5.6%), 54 were married (43.5%), 6 were legally separated

(4.8%), 17 were legally divorced (13.7%), and 14 were











widowed (11.3%). In terms of their total household income,

27 participants had incomes of $15,000 or less (21.8%), 31

had incomes between $15,001 and $30,000 (25.0%), 28 had

incomes ranging between $30,001 and $45,000 (22.6%), and 38

had incomes greater than $45,000 (30.6%).

Instruments

The Health Behavior Assessment Battery (HBAB) used in

this study included the following instruments:

1. Demographic Questionnaire (DQ) (Appendix B)

The Demographic Questionnaire was used to obtain the

following information on each participant: age, paid

employment status, relationship status, and total household

income.

2. African-American Acculturation Scale (AAAS; Landrine &

Klonoff, 1994) (Appendix C)

The African-American Acculturation Scale was used to

assess degree of acculturation among African-American women.

The AAAS is a 74-item scale consisting of eight subscales

which are: (1) Preference for African-American Things (e.g.,

"I feel more comfortable around Blacks than around

Whites."); (2) Traditional Family Practices and Values

(e.g., "When I was young, my cousin, aunt, grandmother, or

other relative lived with me and my family for a while.");

(3) Traditional Health Beliefs, Practices, and Folk











Disorders (e.g., "I was taught that you should not take a

bath and then go outside."); (4) Traditional Socialization

(e.g., "I went to a mostly Black elementary school."); (5)

Traditional Food and Food Practices (e.g., "I save grease

from cooking to use it again."); (6) Religious Beliefs and

Practices (e.g., "The church is the heart of the Black

community."); (7) Interracial Attitudes (e.g., "IQ tests

were set up purposefully to discriminate against Black

people."); and (8) Superstitions (e.g., "I eat black-eyed

peas on New Year's Eve.").

The eight theoretically defined subscales of the AAAS,

as well as the total AAAS score, have group differences and

concurrent validity. The subscales also have high internal

consistency reliability, ranging from .71 to .90. The scale

as a whole has high split-half reliability (r=.93) (Landrine

& Klonoff, 1995).

Items on the scale are rated on a 1-7 scale with 1 = "I

totally disagree Not true at all" and 7 = "I strongly

agree Absolutely true." All eight subscales were used for

this study; thus, a total sum score was calculated across

each of the eight subscales for each participant's score.

Participants' scores ranged from a low of 247 to a high of

464 out of a possible score of 518. The sample mean score

was 356.19 and the standard deviation was 44.77. According











to Landrine and Klonoff's (1996) validation study, the mean

and standard deviation as reported for Blacks is 343.01 and

60.76 respectively.

According to Landrine and Klonoff (1996), very high

scores on the AAAS indicate that a person is highly

traditional, whereas very low scores indicate that a person

is highly acculturated. Because the AAAS only measures

degrees of immersion in the African-American culture,

comparisons can be made only between the highly acculturated

and the highly traditional. This particular scale cannot

address biculturalism, multiculturalism, or the meaning of

midrange scores on the scale (Landrine and Klonoff, 1996).

In order for this to occur, scores on the AAAS would have to

be compared to scores on an instrument that measured

European American culture. Landrine and Klonoff believe

that those individuals who scored equally high on both a

European acculturation scale as well as the AAAS could then

be classified as bicultural. Those who scored low on both

scales would be classified more or less as marginal.

3. The Multi-Dimensional Support Scale (M-DSS; Winefield,

Winefield, & Tiggemann, 1992) (Appendix D)

The Family and Friends subscale of the Multi-

Dimensional Support Scale was used to assess level of social

support. The M-DSS measures perceived availability of











social support and satisfaction with social support from

three potential sources family and friends (or other

confidants and/or attachment figures), peer group (or those

who are on a similar level or facing the same challenges as

the respondent), and experts (or those who have some

official role over the respondent such as a supervisor).

The M-DSS is described as a flexible instrument as it

can be adapted for use with various populations. Thus, when

using this scale, it is imperative that potential sources of

support be selected on the basis of the recipient group

being studied (Winefield, Winefield, and Tiggemann, 1992).

The Family and Friends subscale of the M-DSS was chosen for

this particular study because the participants were

volunteers from the general population. As such, not all of

them would be in a situation where they had either a peer

group or a person in authority over them (such as at school

or at work).

The M-DSS as a whole is a 38-item scale consisting of

19 two-part core questions. The Family and Friends subscale

consists of 7 two-part questions. The first part of each

question is answered on a scale of 1 to 4 with 1 = Never, 2

= Sometimes, 3 = Often, and 4 = Usually/Always (e.g., "Think

of your family and close friends, especially the 2-3 who are

most important to you. How often did they really listen to











you when you talked to them about your concerns or

problems?"). The second part of each question is answered

by choosing one of the following three choices: "more

often"; "less often"; or "it was just about right" (e.g.,

"And you would have liked them to do this"). In scoring,

"more often" = 1, "less often" = 2, and "it was just about

right" = 3. A total score is obtained for each participant

by adding together the scored responses from each two-part

question. Higher scores on the M-DSS indicate higher levels

of perceived social support.

The M-DSS is reported to be both sensitive and

reliable. Internal reliability coefficient alphas range

from .80 to .91. Subscale scores have been found to

correlate significantly with affect measures. While there

are currently no normative data available, some norms are in

preparation. The sample mean of the Family and Friends

subscale was 36.76 and the standard deviation was 7.92.

Sample scores ranged from a low of 19 to a high of 49 out of

a possible 49.

4. The RAND 36-Item Health Survey 1.0 (RAND; Ware, Jr. &

Sherbourne, (1992) (Appendix E)

The RAND 36-Item Health Survey 1.0 was used to measure

overall level of health-related quality of life (as well as

physical health and psychological well-being two separate











components of overall health-related quality of life). The

RAND is a 36-item self-report inventory that assesses eight

health concepts that often are used to determine one's

health-related quality of life. The eight concepts or

subscales included in this scale are as follows: physical

functioning, bodily pain, role limitations due to physical

health problems, role limitations due to personal or

emotional problems, emotional well-being, social

functioning, energy/fatigue, and general health perceptions.

The RAND utilizes Likert-type scales for scoring

purposes with responses ranging from a score of 0 to a score

of 100. A sample item from the RAND reads "Compared to one

year ago, how would you rate your health in general now?".

This particular question is scored on a scale of 1 to 5

where 1 =100, 2 = 75, 3 = 50, 4 = 25, and 5 = 0. Alpha

reliabilities for the subscales of the RAND range from .78

to .93. For purposes of this research, a mean score was

averaged across all eight subscales for an overall health-

related quality of life score. No normative data was found

for this scale. The sample mean for the overall health-

related quality of life scale was 78.92 with a standard

deviation of 11.60.

In addition to the overall health-related quality of

life score, two separate scores were obtained from two of











the eight subscales: Physical Functioning and Emotional

Well-being. The Physical Functioning subscale was used to

look at physical health a component of health-related

quality of life. The Physical Functioning subscale is an

indirect measure of physical health in that it measures

one's perception of limitations in various physical roles or

functions as a result of her health. The assumption is that

those who are in better physical health will experience

fewer physical limitations as a result of their health. In

addition, the Emotional Well-being subscale was used to

examine level of psychological well-being also a component

of health-related quality of life. Due to the various role

demands on African-American women and the physical and

psychological stress that results from these demands, it was

deemed that the physical functioning and emotional well-

being subscales would be of particular interest in terms of

understanding more fully respondents' perception of their

health-related quality of life.

The alpha reliability for the Physical Functioning

subscale (physical health) is .93, and the alpha reliability

for the Emotional Well-Being subscale is .90. The mean and

standard deviation for the Physical Functioning subscale as

reported in the Medical Outcomes Study are 70.61 and 27.42

respectively. Likewise, the mean and standard deviation for









51

the Emotional Well-Being subscale as reported from the same

study are 70.38 and 21.97 respectively. The sample mean and

standard deviation for the Physical Functioning subscale is

84.96 and 18.25 respectively. The sample mean and standard

deviation for the Emotional Well-Being subscale is 78.32 and

13.05 respectively. Higher scores on the RAND reflect

better or more positive perceptions of health-related

quality of life.

5. The Health-Care-Seeking Behavior Questionnaire (HCSBQ),

(Appendix F)

The Health-Care-Seeking Behavior Questionnaire is a

six-item scale that was developed by the researcher to

assess health-care-seeking behaviors among African-American

women. Using a five-point Likert-type scale where 1 = very

likely and 5 = not likely at all, participants reported how

likely they are to engage in health-care-seeking behaviors

such as visiting the doctor or other health-care provider

for a routine check-up or visiting the doctor or health-care

provider when they feel ill. A sample item is "How likely

are you to go to a primary health-care provider for a yearly

physical check-up?" The items are specifically geared

toward visits to a primary care provider for routine visits

and preventive health-care.











Internal consistency reliability yielded a Cronbach

Coefficient Alpha of 0.76. The participants' scores ranged

from a high of 4.33 to a low of 1. The lower a

participant's score, the more engaged she was in health-

care-seeking behaviors. The sample 7--.n' the HCSBQ was

1.82 and the standard deviation was 0.73.

The HCSBQ originally started out as a ten-item scale.

However, a factor analysis of the HCSBQ indicated that there

were three separate factors within this questionnaire.

Questions 1, 2, 3, 4, 5, and 8 loaded onto factor one and

seemed to specifically measure health-care seeking behaviors

as defined by this study. Thus, only these items (i.e.,

factor one items) were used as the final Health-Care-Seeking

Behavior Questionnaire (HCSBQ). The mean ratings of these

items by each participant was the HCSB score for that

participant.

Questions 6 and 7, which loaded onto factor two, seemed

to measure more specifically the availability of services,

or the degree to which a participant thought medical

services were available to her. Questions 9 and 10 loaded

onto factor three, which seemed to measure compliance with

medication. In other words, it measured whether or not

participants were likely to get prescriptions filled and











whether or not they were likely to take all of their

prescribed medication.

6. The Marlowe-Crowne Social Desirability Scale, short-

form, (M-C SDS [20]; Marlowe & Crowne, 1960)

(Appendix G)

The Marlowe-Crowne Social Desirability Scale was used

to measure the amount of variance in the data due to the

participants' desire to respond to questions in a culturally

appropriate manner. The short version of the Marlowe-Crowne

(20 items) is based on the original 33-item instrument. The

scale consists of items that are culturally supported but

are unlikely to occur. A sample item reads "I am always

willing to admit it when I make a mistake." The scale is

scored based on the number of "true" or "false" responses to

each question. Sample scores ranged from a low of 0 to a

high of 20 out of a possible 20. The sample mean was 12.52,

and the sample standard deviation was 4.27. The higher a

respondent's score, the more likely she was responding in a

socially desirable manner.

The Kuder-Richardson formula 20 (K-R 20) reliability

coefficients for the 20-item instrument (.78 for university

males and .83 for university females) are similar to the K-R

20 reliability for the original 33-item inventory (.83 for

university males and .87 for university females). Pearson











product-moment correlations between the 20-item scale and

the 33-item instrument were as high as .98, indicating

adequate construct validity for the shorter version (Fraboni

& Cooper, 1989; Strahan & Gerbasi, 1972). Participant

responses on the M-C 20 indicated whether or not the

responses to the Health Behavior Assessment Battery were

likely to be valid.

A debriefing form describing the nature and purpose of

the study was read and signed by all participants at the

conclusion of the study (see Appendix H).

Procedure

This research involved three phases: (1) training of

research assistants, (2) recruitment of participants, and

(3) collection of data.

Phase I: Training of Research Assistants

Phase one of the research consisted of training three

African-American female research assistants to assist in the

group administration of the Health Behavior Assessment

Battery (HBAB). Three adult women from the African-American

communities sampled were trained as research assistants in

this study. This training was conducted by the principle

investigator and lasted approximately one hour. During the

training session, each research assistant was shown how to

administer the HBAB using a culturally sensitive approach.









55

Specifically, each research assistant was given a set of

verbal instructions to use with the participants in each

group session. These verbal instructions were as follows:

Hello, my name is You have verbally
agreed to participate in a study that is designed to
examine health problems that are often experienced by
African-American women and how African-American women
cope with these health problems. Before we begin, I
would like for you to read and sign the consent form
which will now be passed out to you. If you need help
reading the form, I will be happy to assist you. There
are two copies of the consent form one for you to
keep and one to be returned to me. These forms will be
kept separate from the questionnaires you will complete
later.
This study will involve completing a series of
health-related questionnaires. There are seven
questionnaires in all. Each questionnaire has a
separate set of instructions at the beginning, so
please be sure to read all instructions before starting
the questionnaire. You should complete the
questionnaires in the order that they are given to you.
If you need assistance reading the questionnaires, I
will be available to assist you. Please do not put
your name on any of the questionnaires.
Please be sure to answer all items to the best of
your ability. It usually takes 45 minutes to an hour
to complete the questionnaires; however, you may take
as much time as you need. Once you have finished your
packet of questionnaires, you are to paper clip them
together as you found them and place them in the
"Completed Questionnaires" box located in the front of
the room. You are to then read (or have read to you) a
debriefing form, which you will sign, stating that you
are aware of the purpose of this study. Should you
want group results of this study, you will need to
complete your name and address on the "Request for
Results" form. You will receive $10.00 for your
participation in this study after you have turned in
your questionnaires. You will have completed this
study once you have completed all of the questionnaires
and signed all appropriate forms. Are there any
questions? If not, you may begin.











The research assistants also were shown how to (1)

address questions pertaining to the HBAB in the event that

someone could not read and/or understand a particular

question, (2) collect the completed HBAB, (3) debrief the

participants, and (4) collect the "request for results"

sheets. Finally, each research assistant was instructed as

to how to compensate each participant for her participation.

After the training was completed, each research assistant

conducted a "mock" session with the principal investigator

to confirm that she understood how to conduct the group

sessions.

Phase 2: Recruitment of Participants

Participants for this study were solicited from the

community at large as well as community churches and other

community organizations/centers that have high

concentrations of African-American women. The recruitment

phase consisted of three separate methods or strategies:

(1) Newspaper announcement; (2) flyers; and (3) personal

appeals.

Method 1:

Participants from the community at large were recruited

via a newspaper announcement about the study. In the

announcement, African-American women 21 years of age or

older were invited to call the principal investigator or a









57

research assistant in order to sign up for one of four group

sessions, using their initials only. Initials were used in

order to protect anonymity. Respondents could call at any

time to sign up for a group session. Group session times

were set according to times thought to be convenient for

most people.

Approximately 70 women responded via phone to the

newspaper announcement. Verbal instructions were given to

the respondents regarding the date, time, and location of

the study. In addition, they were told that the study

consisted of completing a packet of questionnaires about

African-American women and health that would take

approximately 45 minutes to an hour to complete. Finally,

respondents were told that they would receive $10.00

compensation for their participation in the study.

Fifty-six women of the 70 who responded via phone

actually showed to participate in the study (80%). A total

of four group sessions were conducted for participants who

responded via phone to the newspaper ad. All group sessions

were conducted at a community church.

Method 2:

Participants also were recruited from churches and

other community organizations via flyers about the study.

Flyers were posted in places heavily frequented by African-











American women, such as a community health agency, a local

school in an African-American community, an African-American

managed business, a two-year community college, and an

African-American church. Flyers announced that African-

American women 21 years of age or older were invited to call

the principal investigator or a research assistant in order

to sign up for one of four group sessions, using their

initials only. Respondents could call at any time. Again,

group session times were set according to times thought to

be convenient for most people.

Several women responded to the flyers regarding the

study both by phone and in person when the principal

investigator or a research assistant was available. Verbal

instructions regarding the date, time, and location of the

study were given to those who responded by phone. To those

responding in person, a reminder slip was given to them that

contained information regarding date, place, and time. All

respondents were told that the study consisted of completing

a packet of questionnaires about African-American women and

health. In addition, they were told that the study would

take approximately 45 minutes to an hour to complete.

Finally, respondents were told that they would receive

$10.00 compensation for their participation in this study.











While the flyer initially stated that four group

sessions would be held, a total of 6 group sessions were

held in order to accommodate those who were willing to

participate but could not attend the pre-determined group

sessions. Group sessions were conducted as follows: Two

group sessions were conducted at a community health agency,

one at a local business, one at a local school, one at a

local community college, and one at a local church. A total

of 60 African-American women participated in these 6 group

sessions.

Method 3:

In addition to a newspaper ad and flyers about the

study, participants also were recruited via personal appeals

at churches by the principal investigator. In the personal

appeal, potential participants were asked to participate in

a study on African-American women and health. They were

told that the study would involve completing a set of

questionnaires that would take approximately 45 minutes to

an hour to complete. In addition, they were told that

participants would receive $10.00 compensation for their

participation in this study.

Participants who were recruited via personal appeals

signed up for one of two group sessions. Both of these

times were selected according to normal meeting times for











the church members. Participants signed up using their

initials only to protect their anonymity. In addition, the

place for both group sessions was indicated on the sign-up

sheet so that all participants would know where as well as

when to report to participate in this research project.

However, each person who signed up to participate in the

study was given a reminder sheet indicating the date, time,

and place of the group sessions. A total of 30 African-

American women participated in these two group sessions.

Phase 3: Data Collection

A total of twelve group sessions were conducted in this

study. In all twelve group administration sessions,

participants completed the Health Behavior Assessment

Battery (HBAB). All instruments in the HBAB, with the

exception of the Demographic Questionnaire, were placed in

random order within and across group settings in order to

control for any order effects on the results. The

Demographic Questionnaire was placed last in each packet in

case any participant found the information requested too

personal, thereby influencing her approach to the study as a

whole.

Before receiving the HBAB, participants in each group

were told the purpose of the study by either the principal

investigator or a research assistant. Specifically, they











were told that the purpose of this study was to examine

health problems African-American women often experience and

how African-American women cope with these problems.

Individuals who agreed to participate in the study then were

asked to read (or have read to them) and sign an informed

consent form that was collected by the research assistant.

Only those individuals who signed an informed consent form

were allowed to participate in the actual study. All

individuals who showed up for the study chose to

participate. Verbal instructions specifying the procedures

for data collection were stated previously in the section

labeled "Phase 2." Only one participant had difficulty

reading the questionnaires and therefore required verbal

administration of the HBAB.

Participants were given $10 compensation for their time

and participation. These incentives were especially needed

to recruit African-American participants, given their well-

documented reluctance to participate in research. The

compensation was given immediately after the participants

turned in the HBAB and signed the debriefing form.

A debriefing form describing the nature and purpose of

the study was read by each individual and signed by all

participants at the conclusion of the study. Again, in the

case where the participant could not read, the form was read











to her by a research assistant. Prior to starting,

participants were told that they could request a report

containing a summary of the results from the study and that

this summary, if requested, would be mailed to them. In

order to facilitate action on this request, participants

were given the opportunity to completed a "Request for

Results" form on which they placed their name and address.

These forms were given to each participant when she placed

her completed HBAB in the "Completed Questionnaire" box.

Fourteen participants requested results of the study.

Participants took approximately 20 to 50 minutes to

complete the HBAB. Participation in this study was

voluntary; as such, participants could withdraw their

consent at any time without prejudice. No one opted to

withdraw. All data collection sessions occurred within a

four-week period.
















CHAPTER 4
RESULTS

The descriptive data, hypotheses and research questions

of interest in this study are discussed in this chapter.

The results are delineated in several parts. First,

descriptive data on all major variables are reported.

Second, the results of a preliminary Pearson correlational

analysis to assess relationships among the measures of the

dependent variable in the hypotheses and scores on the

Marlowe-Crowne Social Desirability Scale are reported.

Third, results of the correlational analyses to test the

hypotheses are reported. Finally, the statistical analyses

used to address the research questions are examined.

Descriptive Data on all Major Variables

Table 4.1 presents the descriptive statistics on all

major variables for the total sample. Normative data on the

major variables in this study are not available.

Results from the Preliminary Pearson Correlational Analysis

A preliminary Pearson correlational analysis was

performed to determine if there was an association between

scores on the Marlowe-Crowne Social Desirability Scale (M-C













Table 4.1

Descriptive Statistics on All Major Research Variables


Sample Sample Sample Sample
N Mean Std Dev Min Max


AC 124 356.19 44.77 247.00 464.00
SS 124 90.60 17.24 50.00 133.00
HRQL 124 78.92 11.60 46.25 100.00
PSYCH 124 78.32 13.05 44.00 100.00
PHYS 124 84.96 18.25 30.00 100.00
HCSB 124 1.67 0.55 1.00 4.00


Notes: AC = Acculturation
SS = Perceived Social Support
HRQL = Health-Related Quality of Life
PSYCH = Psychological Well-Being
PHYS = Physical Health
HCSB = Health-Care Seeking Behaviors


SDS) and scores on the major variables of study:

(1) acculturation (AC), (2) perceived social support (SS),

(3) overall health-related quality of life (HRQL), as well

as physical health (PHYS) and psychological well-being

(PSYCH), and (4) health-care-seeking behaviors (HCSB).

Results indicated that social desirability (SD) as measured

by the M-C SDS was significantly associated with

three of the major variables in the study the

psychological well-being (PSYCH) component of health-related

quality of life (HRQL) as measured by the RAND 36 r = .23, p

<.01), the physical health (PHYS) component of health-











related quality of life as measured by the RAND 36 r = -.20,

ep .02), and health-care seeking behaviors (HCSB) as

measured by the HCSBQ r = -.20 p .03). Consequently, the

analyses used to test hypotheses 1, 2, and 3 controlled for

the influence of social desirability. The Pearson

correlation coefficients for correlations between the

Marlowe-Crowne Social Desirability Scale scores and the

scores on the major variables of the study are summarized in

Table 4.2.


Table 4.2

Pearson Correlation Coefficients for the Marlowe-Crowne
Social Desirability Scale (SDS) Scores and Measures of
Acculturation (AC), Perceived Social Support (SS), Health-
Related Quality of Life (HRQL, PSYCH, and PHYS), and
Health-care-seeking Behavior (HCSB)


SDS AC SS HRQL PSYCH PHYS HCSB
SDS -0.079 0.080 -0.117 0.231** -0.202* -0.201*

Note: *~ < .05, **p < .01


Results from the Analysis to test Hypotheses 1,2, and 3

This study examined the following three hypotheses:

(1) Degree of acculturation (AC) is significantly

associated with health-care-seeking behaviors (HCSB) such

that the greater the degree of accu1' .L- 'Lr,, the greater

the level of engagement in health-care-seeking behaviors;











(2) Perceived social support (SS) is significantly

associated with health-care-seeking behaviors (HCSB) such

that the greater the level of perceived social support, the

greater the level of engagement in health-care-seeking

behaviors; and

(3) Overall health-related quality of life (HRQL), as well

as physical health (PHYS) and psychological well-being

(PSYCH), is significantly associated with health-care-

seeking behaviors (HCSB) such that the greater the level of

overall health-related quality of life, (as well as physical

health and psychological well-being), the greater the level

of engagement in health-care-seeking behaviors.

To test hypotheses one, two and three, a partial

correlation analysis was performed. A partial correlation

was used in order to control for the influence of social

desirability as it was found to be significantly correlated

with three of the major variables in the study. Results for

hypothesis one indicated that there was no significant

relationship between level of acculturation (AC) and level

of engagement in health-care-seeking behaviors (HCSB).

Thus, hypothesis one, which stated that higher levels of

acculturation would be significantly associated with higher

levels of engagement in health-care-seeking behaviors, was

not supported.











No significant correlation was found between level of

perceived social support and level of engagement in health-

care-seeking behaviors r = -.03, p < .73). Thus, hypothesis

two, which stated that higher levels of perceived social

support would be significantly associated with higher levels

of engagement in health-care-seeking behaviors, was not

supported.

Results of the partial correlation analysis to test

hypothesis three also revealed no significant correlation

between overall health-related quality of life and health-

care-seeking behaviors (r = -.06, 2 < .48), physical health

and health-care-seeking behaviors r = .02, 2 < .83), or

psychological well-being and health-care-seeking behaviors

r = -.04, p < .633). Thus, hypothesis three, which stated

that higher levels of overall health-related quality of

life,(as well as physical health and psychological well-

being), would be significantly associated with higher levels

of engagement in health-care-seeking behaviors, was not

supported.

Additional results from the partial correlation

analysis revealed a low but significant negative correlation

between level of acculturation and psychological well-being

r = -.24, p < .007). Lower scores on the African-American

Acculturation Scale indicate higher levels of acculturation,











whereas higher scores on the Emotional Well-Being subscale

of the RAND 36 suggest higher levels of psychological well-

being. Thus it appears that African-American women who are

more acculturated experience higher levels of psychological

well-being.

Finally, results from the partial correlation analysis

suggested that there was a low but significant positive

correlation between perceived social support and

psychological well-being r = .19, p < .04). This suggests

that African-American women who have social support systems

also tend to experience higher levels of psychological well-

being. The partial correlation coefficients from the

partial correlation analysis performed to test hypotheses

one, two, and three are presented in table 4.3

Results from the Analyses to test Research Questions 1,2,3 & 4

Research question 1 asked whether there is a

significant difference in level of acculturation (AC) in

association with age (AGE) or total household income

(INCOME). An Analysis of Variance (ANOVA) was used to

examine this research question. The dependent variable was

acculturation. The independent variables were AGE, INCOME,

and AGE x INCOME. Results of the ANOVA indicated that there

were no significant differences in acculturation in

association with age, income, or the age by income

interaction [F(l, 19) = 1.13, p = .34)].











Research Question 2 asked whether there is a

significant difference in level of perceived social support

and level of overall health-related quality of life (HRQL),

as well as physical health (PHYS) and psychological well-

being (PSYCH), in association with age (AGE), total

household income (INCOME), and age (AGE) x total household

income (INCOME). Two separate MANCOVAs were used to examine

this research question. Social desirability (SD) was

entered as a covariate to control for its influence as it

was significantly correlated with both PSYCH and PHYS.

In the first MANCOVA, the dependent variables were

physical health, psychological well-being, and perceived

social support. The independent variables were AGE, INCOME,

and AGE x INCOME. Results of the MANCOVA showed that there

were no significant main effects for age [F (12, 268) =

1.43, p = .15)] or income [F (9, 246) = 1.68, p = .09].

Similarly, there was no significant interaction effect for

age x income [F (36, 299) = 1.42, p = .07]. These findings

suggest that there are no significant differences in

perceived social support, physical health, or psychological

well-being in association with age, income, or the age by

income interaction.

In the second MANCOVA used to examine research question

two, the dependent variables were overall health-related











quality of life (HRQL) and perceived social support (SS).

The independent variables were AGE, INCOME, and AGE x

INCOME. Results revealed that there was no significant main

effect for age [F (8, 204) = .94, 2 = .49] nor a significant

interaction effect for age x income [F (24, 204) = 1.49, p =

.07]. There was, however, a significant main effect for

income [F (6, 204) = 2.21, p = .04]. Only the ANCOVA with

overall health-related quality of life was significant [F

(6, 204) = 3.00, p = .03]. In this ANCOVA, income was the

only significant independent variable (p < .03 ). Follow-up

Tukey's analysis indicated no differences in overall health-

related quality of life for African-American women based on

income.

Research question 3 asked whether there is a

significant difference in level of health-care-seeking

behaviors (HCSB) in association with age, total household

income, or the age by income interaction. An Analysis of

Covariance (ANCOVA) was used to examine this research

question. The dependent variable was health-care-seeking

behaviors (HCSB). The independent variables were AGE,

INCOME, and AGE x INCOME. Additionally, social desirability

(SD) was entered as a covariate to control for the influence

of this variable as it was significantly correlated with the











health-care-seeking behavior variable. Results of the

ANCOVA were not significant [F (1, 20) = 1.00, p = .47].

Research question 4 asked whether acculturation (AC),

perceived social support (SS), or health-related quality of

life (HRQL) overall, as well as physical health (PHYS) and

psychological well-being (PSYCH) predict health-care-

seeking behaviors (HCSB). Two multiple regression analyses

were performed to address this research question. Social

desirability scores (SD) from the Marlowe-Crowne-Social

Desirability Scale were entered as a covariate in both

multiple regressions to control for the influence of social

desirability.

The first multiple regression model with health-care

seeking behaviors (HCSB) as the criterion variable and

acculturation (AC), perceived social support (SS), and

overall health-related quality of life (HRQL) as the

predictor variables was not significant [F (4, 123) = 1.41,

p = .24, r = .05]. This finding suggests that among the

African-American women in this research, level or frequency

of health-care-seeking behaviors is not significantly

associated with levels of acculturation, perceived social

support, or overall health-related quality of life.

The second multiple regression model with health-care

seeking behaviors (HCSB) as the criterion variable and









72

acculturation (AC), perceived social support (SS),

psychological well-being (PSYCH), and physical health (PHYS)

as the predictor variables also was not significant [F (5,

123) = 1.01, p = .39, r2 = .04]. This finding also suggests

that among African-American women in this research, level or

frequency of health-care-seeking behaviors is not

significantly associated with levels of acculturation,

perceived social support, psychological well-being, or

physical health.















CHAPTER 5
DISCUSSION

While African Americans in general appear to experience

a greater number of health problems in comparison to their

White counterparts, the research on their health behaviors

and/or health-care-seeking behaviors has been limited. In

response to the limited research in this area, the current

study was designed to examine physical, social, and

psychological factors that influence African-American

women's health-care-seeking behaviors.

Specifically, this study examined the relationship

among acculturation, perceived social support, overall

health-related quality of life, as well as physical health

and psychological well-being, and health-care-seeking

behaviors. In addition, this study examined whether level

of acculturation, level of perceived social support, or

level of overall health-related quality of life, as well as

physical health and psychological well-being, predict

health-care-seeking behaviors. Finally, this study assessed

whether there was a significant difference in level of

acculturation, level of perceived social support, or level

of overall health-related quality of life, as well as









74

physical health and psychological well-being, in association

with age and total household income.

This chapter presents a summary and interpretation of

the results, the limitations of the study, the implications

for practice and future research, and a conclusion.

Summary and Interpretation of the Results

Results of the preliminary Pearson correlational

analysis revealed that social desirability scores (SD) were

significantly correlated with three of the main variables in

the study: health-care seeking behaviors (HCSB), physical

health (PHYS), and psychological well-being (PSYCH). These

results indicate that a response bias may have threatened

the validity of the measure of the health-care-seeking

behaviors variable as well as the two specific health-

related quality of life variables physical health and

psychological well-being.

A significant negative correlation was found between

social desirability (SD) and health-care-seeking behaviors

(HCSB). The lower the HCSB score, the more likely

participants felt they were to engage in health-care seeking

behaviors. Thus, if a response bias or social desirability

bias were indeed in effect, it would have lowered the

health-care seeking behavior scores of some respondents,

systematically biasing the correlations investigated. As











such, the African-American women in this study may have

reported themselves as engaging in more health-care-seeking

behaviors than they actually do.

Similarly, a significant negative correlation was found

between social desirability (SD) and physical health (PHYS).

The higher the physical health score, the more likely

respondents feel that they were not inhibited in any of

their physical functioning as a result of health. Since

higher PHYS scores indicate fewer health-related problems,

the response bias here seems to be that of African-American

women reporting themselves as having more health-related

problems than perhaps they actually do. This finding is

surprising in that if a true social desirability bias were

in effect, it would have raised the physical health scores

of some respondents, systematically biasing the correlations

investigated.

A significant positive correlation also was found

between social desirability (SD) and psychological well-

being (PSYCH). This significant correlation also suggests a

response bias in responding to the psychological well-being

component of the health-related quality of life measure;

thus participants may have presented themselves as being

more psychologically healthy than they really are.









76

The finding that social desirability was significantly

correlated with three major variables health-care seeking

behaviors, physical health and psychological well being is

an important finding in this research, given that previous

research investigating health behaviors and/or health-

related quality of life and its various components failed to

control for the effects of social desirability.

Specifically, previous research investigating health-related

quality of life among various populations failed to control

for response bias among subjects (Weinberger, Nagle, Hanlon,

Samsa, Schmader, Landsman, Cowper, Cohen, & Feussner, 1994;

Yamaoka, et al., 1998; Barofsky et at., 1997).

As a means for controlling for response bias in the

current study, partial correlations that controlled for

social desirability were used to examine hypotheses 1,2 and

3. In addition, the social desirability variable was used

as a covariate in each of the univariate and multivariate

analyses used to investigate research questions 1,2 and 3,

as well as in the multiple regression to investigate

research question 4. Even though the influence of social

desirability was controlled for statistically, this is an

artificial control that does not allow us to know the degree

to which the results are reliable.











Findings in this research provide no support for

hypothesis 1. Hypothesis one stated that a greater degree

of acculturation would be significantly associated with a

greater level of engagement in health-care-seeking

behaviors. A partial correlation analysis performed to

examine hypothesis 1 revealed no significant association

between reported level of acculturation and reported level

of health-care-seeking behaviors; thus this hypothesis was

not supported.

The finding that acculturation and health-care-seeking

behaviors were not significantly correlated in this study

suggests that African-American women who are more

acculturated do not necessarily engage in higher levels of

health-care-seeking behaviors. This finding differs

somewhat from earlier research findings that have reported

that African Americans who were more acculturated did indeed

show a greater level of health status and/or engagement in

health-related behaviors. Specifically, Landrine and

Klonoff (1996) found that African Americans (both men and

women) who were more acculturated did not tend to suffer

from hypertension, whereas those who were less acculturated

(more traditional) did. Similarly, a related study

conducted by Landrine and Klonoff (1996) showed that African

Americans nonsmokers tended to be more acculturated than











their smoking counterparts, who tended to be more

traditional.

The results of this study also differed from the

results obtained by Maxwell, Bastani, and Warda (1998) in

their study of factors affecting mammogram utilization a

health-care-seeking behavior as defined by this study.

Maxwell, Bastani, and Warda found a significant positive

relationship between mammography utilization and level of

acculturation. Specifically, the greater the level of

acculturation, the more likely the women in this study were

to have had a mammogram. One difference between studies

however, is the fact that Maxwell, Bastani, and Warda used

Korean-American women whereas this study examined African-

American women. Cultural factors may indeed be a confound

when comparing results across cultures.

The current study may have failed to support earlier

findings from similar research that included African

Americans because the previous studies focused on measures

of health status or general health behaviors and their

relationship to acculturation, whereas the current study

focused on specific health-care-seeking behaviors. The

latter was defined as behaviors associated with seeking

preventative and/or follow-up care from a health-care

provider. There may or may not be a significant











relationship between health behaviors or health status and

health-care-seeking behaviors as defined by this study; thus

it is difficult to compare the results of research involving

these different but seemingly related constructs.

Hypothesis 2 in this study stated that a greater level

of perceived social support would be significantly

associated with a greater level of engagement in health-

care-seeking behaviors. As with hypothesis 1, results of

the partial correlational analysis to test hypothesis 2

revealed no significant association between perceived social

support and health-care-seeking behaviors. Thus, this

hypothesis was not supported.

The finding that perceived social support and health-

care-seeking behaviors were not significantly correlated in

this study suggests that individuals who perceive themselves

as receiving an adequate level of support from various

sources in their lives do not necessarily engage in more

health-care-seeking behaviors. This finding is surprising

in light of the fact that previous research has found fairly

significant relationships between health status/health

behaviors and social support. More specifically, earlier

research findings have shown that individuals who

experienced higher levels of social support engaged in more

positively oriented health behaviors. For instance, it has











been found that individuals who have a significant social

network tend to exercise more frequently (Gottlieb & Green,

1984; Steptoe et al., 1997). In addition, Sorensen,

Stoddard and Macario (1998) found that those individuals who

had adequate support systems were more likely to make

positive dietary changes in their lives as needed.

One reason for the lack of the predicted correlation

between perceived social support and health-care-seeking

behaviors is that this study used a multi-dimensional

measure of social support. As such, various behaviors from

various sources were presented for endorsement. In

addition, satisfaction with the amount of support received

from these sources was factored into the overall perceived

social support score. Thus, individuals may have perceived

themselves as receiving an adequate amount of social

support; however, their level of satisfaction with this

support may not have been strong enough to exert a positive

influence on their level of engagement in health-care-

seeking behaviors. Future research should focus on whether

health-care-seeking behaviors in African-American women are

significantly associated with both their levels of

satisfaction with their social support and their overall

levels of perceived social support.











Hypothesis 3 in this study stated that overall health-

related quality of life, as well as physical health and

psychological well-being two components of health-related

quality of life would be significantly associated with

health-care-seeking behaviors. Results of the partial

correlation analysis to test hypothesis 3 revealed no

significant association between overall health-related

quality of life or its components (i.e., physical health and

psychological well-being) and health-care-seeking behaviors.

Thus, hypothesis 3 was not supported.

The finding that neither overall health-related quality

of life nor its components and health-care-seeking behaviors

were significantly correlated in this study suggests that

individuals who experience themselves as having better

health-related quality of life do not necessarily engage in

more health-care-seeking behaviors. As most of the research

reviewed on health-related quality of life focused on its

association with health status as opposed to health behavior

or health-care-seeking behaviors, it is difficult to compare

the results of this study with previous research.

This study may have failed to find a significant

association between health-care-seeking behaviors and

health-related quality of life because it failed to assess











dimensions of health-related quality of life that are

relevant for the African-American women in this research.

According to Berger (1989), quality of life [as well as

health-related quality of life] must be assessed specific to

a particular population. While some basic quality of life

measurements might be applicable to everyone in all

situations, many more are relevant only to a particular

group of people. Thus if the instrument used in this study

failed to assess the measurements that were salient for the

African-American women participants in this study, it would

not adequately have captured a true measure of health-

related quality of life among these women. Future research

should focus on developing a measure of health-related

quality of life that is geared specifically for African-

American women such as the research participants in the

present study.

Research question 1 investigated differences in level

of acculturation in association with age and total household

income. Results of the ANCOVA used to address this research

question indicated that neither age nor total household

income were associated with differences in level of

acculturation among the African-American women in this

research. These results are in keeping with the results

obtained by Landrine and Klonoff (1995) in their cross-











validation of the African-American Acculturation Scale

(AAAS). When Landrine and Klonoff compared the scores of a

young college student sample with those of an older

community sample, the scores were statistically similar for

the two populations. Additionally, in their validation of

the AAAS, Landrine and Klonoff (1994) found that scores on

the eight subscales were not related to income, city of

origin or education. As the authors stated, the purpose of

the AAAS is to measure acculturation the extent to which

individuals are immersed in their culture of origin. As

such, acculturation should not be related to income or other

status variables. Because no significant relationships were

found in this study between acculturation and status

variables, it can be surmised that an accurate measure of

acculturation as opposed to socioeconomic status was

obtained.

The second research question investigated differences

in level of perceived social support and level of health-

related quality of life (as well as physical health and

psychological well-being) in association with age and total

household income. Results indicated that there were no

significant differences in level of physical health

according to age or income. These findings are inconsistent

with findings from previous studies that revealed that women











who had lower incomes either experienced poorer health

status or engaged in preventive health measures less

frequently than those with higher incomes (Mickey, Durski,

Worden, & Danigelis, 1995; Mutchler & Burr, 1991; Keith &

Jones, 1990). One reason cited for such income-related

differences is the fact that individuals with lower incomes

do not have the resources available to them (i.e., money or

health insurance) to obtain the medical care that they need.

As such, many individuals without needed resources delay

medical attention until their pain or symptoms are much

advanced.

The finding that there was no significant relationship

between income and physical health in this study could be

attributed to the fact that there was a significant negative

relationship between the physical health and social

desirability scores. If a response bias were indeed in

effect, it seems plausible that the physical health scores

would have been systematically lowered across the range of

income levels, thus making it difficult to detect true

differences in perceived physical health between the income

categories.

One reason for the lack of predicted association

between psychological well-being and income could also be

due to the fact that there was a significant positive











relationship between psychological well-being and social

desirability. Again, if a response bias were in effect,

participants would have portrayed themselves as being more

psychologically healthy than they really are across all

income categories. Thus it would be difficult to tell if

there are indeed differences in level of psychological well-

being in association with income.

The finding that there were no significant differences

in physical health or psychological well-being in

association with age is surprising given that physical and

emotional health is often seen as declining with age. A

possible reason for the lack of association in this study is

that the majority of the participants (n = 90) were age 50

and below. Future research should focus on obtaining a more

representative sample across all age categories.

Research question 3 asked whether there was a

significant difference in level of health-care-seeking

behavior in association with age and total household income.

Results indicated that age nor total household income was

associated with health-care-seeking behaviors. This lack of

association was not expected due to the fact that the

literature, as mentioned before, reports income as being one

of the determinants of both health status and health-care









86

services utilization (Mickey, et al., 1995; Mutchler & Burr,

1991; Keith & Jones, 1990).

One possible explanation for not finding significant

differences in level of health-care-seeking behaviors in

association with age and income is that health-care-seeking

behaviors was found to be significantly correlated with

social desirability. As such, this would systematically

bias the health-care-seeking behavior scores of African-

American women in this study. In other words, participants

may have presented themselves as engaging in more health-

care-seeking behaviors than they actually do, thus making it

difficult to distinguish differences across the age and

income spectrum.

Research question 4 asked whether level of

acculturation, level of perceived social support or level of

health-related quality of life predict health-care-seeking

behavior. Results indicated that neither acculturation,

perceived social support nor health-related quality of life

predicted health-care-seeking behaviors. This finding

suggests that health-care-seeking behaviors, for those who

participated in this study, are likely due to factors not

investigated in this study. Further research is needed to

ascertain factors that might be predictive of African-

American women's health-care-seeking behaviors in general.











Additional results from the study indicated that the

African-American women who participated in this study

experienced higher levels of psychological well-being in

association with higher levels of perceived social support.

In addition, the more acculturated these participants were,

the greater their levels of psychological well-being. These

findings suggest that for the participants in this study,

having a functional social support system in their lives is

important to their overall level of psychological well-

being. Further research is needed to determine the role

acculturation plays in the psychological well-being of

African-American women.

Limitations of the Study

There are several limitations of this study that

warrant discussion. First, only self-report measures were

used in the actual data collection. While self-reports are

a primary source of data in both psychological and social

sciences (Schwarz, 1999), participants may have under- or

over-reported the physical, psychological or social

variables examined in this study due to their tendency to

respond to questions in a socially desirable manner. This

tendency may have been true in this study as results

indicated that there were indeed significant correlations

between social desirability and three major variables in











this study (health-care-seeking behaviors, physical health,

and psychological well-being). In an effort to address this

potential limitation, social desirability was controlled for

in the present study.

In addition, self-reports can be seen as a fallible

source of data as even minor changes in how the questions

are worded or formatted can lead to major changes in the

obtained results (Schwarz, 1999). As referenced in Schwarz,

when a group of participants were asked how successful they

had been in life, results varied significantly based on the

rating scale used. Specifically, when a scale of -5 to 5

was used, 34% of the respondents reported high success.

However, when the rating scale was changed to 0 to 10, only

13% reported high success. According to Schwarz, factors

involved in self-report research that can significantly

influence results are those such as making sense of the

questions asked and types of response alternatives. Even

so, given the paucity of research in the area of African-

American women's health, self-report questionnaire research

is indeed a first step.

Another major limitation of this study is the small

sample size (N = 124). With a larger sample, perhaps

statistical results that approached significance would have

actually been significant. This brings up the larger issue











of how researchers might successfully recruit African

Americans as participants in various research efforts.

Extensive efforts were expended to recruit participants for

this study; however, these efforts resulted in a limited

number of participants. As part of the recruitment efforts

in the present study, a monetary compensation ($10) was used

as a participation incentive. While $10 may have been a

strong incentive for some individuals, others may have

needed a clearer understanding of and acknowledgment of the

importance of health-related research to be inspired to

participate in this research.

Another possible limitation in this study is the

appropriateness of the inventories for African Americans.

While the African-American Acculturation Scale was designed

specifically for African Americans, the other inventories

were universal in nature. Often there is an assumed

universality to research inventories at the expense of their

cultural relevance. As stated by Ibrahim and Arrendondo, it

is imperative that cultural issues be taken into account

when conducting research (1986). This attention to culture

would include the use of appropriate instrumentation.

Finally, limitations in this study might have occurred

as the result of the use of the Health-Care-Seeking Behavior

Questionnaire (HCSBQ) constructed by the researcher. As











there was no test-retest data or validity data on the

instrument, it might not have captured the true meaning of

health-care-seeking behaviors that the researcher intended.

Future research with this measure would need to assess the

reliability as well as the validity of this measurement.

Implications for Practice and Future Research

Limitations did indeed exist in this study; yet, these

findings have implications for counseling psychologists.

Because many of the findings were inconsistent with related

previous research involving both African Americans in

general and African-American women in particular, a

necessary first step would be to see if the findings in this

research were a result of issues expounded on in the

limitations section. If this study could be replicated with

a larger sample size and with a more culturally sensitive

design (e.g., use of culturally specific instruments or

measures), findings might indeed support previous literature

showing that acculturation, perceived social support and

health-related quality of life are associated with some form

of positive health behavior.

According to Casas and Thompson (1991), research that

takes into account the values and diversity of worldviews

held by other cultures would challenge the researcher to

work within the community itself in order to solve real-











world problems. In addition, it would promote an active

role of leaders within minority communities in order to

identify those research projects that are deemed important

by the community. As such, future research examining the

relationship among acculturation, perceived social support,

health-related quality of life and health-care-seeking

behaviors might focus on the use of trained community

leaders within the African-American community to interview

African-American women about their health-care-seeking

behaviors.

Conclusion

The current study examined acculturation, perceived

social support and overall health-related quality of life

(as well as physical health and psychological well-being) as

factors in health-care-seeking behaviors among African-

American women. Results of this research clearly show that

there is indeed a need for more in-depth research regarding

African-American women and their health-care-seeking

behaviors. This need is reflected in the fact that the

factors in the health-care-seeking behaviors of these

African-American women remain unclear. The examined factors

together accounted for only 5% of the variance in health-

care-seeking behaviors as measured in the current study.









92

Given that African-American women have higher health

risks than White women, future research to further

investigate the factors in health-care-seeking behavior of

African-American women is clearly needed. This future

research will benefit from efforts to promote honest rather

than socially appropriate responses to assessments used in

such research. In addition, this research will be enhanced

by the use of culturally sensitive assessments of the

investigated predictors of health-care-seeking behaviors.

















APPENDIX A
INFORMED CONSENT FORM

You are being asked to volunteer as a participant in a
research study. This form is designed to inform you about
the nature of the study. The purpose of this study is to
help psychologists learn more about how African-American
women cope with health problems.

Participants in this study will be asked to complete a set
of questionnaires. The questionnaires can be completed in
approximately one hour, but you will be allowed as much time
as you need. You will be asked not to put your names on the
questionnaires so that your right to confidentiality is
protected. Instead, all questionnaires in each packet will
have the same code on it for matching purposes. Completed
questionnaires will be stored in a locked filing cabinet in
the psychology building at the University of Florida.

As immediate compensation for your participation in this
study, you will receive $10.00 cash.

Individual results from this study will not be available;
however, group results will be made available via mail upon
your written request. A "Request for Results" card will be
provided for your signature. There are no risks or
discomforts anticipated for participants in this study. You
may benefit from your participation in this study by
learning more about your own health and your health
practices. If you wish to discuss any discomforts you may
experience, you may call Ms. Alaycia D. Reid, Principal
Investigator, at (614) 292-5766, or Dr. Carolyn M. Tucker at
(352) 392-0601, ext. 260. Please read the statement below
and sign the form.

I have been fully informed of the procedure for the above-
described study and understand its possible benefits and
risks. I also understand that I will receive $10.00 as
compensation for my participation in this study. I
understand that I am free to discontinue my participation in




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EMH9QIK4D_NJJ79P INGEST_TIME 2013-03-25T15:05:28Z PACKAGE AA00013607_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES