":31

7 . Q

9
Ilnh»;
‘r-f.

‘ r

. 5.5?
o 'g f
#1:; ’

S.-

. .V
‘ A .. V
4"" , J A

 

 

.
u

1‘
IA?

' ~......‘
I.
u

.THEms

iGAN STAT

IHIIHIIIHHIZII ll'

 

 

 

 

‘HIUIIIJHIIJJIHHIIHIUHIIIUUII

301389 8873

This is to certify that the

dissertation entitled

Low Birthweight, Infant Mortality, Acculturation,
and Nutrition: An Explanation of Between Group
Differences Among Latinos

presented by

David Anthony Lopez

has been accepted towards fulfillment
of the requirements for

.211 ._D_.____ degree in __§_Q_¢ig_1_qg_y

%«/%

Major professor

Date WAS: {77‘

MSU is an Afflrmativr Action/Equal Opportunity Institution 0- 12771

 

LIBRARY
Michigan State
University

 

 

 

PLACE ll RETURN BOX to romovothb ohookoutfnom your rooord.
TO AVOID FINES Mum on or Moro duo duo.

DATE DUE DATE DUE DATE DUE

be: 939° L
[‘42: 939.. 117.9711 I

5., 4‘

 

 

 

 

 

 

   

 

 

 

r—U
Pam __IE

MSU lo An Aflinnotivo Action/Equal Opportunity lnotiMion
W

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ulla-9.1

 

 

 

LOW BIRTHWEIGHT, INFANT MORTALITY, ACCULTURATION, AND
NUTRITION: AN EXPLANATION OF BETWEEN GROUP DIFFERENCES
AMONG LATINOS

BY

David Anthony Lopez

A DISSERTATION
Submitted to
Michigan State University
in partial fuifliiment of the mquiremonts
for the degree of

DOCTOR OF PHILOSOPHY

Department of Sociology

1996

ABSTRACT

Low Blrthweight, Infant Mortality, Accuituration and Nutrition:
An Explanation of Between Group Differences Among Latinos

BY

David Anthony Lopez

This study provides an analysis of the correlates of low birthweight
among Latinos, Non-Latino Whites, and African Americans with its primary
focus on Latinos. JOIN birthweight is defined as weighing less than 2,500
grams at birth] The study tests two hypotheses. Hypothesis one proposes that
the nutrient intake of Puerto Rican women is less than the nutrient intake of
Mexican and Cuban women. Hypothesis two proposes that nutritional intake
effects low birthweight outcomes more for Puerto Rican women than Mexican
and Cuban women. The Hispanic Health and Nutrition Examination Survey,
1982-84, serves as the data source. The data contains information on
nutritional intake for the Latino subgroups, low birthweight outcomes among the
Latino subgroups, and social and economic information. Several multivariate
statistics are used to test the hypothesis. The hypothesis is not supported.
Puerto Ricans are not undernourished as compared to Mexicans and Cubans.
However, differential patterns in nutritional intake are found to exist among
the Latino subgroups. Differences are also found in low birthweight outcomes.
Puerto Ricans have the highest rate of preterm low birthweight followed by

Cubans. Mexicans have the lowest rate of preterm low birthweight. I argue

that Puerto Ricans have a high rate preterm low birthweight because of poor

David Anthony Lopez

social and economic conditions and a history of patriarchal dependency on the
United States. I suggest that the Cubans who are having the preterm low
birthweight babies are those who are recent immigrants to the United States. I
propose that the positive outcomes for Mexicans is due to the Mexican
experience which is a function of their particular historical circumstances. i
suggest that the Mexican experience promotes an environment that results in

positive low birthweight outcomes.

Para mi familia.

Past and present.

iv

ACKNOWLEDGEMENTS

i would like to acknowledge and thank my advisor, Dr. Refugio l. Rochin,
and The Julian Samora Research Institute for the assistance that was given me
in completing this dissertation. The help I received was indispensable.

i wish to thank my committee members, Drs. Clifford Broman, Thomas
Conner, and Francisco Villarruel for their help and guidance in this dissertation.

I also wish to thank Christopher Vanderpool and the Department of
Sociology for their support in this dissertation and throughout my graduate
career at Michigan State.

Lastly, I wish to acknowledge Ms. Marie Hansen, Acting Director of
Student Life, and the Judicial Affairs Office for their support over the last three
years. My position in Judicial Affairs was instrumental in allowing me to focus

on and complete this dissertation.

TABLE OF CONTENTS

Chapter
1. introduction
Selection of Topic
Strategy for Literature Review
Organization of Literature Review
2. Review of Literature

Relationship Between Low Birthweight and

\

Infant Mortality Rate

Risk Factors for Low Birthweight and Infant
Mortality . . . .

Preterm Low Birthweight versus Fullterm Low
Birthweight

Low Birthweight and African Americans .
Environment and Poverty .

Marital Status

Low Birthweight and Latinos

Infant Mortality Rate and Latinos.

An Overview of Latino Heterogeneity and
Low Birthweight .

Literature on Latinos and Low Birthweight

vi

Page

12

14
18
19
19
21

21

21

22

Literature on Acculturation Effects and Low
Birthweight

implications .

3. Acculturatlon and Acculturatlon Processes.

Life-Worlds
Acculturation Processes
The Role of Family .

Gender Roles

4. Food Habits, Culture, and Low Birthweight.

Food Habits .
Nutritional Anthropology
Culture and Food Habits

Nutrition and Low Birthweight

5. The Theory and the Operational Model.

Conceptual Framework: Functional Relationships and
Theory . .

Expectation of Effect of Variables on
Low Birthweight .

Operational Model .

independent Variables

Dependent Variables

Functional Relationships of Variables
Hypotheses and Subhypotheses .

Operational Definitions

vii

25
28
31
32
37
38
41

46

47
48
51

55

55

57
57
58
58
58

59

7.

Data

The Hispanic Health and Nutrition Examination
Survey, 1982-84

Limitations
Strengths
Data Analysis
The Sample .
Statistical Procedures
Logistic Regression Models
Relation of the Model to the Theory
Collinearity
Results

Low Birthweight, Gestation Length, and Racial and Ethnic
Identity . . . . . .

4:2, Social and Economic Characteristics of Low Birthweight
g ) Households .

Test of Hypotheses
interpretation of Results and Conclusions .
The Objectives

Gestation Length, Latinos, and
Low Birthweight.

‘, J, ‘\, Low Birthweight and Social and Economic
4’7 Characteristics

Hypothesis One

Hypothesis Two

viii

60

6O
61
62
62
63
65
68
69
71

73

73

74
77

85

87

88

89

90

The Effect of Education and Language on Low

Birthweight . . . . . . 91
incidence of Low Birthweight and

the Theory . . . . . . . 91
The Mexican Life-World and How do you Test

This? . . . . . . . . 94
A Reasonable Question . . . . . 96

Societal Transition and Latino

Food Habits . . . . . . . 97
Relevance of Study . . . . . 102
The Importance of Culture in Low
Birthweight Research . . . . . 104
Further Research . . . . . . 105
Conclusion . . . . . . . 106
Appendices.
Appendix A: Key Concepts . . . . . 107

Appendix 8: Significant Risk Factors for Low Birthweight
Preterm and Fullterm Infants
(Michielutte et al. 1993) . . . . . . 109

Appendix C: Summary of Selected Low Birthweight (LBW)
Studies . . . . . 110

Appendix D: Summary of Studies on Latinos and Low
Birthweight (LBW), Pregnancy, and Childbirth . . . 112

Appendix E: Historical and Geographical Circumstances
of Mexicans, Cubans, and Puerto Ricans . . . 116

Appendix F: Marin and Marin' s (1991) Discussion of Cultural
Similarities Among Latinos . . . . 123

ix

Appendix G: Social and Economic Characteristics of Latino
Subgroups and the Acculturation Process . . . 127

List of References . . . . . . . 139

LIST OF TABLES

Table

1. Low Birthweight Differences Among Latino
Subgroups . .

2. Major Food Groups and Food Practices by
Latino Subgroups . .

3. Expectations of Relationships between Independent
and Dependent Variables . .

4. Correlation Matrix of independent Variables

5. Crosstabs of Birthweight and Gestation Length
by Ethnic Group . .

6. Social and Economic Characteristics of Households
with Low Birthweight Children

7. Social and Economic Characteristics of Households
with Low Birthweight Children Born Preterm

8. Social and Economic Characteristics of Households
with Low Birthweight Children Born Fullterm

>15 9. Crosstabs of Ethnic Group by Birthweight Category
10. Means of Birthweight Categories by Ethnic Group

11. One-way Analysis of Variance of Latinas by Nutrient Intake
for Birthweight Category

12. Logistic Regression Coefficients for independent Variables
on Birthweight Category for Latinas . .

xi

Page

22

50

70

71

74

75

76

76
77

78

79

82

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Population Estimates of New Spain Colonies Prior to
1776 .

Latinos by Region (Selected Cities)

Social and Economic Characteristics of Cuban and Mexican
Women Who Immigrated. to the U. S. Between 1960 and
1970 . . . . .
Educational Attainment and Those in Professional
Occupations: Cuban, Mexican, and Puerto Rican
Women .

Economic Characteristics of Cubans and Mexicans
Percent of Cubans, Mexicans, and Puerto Ricans Who
are College Graduates (persons aged 25 years or
older) . . . . . .

Economic Characteristics of Cubans, Mexicans, and
Puerto Ricans ‘

Total Family Annual Income for Cubans, Mexicans, and
Puerto Ricans

Occupational Distribution for Cubans, Mexicans, and

Puerto Ricans (96 per 1,000 in population for persons aged

16 years and older)

Socioeconomic Status Measures for Latino Children
by Generation

xii

116

121

127

128

128

129

130

131

132

134

LIST OF FIGURES

Figure Page

1. Path Chart of Acculturation Processes . . . . 138

xiii

CHAPTER 1

iNTRODUCTION

infant mortality is a common indicator of the health status of a society
(Cockerham 1986). Low birthweight is the best predictor of infant mortality
(Hogue, Buehler, Strauss, and Smith 1987). Low birthweight is defined as a
birthweight of less than 2,500 grams, approximately 5.5 pounds (Brooks-Gum,
McCormick, and Heagarty 1989; Hogue, et al. 1987; Kramer 1987; Michielutte,
Ernest, Moore, Mies, Wells, and Buescher 1992, Rumbaut and Weeks 1993;
Taffel 1990). Cubans and Mexicans have better low birthweight outcomes than
Puerto Ricans and the low birthweight outcomes of Cubans and Mexicans are
comparable to the outcomes of Non-Latino Whites (Becerra, Hogue, Atrash,
and Perez 1991; Mendoza, Venture, Valdez, Castillo, Saidivar, Baisden, and
Martorell 1991; Health US. 1990 1992; CDC 1993). The guiding research
question in this study is: to what extent does culture effect differential low
birthweight outcomes? Exploring this question will help researchers learn more
about low birthweight and this can assist in reducing infant mortality.

The population known as "Latinos" (see Appendix A) is the primary
interest of this study. The focus of the study is on nutrition as embedded in
culture. It argues that food habits best reflect culture. The study suggests that
differential food habits (as a measure of culture) result in differential low

birthweight outcomes. Previous studies on low birthweight and Latinos have

2

not conceptualized measuring culture in this manner. The Hispanic Health and
Nutrition Examination Survey, 1982-84, serves as the study's data source.

If it is shown that there are aspects about Mexican and Cuban culture
that contribute to favorable low birthweight outcomes, then social programs and
policies can be implemented that would serve to Ieam more about these
cultures. The knowledge gained about these cultures would in turn help to
reduce low birthweight outcomes for the larger society.

The benefits of this study are: (1) it acknowledges Latino heterogeneity,
(2) positive attributes of Latino culture are recognized, (3) it increases
knowledge regarding low birthweight and the infant mortality rate, and (4)
attributes of Latino culture are viewed as a model for Non-Latino culture.

The research is guided by three main objectives: (1) to demonstrate
between group differences among Latinos regarding low birthweight and the
degree to which these differences exist, (2) to develop an improved measure
of cross-cultural comparison and to systematically combine factors that have
not been previously related to low birthweight, and (3) to test hypotheses
derived from research on biomedical, demographic, behavioral and cultural
correlates of low birthweight and to narrow these hypotheses to a reduced set
of variables that would best explain differential low birthweight among Latinos.
Due to the cross-disciplinary nature of the study, key concepts are defined in

Appendix A.

§giggtion of Topic
I discovered between group differences in low birthweight while taking a

graduate seminar at Michigan State in the fall of 1993. This seminar sparked
my interest. The seminar was titled "Topics in Social Inequality" and was
taught by Dr. Ruben Rumbaut. Its focus was on global immigration issues and
the continuing changing face of the world's population. As part of this
discussion, issues associated with immigration and acculturation were
addressed. One of these issues was low birthweight. l was not so interested
in the medical aspects of low birthweight, but more in the cultural aspect. i had
to ask, ”what is going on here?" I then began to read more about it and
concluded that this was a tepic worthy of a dissertation.

r it r u vie

i needed to learn more about the topic. To do this, I had to consult the
experts. In my seminar, I had learned that there was a professor on campus
who studies low birthweight issues. This professor is Dr. Nigel Paneth,
Director, Program in Epidemiology at Michigan State University. He was the
first stop in my fennel literature review.

There were three main goals of the literature review. The first goal was
to become familiar with and review the biomedical and risk factor literature on
the phenomenon of low birthweight. The second goal was to become familiar
with and review the literature on low birthweight and Latinos. The third goal

was to examine the role of acculturation in relation to low birthweight with a

4

special emphasis on nutrition as embedded in culture. As the literature review
progressed, it became clear that the literature pertaining to acculturation, low
birthweight, and nutrition was so important to the primary focus of the study,
that separate chapters were developed. Chapter 3 discusses acculturation and
Chapter 4 discusses food habits (i.e., nutrition), culture, and low birthweight.
Regarding the first goal of the literature review, Dr. Paneth was consulted.

Dr. Paneth suggested three important issues in considering low
birthweight. One, the phenomenon is enormously complex and a cause
remains largely unexplained. Several risk factors have been identified in
researching potential causes. Secondly, he indicated that infant mortality is
directly related to low birthweight end is the primary reason that low birthweight
is of importance. Lastly, he suggested that a distinction should be made
between preterm low birthweight and term low birthweight when conducting
research. Term low birthweight is when an infant is born at less than 2,500
grams after 37 weeks gestation. Preterm low birthweight is when an infant is
born at less than 2,500 grams prior to 37 weeks gestation (Michieiutte et el.
1992). A distinction needs to be made due to the confounding effect of preterm
delivery on low birthweight. Dr. Paneth referred me to important studies in the
biomedical field relating to the above issues which are reflected in this
dissertation (Hogue et al. 1987; Kramer1987).

Regarding low birthweight and Latinos, l was familiar with some of the

literature as a result of the previously mentioned seminar. However, this

5

literature pertained mostly to immigration issues and low birthweight. For
example, the research of Rumbaut (1992), Rumbaut and Weeks (1989; 1993)
and Rumbaut, Chavez, Moser, Pickwell, and WIshik (1988), discussed the
assimilation of immigrants and their lack of access to health care and
subsequent health outcomes. I conducted a more comprehensive library
search using the words "Hispanic" and "weight" as my "key" phrases. i also
obtained research data from governmental and health agency sources by
ordering works directly (e.g., Center for Disease Control, COSSMHO, and the
US Census Bureau). Additionally, discussions with various scholars at
Michigan State directed me to pertinent articles and research. For example, Dr.
Refugio Rochin, my adviser, familierized me with the works of Drs. Eunice
Romero Gwynn and Douglas Gwynn, and Dr. Francisco Villarruel, a member of
my committee, acquainted me with the works of Dr. Cynthia Garcia Cell.
Finally, in the course of reviewing the literature, I became interested in studies
en Mexico and Mexicans regarding low birthweight and these are included in
the review. These sources yielded a substantial accumulation of literature on
low birthweight and Latinos.
Qrggnjgatieg of Literature Reyiew

i first discuss the biomedical aspect associated with low birthweight.
This discussion focuses on the relationship between infant mortality and low
birthweight. i then address risk factors known to be associated with low

birthweight which leads to a discussion on the role of gestation length and low

birthweight outcomes.

i then turn to examining the literature on low birthweight as related to
Latinos. However, prior to this section, I discuss low birthweight among African
Americans. I do this as a basis of comparison. Also, since Latinos are defined
by ethnicin and can be of any racial group, a discussion of low birthweight and
African Americans seems in order. African Americans have the highest rate of
low birthweight of any group in the US. (US. Bureau of the Census 1992;
Health United States 1990 1992; From the MMWR 1993). This factor may
confound differential outcome rates of low birthweight for Latinos and this is
why rates for African Americans are considered in the literature review.

After the discussion of African Americans, Latinos and low birthweight is
examined. The literature review on Latinos and low birthweight demonstrates
that differential low birthweights do exist among Latinos.

i then discuss acculturation. Since acculturation is such an important
factor in relation to health outcomes, Chapter 3 is devoted to its discussion. I
view acculturation as a process that incorporates several variables. My
definition of acculturation is taken from Garcia Cell and Meyer (1993211) who
write that "acculturation may be conceptualized as a process of leeming about
a new culture, and deciding what is to be saved or sacrificed from the old
culture.” The variables that I focus on in my examination of the acculturation
process are gender roles, family, and social and economic conditions.

Following the discussion on acculturation, the relationship between food habits,

7

culture, and low birthweight is examined in Chapter 4.

Next, in Chapter 5, l propose a theory of differential low birthweight
outcomes among Latinos and develop an operational model. This theory
suggests a series of functional relationships between the variables of nutritional
intake, food habits, culture, socioeconomic conditions, and history (as related to
geography). In the development of the operational model, the methodology of
the study is addressed. In this development, formal hypotheses are suggested,
variables are operationalized, and the data set is discussed as is the data
analysis. Results are presented in Chapter 6. Lastly, the results are

interpreted and conclusions are drawn in Chapter 7.

CHAPTER 2

REVIEW OF LITERATURE

The study of low birthweight is epidemiological because its study draws
from different academic fields. Epidemiologists explore "human ecology as it
relates to the health of human beings and their environment" and epidemiology
"draws upon the knowledge and techniques of several scientific fields"
(Cockerham 1986214). Since the study of low birthweight is epidemiological, all
possible effects contributing to low birthweight need to be considered. For this
reason, biomedical and risk factors are addressed. Low birthweight is related
to intrauterine maldevelepment and retarded growth (Yerushelmy 1967). An
infant that is born with low birthweight has not developed to its fullest potential
in the womb and thus its growth is "retarded." Low birthweight is the most
important predictor for infant survival (Hogue et al. 1987). Infants born with low
birthweight (2,500 grams at birth) are "five to 10 times more likely to die within
the first year of life" (CDC 1994). Additionally, "infant mortality is considered to
be one of the most sensitive indicators of the health status of a population"
(Garcia Coll 1990275). The relationship between low birthweight and infant
mortality makes the study of low birthweight of importance to a population.
Relationship Between ng B'flgweight gnd Infant Mortality Rgt_e

Low birthweight is the most important predictor of infant mortality. The

more we learn about low birthweight, the more we can apply knowledge to

9

reduce infant mortality. Since Mexicans and Cubans have better low
birthweight outcomes than Non-Latino \Miites (Becerra, Hegue, Atrash, Perez
1991; Mendoza Venture, Valdez, Castillo, Saidivar, Baisden, and Martorell
1991; Health US. 1990 1992; CDC 1993), using these groups as models, may
reduce the incidents of low birthweight (and infant mortality) in the larger
pepulation.

Dr. Paneth (1995) suggested that one of the most comprehensive
studies on infant mortality was conducted by Carol Hogue and her colleagues
(Hogue et al. 1987). These researchers argue that the most effective method
of reducing infant mortality is to reduce low birthweight (Hogue et al. 1987).
They collected data from 50 states, New York City, the District of Columbia,
and Puerto Rico. Data consisted of birth and death certificates for infants born
in 1980 and who died in the first year of life. For the 1980 cohort (all races), it
was found that 97.6% of the infant deaths per 1,000 live births were low
birthweight infants (Hegue et al. 1987). Six risk factors for infant mortality were
identified. These factors were (1) gender, (2) gestational age, (3) live birth
order, (4) matemel age, (5) matemel education, and (6) prenatal care.

According to Hogue et al. (1987) more low birthweight males died than
low birthweight females. As gestational age increased, birthweight increased.
As birthweight increased, infant mortality decreased. Low birthweight infants
born second had the lowest infant mortality rate among low birthweight infants.

For infants over 2,500 grams, those born first had the lowest infant mortality

10

rate. The infant mortality rate decreased as the age of mothers increased
through 34 years of age. However, the infant mortality rate increased for births
to mothers over 35 years of age. Infant mortality associated with younger
females is related primarily to low birthweight. As matemel education
increased, infant mortality rate decreased. When prenatal care was received in
the first trimester of pregnancy, the infant mortality rate was substantially lower.
The Hogue et al. (1987) comprehensive study makes clear the association
between risk factors, such as low birthweight, and infant mortality.

Unfortunately, Latinos were not included in the Hogue et al. (1987) study.
Hogue et al. state that "Hispanics have emerged as a major ethnic group in the
United States having distinct reproductive life histories. However, because
states use a variety of methods to define Hispanics, a national study of
Hispanics could not be launched with the NIMS data” (Hogue et al. 1987). This
emission is indicative of early research on Latinos (i.e., they were ”left out").

A more recent study (using available 1990 data) on infant mortality rate
(From the MMWR 1993) found the infant mortality rate for the US. to be 9.2
infants per 1,000 births. The data was derived from the National Center for
Health Statistics and the Center for Disease Control. For the total population,
disorders relating to short gestation and low birthweight were the third leading
causes of death. For African American infants, disorders relating to short
gestation and low birthweight were the first leading causes of death and were

the third leading causes of death for Non-Latino White infants (From the

1 1
MMWR 1993:428). in 1987 (the most recent data at the time of the study),

95% of infant deaths were low birthweight babies. This is a slight decrease
from the 1980 statistics used by Hogue et al. (1987).

Infant mortality is not only effected by low birthweight, but a relationship
exists between degree of low birthweight and incidence of infant mortality. The
lower the low birthweight, the more likely an infant will die. Rumbaut and
Weeks (1989) in a study on infant mortality among lndochinese dichotomized
their sample by four birthweight categories for this very reason. They write:

We have chosen birthweight groups (<500, SOC-1,499, 1,500—2,499, and

>2,500 grams) that exhibit markedly different risks of deeth--reflecting the

geometric increase in death rates as birthweights decline (Rumbaut and

Weeks 1989156)

The methodology of Rumbaut and Weeks (1989) suggests that when
assessing the effect of independent variables upon the dependent variable of
low birthweight, if a researcher is interested in variance in low birthweight, low
birthweight should be dichotomized by categories as they are related to infant
mortality.

The research on infant mortality and low birthweight to date suggests
that a clear relationship exists between the two. Low birthweight babies are
more likely to die than babies who are not bom with low birthweight.
Furthermore, there are specific indicators of low birthweight (e.g., gender, birth
order, mothers age and education, and access to prenatal care). Male low

birthweight babies tend to die more often than female babies. Fullterm babies

weigh more than preterm babies. First born low birthweight babies are more

12

likely to die than second bom low birthweight babies. Mothers who are young,
undereducated, and have less access to prenatal care are more likely to have
low birthweight babies and babies who die.
Risk Factors for ng Bi'mWeight and Infant Mortality

Certain behaviors and conditions have been demonstrated to bear on
low birthweight. These behaviors and conditions are new examined. Two
studies, one international and one national, examined the effect of weight gain
and nutrition on low birthweight. Kramer (1987), in a meta-analysis of French
and English studies conducted between 1970 and 1984, found that weight gain
during pregnancy effected intrauterine growth retardation (IUGR). The effect of
lUGR was greater for women who were undernourished and for women who
experienced "acute nutritional stress" during food shortages. These women
were mostly from developing countries which had "hungry seasons." In the
United States, Taffel (1980) found that women who gained less than 21 pounds
during pregnancy were 2.3 times more likely to have a low birthweight infant.
Taffel (1980) also found that the babies of these women were 1.5 times more
likely to experience infant mortality. These studies demonstrate the relationship
between nutrition, low birthweight, and the infant mortality rate.

Brooks-Gum et al. (1988:289) identified possible causes of low
birthweight to be: ”(1) demographic risks, (2) medical risks that can be
detected during pregnancy, (4) behavioral and environmental risks, (5) health

care risks, (6) physical factors with a role in low birthweight that Is still being

13
defined." Brooks-Gunn et al. (1988) also suggest that recent studies indicated

environmental stress and no social support, work patterns during pregnancy,
matemel health habits, and poor prenatal care as predictors of low birthweight.
Brooks-Gum et al. (1988) concluded that access to prenatal care and matemel
education programs are needed to reduce the incidence of low birthweight.
However, Brooks-Gum (1988) did not consider the effects of race or ethnicity
on low birthweight outcomes.

Rumbaut and Weeks (1993) in a multivariate analysis found that the
"secondary biomedical" independent variable of previous live births was the
best predictor of "infant health outcomes" (beta=.261). in this study, being bem
later put the infant more at risk for infant mortality. These findings may appear
contradictory to the Hogue at al. (1987) study which found that first born low
birthweight babies were more at risk than second bom low birthweight babies.
The difference in the studies is that Hogue et al. (1987) looked at low
birthweight births where the Rumbaut and Weeks (1993) looked at births In
general.

In critiquing the Rumbaut and Weeks (1993) study, low birthweight was
just one measure of infant health outcomes. The study was based on the San
Diego Comprehensive Perinatal Program 1989-1991. The sample included
Asians, Latinos, Non-Latino Whites, and African Americans. Although the
findings of this study give insight into infant health outcome, it too is limited to a

specific geographic area. Also, the study did not subgroup Latinos.

14
Wrthweight versus Fgllterm Low Birthweigm

The less time an infant has in the womb, the less it fully deveIOps. A
less developed infant is less likely to weigh as much as its fulltenn counterpart.
Very simply, preterm infants experience low birthweight due to less time in
gestation (Paneth 1995). Michielutte et al. state that "failure to distinguish
between preterm and term low birthweight would be highly misleading"
(19922103). Differences in preterm low birthweight infants and term low
birthweight infants exist. Michielutte et al. (1992) address these differences.
The study sample consisted of 25,758 low birthweight singleton births from 20
Counties in North Carolina. Data was collected from physician records. These
records included socioeconomic, physical, and medical history factors
associated with low birthweight. The majority of the subjects were clients of
Public Health Clinics.

Preterm low birthweight was defined as less than 37 weeks gestation
and weighing less than 2,500 grams. Term low birthweight was defined as 37
or more weeks gestation and weighing less than 2,500 grams at birth. One
thousand seven-hundred twenty-two births were preterm low birthweight and
1,098 births were term low birthweight.

Adjusted odds ratios for all low birthweight, preterm low birthweight and
term low birthweight births were conducted. Adjusted odds ratio (AOR) is a
measure of association between dichotomous variables. 1.0 indicates no

relationship. Michielutte et al. (1992) suggest that an AOR of 1.30 and above is

15

a significantly strong association. A positive AOR indicates a positive
relationship. A negative AOR indicates an inverse relationship. Negative AORs
are expressed as numbers less than one (e.g., .79). The final risk assessment
model included all risk factors with an AOR of above 1.30 (or less than .77).
The AOR results of the significant risk factors are given in Appendix B. A
discussion of the results is given below.

Age was found to be factor in both term low birthweight and preterm low
birthweight. The authors write that "it is especially interesting to note that
younger age significantly decreases the risk of term low birthweight, but
increases the risk of preterm low birthweight" (Michielutte et al. 19921104).
Adolescents are more likely to have preterm low birthweight infants but less
likely to have term low birthweight infants. As Hogue et al. (1987) pointed out,
these infants are the most likely candidates for infant mortality. This may be
due to the fact that young adolescents themselves are still growing. For many,
their bodies may not be able to sustain a longer pregnancy Which results in
preterm low birthweight. Garcia Cell, for example, argued that "it may be that
the extra nutritional requirements associated with pregnancy, coexisting with
adolescent growth and economic disadvantage, may be possible additional
growth—retarding factors for infants of teenage mothers" (19882112).

Less education was a factor in low birthweight (Michielutte et al. 1992).
The significantly strong factor of "less than nine years of education" implicates

education as an important factor in determining low birthweight. The education

16

factor may be an element of a lack of access to educational institutions or a
value system which has not internalized matemel education as a value. If this
is the case, receiving education on prenatal care would be less likely. This is
corroborated by Seitz and Apfel (1994) who found that of adolescents who gave
birth to low birthweight babies, 5.2% were in school as compared to 16.1% who
were not in school.

More African Americans had low birthweight infants than Non-Latino
Whites (Michielutte et al. 1992). Being African American and having a greater
percentage of low birthweight infants is a factor that remains largely
unexplained. in the Michielutte et al. (1992) study, even when controlling for
socioeconomic status, physical and demographic characteristics, and social
context factors, low birthweight was present.

Chemical intake also affected low birthweight. Smoking is a behavioral

 

factor that can be modified. This factor has long been recognized as

 

contributing to low birthweight. Smoking is more of a factor for term
birthweight. Perhaps this is due to the fetus being exposed to toxins for a
longer period in the womb.

The physiological processes of previous premature/low birthweight births
(preterm low birthweight and term low birthweight) were also indicated as low
birthweight factors (Michielutte et. al 1992). These factors may be an element
of the woman's physical "stamina" (i.e., the ability to support and recover from

impregnation). Also, a genetic component may be operating. These factors

17

are in the realm of medical science and have yet to be explained.

Nutrient intake was presented as a low birthweight factor (Michielutte et
al 1992). Weight under 100 pounds is a factor for preterm low birthweight.
Women in this category are most likely experiencing some type of nutritional
deficiency. Regarding term low birthweight (in order of most risk) shorter
women who weigh over 100 pounds are at risk (shorter and heavier), as are
taller women who weigh less than 100 pounds (taller and lighter), and women
who are shorter and weigh less than 100 pounds (shorter and lighter). Except
for the shorter and lighter women, these outcomes suggest that body proportion
in relation to weight effects term low birthweight.

Nutrition is a modifiable risk factor. Michielutte et al. propose that "public

AME

health cfinic wears"; raise“ proportion ofwhom are likely to have less access
to economic resources and nutritional information, could benefit most from
dietary intervention" (19922107). This suggests that members from lower
socioeconomic statuses (e.g., Mexicans, see Appendix H) would benefit from
"dietary information."

The Michielutte et al. (1992) study offers some excellent insights into the
issue of preterm low birthweight and term low birthweight and their similarities
and differences. However, the study is limited to North Carolina. A national
study of this type would greatly enhance the literature pool on this topic. Lastly,
and this is related to the regionalism of the study, Latinos are not addressed.

The studies reviewed suggest that a strong relationship exists between

18

low birthweight and infant mortality. Other prominent factors are length of
gestation, maternal education, smoking, weight, medical history relating to
pregnancy, age, education, and nutrition. Appendix C summarizes the findings
of the literature on the biomedical and risk factor aspects of low birthweight.

The focus of my study is on nutrient intake. Nutrition, weight, low
birthweight, and infant mortality rate are interrelated. Poor nutrition can result in
either too little or too much weight gain. Poor nutrition can also result in
intrauterine growth retardation. These factors can result in low birthweight.
Low birthweight increases the likelihood of infant mortality. Like other health
outcomes, the incident of low birthweight and infant mortality among specific
racial and ethnic subgroups needs to be examined.

Lo i h i ht an African A erican

A considerable volume of literature exists regarding low birthweight and
African Americans. This literature is helpful for it frames the low birthweight
issue as it is related to Non-Whites. For this reason, some of the pertinent
research is discussed.

According to the US. Census, in 1989, African Americans had low
birthweight rate of 13.2% as compared to 5.7% for Non-Latino Whites (US.
Bureau of the Census 1992:68). This figure is comparable to the 1988 findings
of the Center for Health Statistics which indicate a low birthweight rate for
African Americans to be 13.3% (Health United States 1990 1992:10). Low

birthweight is related to the infant mortality rate. in 1990, the infant mortality

19

rate for African American infants was 18.0 deaths per 1,000 and for Non-Latino
Writes it was 7.6 deaths per 1,000 (From the MMWR 1993:427).
Environment and Poverty
Using 1982 and 1983 Illinois vital records and US. Census income

data for Chicago, (N=93,101) Collins and David (1992) found that infants whose
I mothers had received prenatal care and came from very poor neighborhoods
had a neonatal mortality risk four times that of Non-Latino White infants. The
authors suggest that "extrinsic factors closely linked to income and education
contribute to the relative birth—weight disadvantage of black infants” (Collins and
David 1992:24). Collins and David (1992) suggest that an impoverished
environment reduces quality nutrition, housing, and safety.

Poverty is a daily reality for many in the African American community.
Forty-six percent of African American children live in poverty in the United
States (Huston, McLoyd, and Cell 1994). Low educational levels, single
parenthood, and being African American seem to be strong correlates of
poverty (Duncan, Brooks-Gunn, and Klebanov 1994). Social and economic
needs have to be considered in effecting change in low birthweight among
African Americans.

ri l t

Marital status also appears to be a factor in low birthweight outcomes for

African Americans. In their sample of 222,810 live births from Georgia (using

Vital Statistics from the Georgia Division of Public Health), Sung et al. (1993)

20

found that married women had lower rates of low birthweight than unmarried
women. Marriage may offer additional financial and emotional support. This
type of support may contribute to a secure and stable environment. Stability
and security are related to the larger socioeconomic condition of African
Americans. Better socioeconomic conditions (e.g., adequate wages, housing,
and education) can provide for an atmosphere of stability. The authors suggest
that "efforts to improve the social milieu will reduce the rate of low birthweight
among both Black and white infants" (Sung et al. 1993:134). However, the
authors also write that in addition to low birthweight being related to
socioeconomic conditions, negative low birthweight outcomes among African
Americans may be due to ”a yet-to-be determined disadvantage related to
being Black" (Sung et al. 19932133).

For African Americans, the environment of poverty and an unstable
social environment (e.g., female headed single parent households) appear to
increase low birthweight rates. However, there is much room to explore the
"yet-to-be determined disadvantage" related to African American heritage. it
may be that years of socioeconomic subjugation have resulted in an unknown
physiological adaptive component that increases the rate of low birthweight.
Although both Latinos and African Americans have histories of subordinated
statuses, African Americans were enslaved. it may be this history of human

captivity and brutality that accounts for the high low birthweight outcome.

21
Low Birthwekght endgame;

As has been discussed, low birthweight is related to infant mortality and
this holds true for Latinos. infant mortality rates are now examined to preface
the discussion on Latinos and low birthweight. A differential infant mortality rate
exists among Latinos.

l ant 0 li Rate and Latinos

The infant mortality rate for Mexicans is 8.8, for Cubans 8.0, and for
Puerto Ricans, 12.3 per 1,000 in the population. The infant mortality rate for
Non-Latino Whites is 9.0 (Health United States 1990 1992). in Brooklyn,
(where 83% of the Latinos are Puerto Rican) the infant mortality rate is 12.3. in
Los Angeies, (where 76% of the Latinos are Mexican) the infant mortality rate is
6.9 (COSSMHO 1994). The differential infant mortality rate for Latinos is
reflected in differential low birthweight data. There is little research that
specifically addresses the infant mortality rate and Latinos.

A verv‘ ti Hater an i and ow Birth ‘ t t e

Do differences in low birthweight among Latinos exist? Table 1 below
suggests that they do. Table 1 presents the results of studies that have
specifically addressed the issue of low birthweight among Latinos. These
studies are presented because they subgroup the Latino population. Other
studies that have not subgrouped Latinos but address Latinos, or, discuss only

one particular subgroup are discussed in the text subsequent to Table 1.

22
Table 1. Low Birthweight Differences Among Latino Subgroups

     

 

  

 

 

 

 

 

 

 

 

 

 

HM
Author(s) Mexican Cuban Puerto Rican
Becerra et al. 4.1% 4.0% 6.6%
1992 *
CDC 1993 5.5% 5.7% 9.0%
Health US. 1990 5.6% 5.7% 9.4%
1992
Mendoza et al. 5.7% 5.9% 9.3%
1991
US. Census 5.6% 5.8% 9.0%
1992

tur n tin a w irth l t

in a San Diego area study, Non-Latino Whites had a low birthweight
percentage of 5.1 as compared to a percentage of 5.2 for Latinos (Rumbaut
and Weeks 1989). The study focuses on the lndochinese population and
Rumbaut and Weeks (1989) suggest that ethnicity factors and language explain
most of the low birthweight among the lndochinese living in San Diego.
lndochinese have different cultures and experiences than Latinos and the
relevance of the Rumbaut and Weeks (1989) study to my study is limited.
Additionally, the results are pertinent only to San Diego due to the fact that this
is from where the data was drawn.

Hayes-Bautista (1992:36) found that "Latinas have the lowest incidence
of low birthweight babies in Los Angeles County: 5.32 percent.” He is quoted

as saying that the low birthweight rate of Latinas is "half the rate of Black

23
babies and a hair better than Anglos" (Spiegel 19912A3). This data is most

likely referring to Mexicans since the majority of Latinos that live in Southern
Caiifomia are Mexican (COSSMHO 1994). However, this is not indicated.

Using the 1983 and 1984 Linked Birth and infant Death Data Sets,
Becerra et al. (1991) found low birthweights of 4.1% for Mexicans, 4.0% for
Cubans and 6.6% for Puerto Ricans. The percentage of low birthweight babies
for all Latinos was 4.6 and for Non-Latino Whites 4.0. Becerra et al. state
"although better nutritional practices, higher regard for parental roles, and lower
rates of smoking and alcohol consumption have been proposed as possible
reasons for the better birth-weight distribution among Hispanics, [as compared
to Non-Latino Whites] these factors partially explain but do not totally account
for the difference" (Becerra et al. 1991).

Data from the 1987 National Vital Statistics System and the Hispanic
Health and Examination Survey, 1982-84, indicated low birthweight percentages
of 6.2 for Latinos and 5.6 for Non-Latino Whites (Mendoza et al. 1991). Low
birthweight for Latino subgroups indicated low birthweights of 5.7% for
Mexicans, 5.9% for Cubans, and 9.3% for Puerto Ricans. in a Chicago area
study comparing African Americans with Mexicans, a review of medical records
indicated that Mexicans had a low birthweight percentage of 5.9 (Dowling and
Fisher 1987).

Recent statistics from the Center for Disease Control reflect low

birthweight for Mexicans to be 5.5%, Cubans 5.7%, and Puerto Ricans 9.0%.

24

The percentage of low birthweight for all Latinos was 6.1 and for Non-Latino
Whites 5.6 (CDC 1993). in another government sponsored study, low
birthweight among Mexicans, Central and South Americans, and Cubans
ranged from 5.6% to 6.0%. Low birthweight among Puerto Ricans was 9.4%
and Non—Latino Whites 5.7%. (Health United States 1990 1992). Latest US.
Census Bureau statistics yield similar results with low birthweight rates for
Mexicans of 5.6%, Cubans 5.8%, and Puerto Ricans 9.5%. The rate for all
Latinos was 6.2% and the rate for Non-Latino Whites 5.7% (US. Bureau of the
Census 1992). in 1991, Ventura and Martin found low birthweight rates of
6.1% for Latinos and 5.7% for Non-Latino Whites. in accounting for group
heterogeneity, low birthweights were 4.8% for Cubans and 7.9% for Puerto
Ricans (in Mendoza 1994). Appendix D summarizes the research reviewed
regarding Latinos and low birthweight.

What does this mean? The literature suggests that Mexicans and
Cubans have the lowest rate of low birthweight among Latinos and Puerto
Ricans have the highest. Heterogeneity exists among Latinos and their low
birthweight outcomes. Why is this important? it is important because this
means that each Latino subgroup has its own distinct set of circumstances that
causes these outcomes. As discussed in the next section, a number of
researchers have argued that acculturation explains these differences in low
birthweight. i now turn to a discussion of the literature on acculturation and low

birthweight among Latinos.

25
Literaturg on Acculturation Efiects_and1_.ow Birthweight

The degree to which one acculturates can effect their behavior. A
change in behavior can have an effect on health outcomes. This is seen in low
birthweight outcomes among Latinos.

in San Diego, foreign born Latinos (1.9% low birthweight) fared better in
terms of low birthweight than US. born Latinos (4.1% low birthweight)
suggesting that "Angloization" results in poorer health outcomes (Rumbaut and
Weeks 1993). Angloization refers to becoming acculturated to the Non-Latino
White culture. Fenster and Coye (1990) found that only 1.8% of babies born to
a sample of Latinas employed in agriculture in Caiifomia (N=1,040) were low
birthweight (Fenster and Coye 1990).

Using the Hispanic Health and Nutrition Examination Survey, 1982-84,
Scribner and Dwyer (1989) found that 4.8% of Mexican infants in the sample
were born with low birthweight. Scribner and Dwyer (1989) argue that as
acculturation increases, so does the incidence of low birthweight. Acculturation
was measured by language preference, nativity, and ethnic identification.

Since the majority of Latinos in the United States are Mexican (US.
Bureau of the Census 1992), and the majority of Latino immigrants are Mexican
(COSSMHO 1994), research on pregnancy and birth issues in Mexico is
considered. Research on pregnancy and birth issues in Mexico is considered
because it gives some insight into possible acculturation effects. Using the

National Addiction Survey (NAS) of Mexico, Borges, Lopez-Cervantes, Medina-

26

Mora, Tapia-Conyer, and Garrido (1993) found an incidence of low birthweight
among their sample (N=5,234) of 3.2%. Women who suffered from Alcohol
Dependence Syndrome were at very high risk of low birthweight and preterm
delivery (odds ratio=12.1). The data on low birthweight was obtained in
interviews and was based on mother's recollection. The data suggests that the
incidence of low birthweight in Mexico is lower than for Mexicans living in the
United States. This supports the notion that Angloization may result in poorer
low birthweight outcomes.

Balcazar (1993) also suggests that Angloization results in poorer low
birthweight outcomes. Using non-marriedstatus, low education, young age at
delivery, at least one pregnancy complication, recent illness, at least one labor
complication, and no prenatal care as risk factors, Balcazar (1993) found that
Mexican Americans had a lower risk for intrauterine growth retardation than did
Non-Latino Whites. Also, in the sample of 25,289 Arizonan Mexican
Americans, "US—bom Mexican American mothers were 1.21 times more likely to
have an intrauterine growth retardation infant than were Mexican-hem mothers"
(Balcazar 1993:172).

Gaviria, Stern, and Schensul (1982), in a sample of 89 women from the
Chicago area, found that recent arrivals to the US. (immigrated in previous two
years) and Mexican American women sought prenatal care later than long term
arrivals (been in the US. more than two years). What is interesting about this

finding is that it lessens the importance of prenatal care in pregnancy outcomes

27
which has been argued by some (Hogue et al. 1987; Brooks-Gum 1988) as

being very important for positive outcomes. However, the Gaviria et al. (1982)
study found that more recent arrivals sought prenatal care later in pregnancy (a
low birthweight risk factor). This suggests that it is not just prenatal care that
accounts for low birthweight outcomes and other factors (e.g., cultural) have to
be considered.

Regarding culture, in Mexico a postpartum practice called cuarentena
exists (Gaviria et al. 1982). In this practice, the mother has a 40-day rest
period following delivery. During this time, the mother focuses on her and her
newbom's health. This may include, but not be limited to, nutritional practices.
Gaviria et. al. write "specific behaviors may include dietary restrictions (e.g.,
chili, pork, tomatoes, citrus fruit, and other 'coid foods') and prescriptions
(chicken, chicken soup, and fried tortillas)" (19822986). The authors state "in
our sample [N=89] we had information about the cuarentena from 56 mothers;
of these 34 (60 percent) were planning to observe the cuarentena or had
observed it" (Gaviria et al. 1988:986). if postpartum practices such as the
cuarentena exist, pregnancy practices must exist that would influence low
birthweight outcomes. These practices may not be what is common to the
dominant culture (e.g., prenatal and postpartum care received by a physician or
health institution).

Weinman and Smith (1994) in a study of US. born and Mexican born

adolescent Latina mothers, found that for both groups postpartum follow up was

28
minimal (17.3% of a sample of 289). Weinman and Smith (1994) suggest that

the two groups are more similar than different. Similarity may be a result of
persistent cultural norms regarding pregnancy and childbirth. The authors write
"Many Mexican-Amencans explicitly prefer the traditional assistance of Pateras
or lay midwives rather than medical institutions" (Weinman and Smith
1994:192).

The literature on acculturation and low birthweight suggests that
acculturation is detrimental to positive pregnancy outcomes (i.e., results in a
higher incidence of low birthweight). This idea of acculturation as detrimental is
especially clear when looking at the low birthweight outcomes for women in
Mexico and recent immigrants from Mexico (acculturation will be fully discussed
in Chapter 3).

Implications

Methodological problems have been identified with the literature
reviewed. Hogue et al. (1987) did not address the role of nutrient intake as a
risk factor for infant mortality rate and did not subgroup Latinos. Michielutte et
al. (1992) discussed the role of weight gain but did not address nutrient intake
specifically as a risk factor of low birthweight. Michielutte et al. (1992) also did
not subgroup Latinos. The neglect to subgroup was a common criticism of
several studies (Brooks-Gum et al. 1988; Dowling and Fisher 1993; Fenster
and Coye 1990; Gaviria et al. 1982; Hayes-Bautista 1992; Rumbaut and Weeks

1989; Rumbaut and Weeks 1993).

29

Many studies were regional in scope (Balcazar 1993; Dowling and Fisher
1993; Gaviria et al. 1982; Hayes-Bautista 1992; Fenster and Coye 1990;
Michielutte et al. 1992; Rumbaut and Weeks 1989; Rumbaut and Weeks 1993;
and Weinman and Smith 1994). Regionalism may bias results and it may
create problems in generalizing results to larger populations. For example,
studies based on data from San Diego may yield results that are different from
studies based in Miami. That is to say, the whole cultural milieu may be
different.

in addition to problems associated with regionalism and neglecting to
subgroup, the role of food habits as related to culture was not discussed in
relation to Latinos and low birthweight. This is problematic for this fails to
recognize Latino heterogeneity in relation to culture and nutrition as related to
health outcomes. In only two articles were food and nutrition specifically related
to health outcomes. These articles were the Gaviria et al. (1982) discussion of
the cuarentena and Garcia Coils recognition of good nutrition as a health
promoting lifestyle.

The literature reviewed has indicated that low birthweight is a complex
phenomenon that includes biomedical and social correlates. What is known is
that low birthweight increases the likelihood of infant mortality and this is why
the issue is of importance. Several risk factors contribute to low birthweight.
Some of these factors are physiological and some are sociological. However, it

is difficult to separate the physical from the social. For example, the rate of low

30

birthweight babies is greater for babies born preterm than fullterm (physiological
correlate). The likelihood of having a preterm baby is effected by social
correlates (e.g., mother's age, low education, smoking).

There are other social correlates that can effect low birthweight. Two of
these correlates are race and ethnicity. African Americans have the highest
incidence of low birthweight. The high rate of low birthweight among African
Americans may be due to poverty and poor environmental living conditions.
Mexicans and Cubans have better low birthweight outcomes than do Puerto
Ricans suggesting cultural heterogeneity.

Wnat is not known about differential outcomes of low birthweight among
Latinos is the role that nutrition plays in effecting outcomes. As yet, we do not
have a clear understanding of the way food habits are rooted in culture.
However, it is plausible to infer a relationship. The next chapter examines

acculturation processes that may be related to food habits.

CHAPTER 3

ACCULTURATION AND ACCULTURATION PROCESSES

Early on in this dissertation, borrowing from Garcia Coll and Meyer
(1993211), l defined acculturation as a "process of leeming about a new culture,
and deciding what is to be saved or sacrificed from the old culture." The key’
word here is process. This chapter examines the process of acculturation as I
believe it to operate. By knowing what it means to be acculturated, we can
begin to make between group comparisons in terms of more or less
acculturation. This is important because knowing about degrees of
acculturation may shed light on why differential low birthweight outcomes exist
between Latino subgroups.

Scribner and Dwyer (1989) in their study of acculturation and low
birthweight among Latinos measured acculturation by developing an
acculturation index using eight variables. Three of these eight variables
measured ethnic identification (Scribner and Dwyer 198921263). The other
variables in the study were language preference and nativity. The ethnic
identity variables were scored from "1 (Mexican-Orientation) to 5 (US-
orientation)" (Scribner and Dwyer 1989:1263). Ethnic identity is similar to
cultural identity as discussed by Bean and Tienda (1987).

Bean and Tienda (1987) propose that historical and social processes

significantly affect cultural identity. These processes are interactive and impact

31

32

the contemporary social and economic situation of Latinos which in turn
influence cultural identity. History and social circumstances assist in
"interpreting the diverse integration experiences and socioeconomic standing of
Hispanics" (Bean and Trends 1987:15). Change in cultural identity is analogous
to acculturation.

My discussion of acculturation processes is similar to the approach of
Bean and Tienda (1987). It is difficult (if not impossible) to discuss
acculturation in isolation. Diaz-Guerrero articulates this when he writes that the
"concept of culture is closely related to the concept of history. Culture at a
given time is the result of history; it is actually, in a sense, history at that time"
(1977:19). The history of Mexicans, Cubans, and Puerto Ricans is diverse.
Each group’s culture is the result of their distinct circumstances. The Mexican
experience is one of conquest and economic exploitation. The Cuban
experience is one of Cold War international politics. The Puerto Rican
experience is one of dependency and patriarchy. Appendix E discusses in
detail the differential historical circumstances of Mexicans, Cubans, and Puerto
Ricans. Although i am not going to elaborate on differential historical
circumstances here, there is a concept related to history that is important to the
discussion of acculturation. This concept is that of "life worlds."
masons:

The interdependent nature of functionalism makes it relevant to this

study. As was seen in the literature review, the phenomenon of low birthweight

33

is related to biological/medical, demographic, behavioral, and cultural factors.
These factors are not isolated events but interact with each other. For
example, Angloizaton (a cultural factor) interacts with behavioral factors that
can increase the incidence of low birthweight births (Balcazar 1993; Borges et
al. 1993; Rumbaut and Weeks 1993). The influence of functionalism is seen in
Jurgen Habermas' (1976) Legitimation Crisis.

In Legitimation Crisis (1976), Jurgen Haberrnas discusses crises as
related to two integration systems; social and system. The system integration
system is representative of the dominant society. It seeks to control and dictate
the society. The social system is related to culture, it is based on symbols.
Society is always evolving. This evolution is based on what best functions for
the dominant society. However, what is best for the dominant society is not
necessarily best for all members of that society and this leads to conflict.

Conflict arises because groups in society seek to maintain their social
system and they do this through life-worlds. Life-wands are the symbols,
rituals, and traditions of a culture. Life-worlds facilitate and allow for
intergenerational communication and the transmission of a cultural identity.
Life-wands develop a person's sense of self. Life-wands are borne out of
history. Since the history of Latino subgroups varies, the life-worlds vary.

Symbols are the cornerstone of the school symbolic interaction.
Symbolic interactionsim finds its roots in the work of Charles Horton Cooley

(1962). Cooley (1962) proposed that a person's sense of identity develOped out

34

of how those in the environment reacted to them. He writes that "self and
society go together, as phases of a common whole" (Cooley 196229). The
person cannot separate themselves from the society of which they are a part.
Cooley developed the concept of the "looking glass self” in which a person
imagines how they appear to others and further imagines how others judge this
appearance. The person then feels either "pride or modification" based on their
sense of how others have judged them (Cooley 1962223).

Symbolic interactionism was further developed by George Herbert Mead.
Mead proposed the concepts of the "Me" and the "i." For Mead, the "Me" was
the "organized set of attitudes of others which one himself assumes"
(196452320). The "I" was "the response of the organism to the attitudes of
others" (Mead 196429). The Me and the i make up the "Self." Mead writes:

When one determines his position in society and feels himself as having

a certain function and privilege, these are all defined with reference to

an "I" . . . The "i" reacts to the salt which arises through taking of the

attitudes of others. Through taking those attitudes, we have introduced

the "Me" and we react to it as an "i" (19642229).

For both Cooley (1962) and Mead (1964) , the individual does not exist
in isolation. The Individual forms their sense of who they are based on how
others respond to them. Symbols play a part in this interaction between the
individual and others. Messages about who the Individual is in relation to
others and the environment can be communicated through symbols. This

communication can then impact a person's sense of self. Symbols, then, are

critical in the formation of one's identity. Habem'ias (1976) recognized this and

35

made the role of symbols of extremely important in his development of the
concept of "life-worlds."

Symbols create and maintain life-worlds. i believe that symbols are an
important aspect of Latino society. I believe this for l have experienced it in my
own life-world. For example, I grew up with pinatas and menudo. I am still
"mijo" to my mother ("mijo" is derived from "ml-hljito" which means "my little
one"). Symbols such as these result in the persistence of Latino life worids.
This persistence enable the regeneration of Latino culture and prevents the
usurpation of the social integration by the system integration. This promotes
the effective internalization of culture. lntemalization of culture reduces the
effect of acculturation processes which can have positive benefits. in
discussing Latinos and acculturation, Magana and Clark (19952105) write, "there
may be particular, compelling, positive elements that prevent acculturation to
beliefs and behaviors that put individuals at risk for serious health problems.“

Others have also viewed symbols as being an important aspect of Latino
society. In discussing the relationship between low birthweight and culture,
James (1993) argues that there is an unexplained factor that contributes to the
low rate of low birthweight among Mexicans. He asks:

is it possible that Mexican American women with a predominately

Mexican cultural orientation derive important psychological benefits from

maintaining contact with identity-preserving symbols? (James 19932132).
James (1993) suggests that psychological benefits are derived from a Mexican

cultural orientation steeped in symbols, and this orientation results in positive

36

health outcomes. James‘ (1993) argument is consistent with Haberrnas'
discussion of life-worlds, particularly the emphasis on symbols in the
maintenance of culture. Religion has also been proposed as important factor in
defining culture.

Magana and Clark (19952103) write that:

One religious symbol uniquely central to Mexicans and Mexican

Americans is the Virgin of Guadalupe . . . She is for Latinos of Mexican

descent a symbol of both cultural and religious identity, usually

inseparable within Mexican American culture.

According to Magana and Clark (1995), the Virgin of Guadalupe is a
religious symbol that provides strength, warmth, and power to Mexicans and
particularly to Mexican women. This is due to the connection of the Virgin of
Guadalupe with motherhood. The Virgin of Guadalupe is representative of
maternal strength, the protection of children, and life itself. Myth has it that the
Virgin of Guadalupe (also known as Our Lady of Guadalupe) appeared to a
Nahuatl lndian named Juan Diego in what is now Mexico City in 1531. She
was said to be the dark-skinned Mother of God (the Virgin Mary). --

Magana and Clark (1995) suggest that it is this symbol that contributes
to the positive birth outcomes for Mexicans. They argue that because of the
maternal association with the Virgin of Guadalupe, Mexican women "are very
likely to have the kind of health behaviors that augur toward better birth
outcomes and more robust infancy" (Magana and Clark 19952106). Specifically,
Magana and Clark (1995) suggest that women who proscribe to the power of

the symbol of the Lady of Guadalupe, in an effort to emulate her, will avoid

37

smoking and alcohol, be modest in their sexual relationships, and maintain a
diet based on traditional foods. Magana and Clarks' (1995) suggestion that
some Mexican women have an especially strong relationship with the symbol of
the Lady of Guadalupe is consistent with a study conducted by Levin (1991)
who found that Mexican women prayed more for their babies during pregnancy
than African American women (although African American women attended
church more often).

Habermas' (1976) concept of life—worlds and the importance of symbols
in the maintenance of life-worlds, James' (1993) discussion of the role of
symbols in promoting favorable low birthweight rates, and Magana and Clarks'
(1995) article on the Lady of Guadalupe as a symbol that contributes to a low
incidence of low birthweight, demonstrate that others' have considered the
importance of culture, symbols, and social psychological phenomenon in
explaining low birthweight. l have reviewed these studies to present a
theoretical construct in which to frame the discussion of acculturation and
acculturation processes. These studies and particularly the concept of life-
worids provide an alternative way of thinking about culture and acculturation.
This is to say that in thinking about acculturation, the role of life-worlds (with its
focus on symbols) can not be overlooked or understated.

WW
Marin and Marin (1991) suggest that there are "cultural values" that are

common to all Latino groups. They do recognize that Latino heterogeneity

38

exists but argue that common values are shared due to a common language,
historical roots of Spanish colonization, and the shared religion of Roman
Catholicism. However, the immediate discussion focuses on acculturation
processes as related to subgroup differences. For this reason, Marin and
Marin's (1991) discussion of Latino similarities can be found in Appendix F.

The majority of the literature regarding acculturation and Latinos
addressees Mexican Americans. This is logical since the majority of Latinos in
the US. are Mexican. According to the US. Census Bureau, 60.1% of the
Latino population in the US. is Mexican, 12.1% Puerto Rican, and 4.7% Cuban
(US. Bureau of the Census 1992221).

R l ami

Martinez (1977230) suggests that Mexican Americans adhere to
"traditional family roles" in which the male is dominate ("machismo" see Marin
and Marin 1991 in Appendix F) and the female submissive. Martinez proposes
that

Utilizing the apparent adherence of Mexican-Americans to traditional

family roles one may begin to develop distinctions between Mexican-

American and Anglo-American family structure that should lead to a

better understanding of how acculturation forces have changed the

Mexican family prototype within the milieu of the current Anglo—American

society (1977:30-31).

in this context, not only can Mexican Americans be compared with
Anglos, but Mexican Americans can be compared with Cubans and Puerto

Ricans. in other words, the degree to which Mexican Americans, Cubans, and

Puerto Ricans conform to “traditional" family roles can be used as a measure of

39

comparison. I accept Martinez' assertion that Mexican Americans adhere to
traditional family roles, but only to a certain extent. I believe that the degree to
which traditional roles are adhered to is a function of acculturation (i.e., the
more they are acculturated, the less they adhere to these traditional roles).

Are Latino families really different than Non-Latino White families in
family practices and child-rearing and does this effect low birthweight
outcomes? Garcia Coll suggests that there are "major sources of influence on
the developmental outcome of minority infants: [these sources are] cultural
beliefs and caregiving practices, health status and health care practices, family
structure and characteristics, socioeconomic factors, and biological factors"
(Garcia Coll 19902271). These factors can influence how an individual deveIOps
and the environment in which they develop. Development over the life span is
effected by these characteristics which in turn effect health outcomes (i.e., the
probability of experiencing risk) (Perkins 1995). Low birthweight is a health
outcome. Garcia Coll suggests that the lower rates of infant mortality and low
birthweight for Mexican-Americans is puzzling given their low socioeconomic
status which points out the "possibility of some protective factors within this
population" (19902275). Dowling and Fisher write that "it is possible that some
‘protective sociocultural effect' exists that attenuate some of the negative factors
that are linked to race, poverty, and pregnancy outcome" (19872157). To
summarize, and to link the concepts together; the developmental outcome of

minority children is influenced by a variety of factors which eifect individual

40

development and environment which can impact exposure to risk (thus resulting
in unhealthy outcomes). There may be certain elements about Latino families
and family practices that provide a context which influences these factors and
development which in turn reduce risk (i.e., they are "protective"). The
influences in Latino culture that may protect children from risk may be in the
cohesion that is brought about by familialism. The concept of protectiveness is
consistent with the concept of familialism as discussed by Marin and Marin
(1991) and Baca-Zlnn (1994) (see Appendix F).

in discussing modern trends in the Mexican American family, Baca-Zinn
(1994) argues that the family should be viewed within the larger context of
adaptation to social and economic conditions. Baca-Zinn (1994) suggests that
traditional Mexican family values of close relationships and male domination are
not deviant, but are adaptive measures in an often inhospitable social
environment. In short, an analysis of the family has to consider culture in
relation to social structure (Baca-Zinn 19942166) because "culture and SES are
inseparably linked" (Fracasso, Busch-Rossnagel, and Fisher). The modem
Mexican family tends to have a lower socioeconomic status than Non-Latino
Whites, are more likely to be married and many at a younger age than Non-
Latino Whites, and have higher fertility rates and larger families than Non-Latino
Whites (Baca-Zinn 1994:166-67).

The intersection between culture and structure is seen in the interaction

between migration, extended family, and economic conditions. Many Mexicans

41

who migrate to the US. already have family in the states. These families form
the basis for an extended family network. in times of economic adjustment, the
extended family may assist financially and help with child care (Baca-Zinn
1994:168-69).
n Roi

Gender roles are directly related to a discussion of the family. Gender
roles are connected to the degree of acculturation of the family because in
Latino culture, women are associated with the propagation of culture. As
gender roles change so does the nature of the family and culture transforms.
The connection between gender roles, family, and culture can be analyzed in
looking at the Mexican experience. Historically for Mexicans, women's "primary
task was to care for their husbands and children and to accept subordination as
a natural condition" (Baca-Zinn 19942164). Not only was a woman's task that of
caregiver, she also promoted culture. Baca-Zlnn writes:

Besides their roles as worker, wives, and mothers, women in particular

guarded Mexican cultural traditions within the family . . . These Included

folklore, songs, and ballads, birthday celebrations, saint's days,

baptisms, weddings, and funerals in the traditional Mexican style.

Through the family, Mexican culture was nurtured (19942165).

Baca~Zinn (1994) suggests that as more Mexican women entered the
workplace, egalitarianism increased in Mexican families. This move toward
egalitarianism was driven by economic imperatives (i.e., the need for women to

contribute to the family income). Although elements of patriarchy still do exist

in attitudes toward family and gender roles, "certain social conditions appear to

42
be associated with greater equality for wives" (Baca-Zinn 1994219). The

relationship between gender roles, families, culture, and economic factors
provides a basis for a comparison of the degree of acculturation between Latino
subgroups. The effect of acculturation on family is seen in a study conducted
by Rueschenberg and Buriel (1995).

In a study by Rueschenberg and Buriel (1995) on acculturation and the
internal and external functioning of Mexican families (N=45), the authors used
measures of acculturation that included "language preference and proficiency,
generational status, and recency of immigration" (Rueschenberg and Burial
1995217). Rueschenberg and Burial (199) found that acculturation increased as
family members became more involved in US. society. Involvement in US.
society (e.g., family members working) changed the family's culture.

The Cuban family experience demonstrates a pattern similar to that of
Mexicans. in pre-Castro Cuba, "the wife and homemaker was the feminine
ideal, and most women sought personal realization in the home rather than the
workplace" (Garcia 1994:208-9). "Homemaker" is a subordinate role and in this
context, it can be argued that the element of machismo (see Marin and Marin
1991 in Appendix F) was present in pro-Castro Cuban culture. Women only
worked when economics so dictated. Garcia (19942208) writes:

A single-income household, determined by choice rather than

circumstance, was the ideal most working-class families aspired to since

it was a symbol of upward mobility; and women often used their incomes

to facilitate their family's entrance into the middle class.

However, emigration to the US. after the Cuban revolution resulted in

43

downward mobility for most Cubans. Although many Cubans were credentialed
professionals, most could not replicate the economic status they had held in
Cuba. Regardless of socioeconomic status in Cuba or skills or educational
level, the majority of the emigres came to the US. penniless. Therefore, many
women were compelled to enter the workplace.

Women began working in menial labor jobs that did not require good
English skills and had minimal contact with the public. In 1964, as a result of
so many Cuban women seeking employment, the Cuban Refugee Center in
Miami instituted a work training program to assist single women and household
heads in becoming more employable (e.g., English lessons, transportation
stipends, and day care centers).

Governmental assistance afforded to Cuban emigres is an example of a
socioeconomic condition which differentiates the Cuban experience with the
Mexican experience. Although both groups experienced a transfon'natlon of
gender roles due to economic conditions, Cubans benefitted occupationally and
educationally from governmental assistance programs. No such programs were
designed and implemented for Mexicans. This governmental assistance
facilitated Cuban women's success in the social and economic world.

Although Cuban women contributed greatly to the economic situation of
the family, like Mexicans, they were the propagators of culture "because of the
traditional association of culture with the female sphere" (Garcia 19942212).

Garcia writes:

44

Cuban women were perhaps most influential in cultural matters,
Specifically preserving cubanidad, or those customs, values, and
traditions associated with being Cuban (19942212).

in this regard, Cubans and Mexicans are similar in viewing women as
the "guardians of culture." However, social and economic forces prompted
changes that differentiated Cuban women from Mexican women. It is my sense
that Cubans are more acculturated than Mexicans and Puerto Ricans and
Mexicans are more acculturated than Puerto Ricans. i think this because of the
greater involvement in of Cubans and Mexicans than Puerto Ricans in
economic and educational institutions. Appendix H demonstrates these
differences.

i have suggested that acculturation is a process that involves several
variables and so far i have focused on family and gender roles. I have further
suggested that when family and gender roles change, then so does culture. l
have a sense that changes in gender roles may be a result of socioeconomic
conditions and these socioeconomic conditions vary by group. However, what I
am not suggesting is that socioeconomic conditions directly change culture.
Socioeconomic conditions are the only variable that I can really measure in my
analysis of culture so to argue that a direct relationship exists, would be in
error. Although I can hypothesize that socioeconomic conditions may effect
family and gender roles and this effects culture, there is no definitive process of
measuring this theoretical relationship.

The pertinent question here is, "what does acculturation have to do with

45
low birthweight?" Low birthweight is a pregnancy outcome. I argue that

differences in these outcomes are related to acculturation. In a study on
matemel-infant attachment and acculturation, Fracasso et al. (1994:145)
suggest that "if patterns of parenting behavior are culturally distinct, then they
should vary with respect to levels of acculturation." i am arguing that if patterns
of low birthweight are culturally distinct (which they are) then they should vary
with respect to levels of acculturation. The Fracasso et al. (1994) study is an
attempt ascertain if acculturation is "relevant in explaining variability in parenting
behavior and attachment." i am attempting to explain whether or not
acculturation is relevant in variability in low birthweight. I use food habits as an
indicator of culture. The relationship between food habits, culture, and low

birthweight is discussed in the next chapter.

CHAPTER 4

FOOD HABITS, CULTURE, AND LOW BIRTHWEIGHT

This chapter examines the relationship between food habits, culture, and
low birthweight. The discipline of nutritional anthropology is introduced and its
relevancy to this study is established. The chapter concludes with a discussion
of the role of nutrients as they relate to positive pregnancy outcomes.

Fogg Habits

Every human has to eat and "food is needed to supply energy for
metabolic processes, growth, and reproduction, and to supply substrates for
building the components of the organism" (Sanjur 198225). Certain dietary
essentials are necessary for the maintenance of the human organism. These
essentials are known as nutrients. Nutrients are composed of "minerals, amino
acids, and the energy sources, fats and carbohydrates" (Sanjur 198225).
However, different human social groups derive these nutrients from different
foods. The foods people eat vary by culture. Sanjur (198223) writes:

Food habits are fundamentally cultural habits. Food habits are culturally

determined; that is, the individual's subcultural background and

orientation, as well as his or her personal characteristics and
perceptions, ultimately determine what his or her dietary patterns will be

(emphasis in original).

Food habits are defined as:
Habits of a group that reflect the way a culture standardizes behavior of

the individuals in the group in relation to food so that the group comes to
have a common pattern of eating (Lowenberg, Todhunter, Wilson,

46

47

Feeney, and Savage 1968285).
Ngtriflgnal Anmrogolggy

The study of culture is associated with the field of anthropology. The
study of foods in relation to culture is known as nutritional anthropology.
Nutritional anthropology is defined as:

The application of anthropological data and methods to the solving of the

cultural aspects of human nutritional problems, or as the study of the

interrelationships between diet and culture and their mutual influence

upon one another (Freedman 1977).

Johnston (1987) elaborates upon the nature of the field of nutritional
anthropology and demonstrates its focus on culture as opposed to biology. He
writes that nutritional anthropology is:

That branch of anthropology which deals with nutrition as a process and

as a science . . . and brings anthropological concerns to the study of

food and, since food is defined culturally (rather than biologically). has a

predominant social and cultural focus (Johnston 19872ix).

Given the above, my analysis of food is in the tradition of nutritional
anthropology, since its focus is on the cultural aspect of food. it should be
noted, however, that the study is not purely cultural anthropology. The
approach is more similar to "social anthropology" since it focuses on
contemporary groups.

Differential historical and geographic conditions have created differential
socioeconomic conditions which have molded and shaped the cultures of

Cubans, Mexicans, and Puerto Ricans. Food is directly related to culture.

Lowenberg et al. (1968) write:

48
Food habits vary from one cultural group to another because each group,
in its own evolution, sets up a complex pattern of standardized
behaviors. individuals within a culture respond to the approved
behavioral pressures by selecting, consuming, and using those foods
which are available (1968287)
Back (1977) suggests that food is not only related to culture, but to one's
identity. He suggests that:
The way in which food is prepared and served, the food people eat
together, the food they do not eat in companionship-all express the
ways in which individuals in different societies project their identities
(Back 1977231).
Therefore, the foods of each Latino subgroup will be different. Differential food
habits will result in differential outcomes.
W
This difference in subgroup food habits is seen in the differential intake
of healthy and unhealthy food between Latino subgroups. Using the Hispanic
Health and Nutrition Examination Survey, 1982-84, Marks, Garcia, and Solis
(1990) found subgroup differences in dietary practices among Latinos. The
authors created a diet index which contained two components. One component
measured the degree to which one's diet was balanced, and one measured
"junk food" intake. A balanced diet meant that the person was frequently
consuming foods from each food group (meets, dairy, fruits, vegetables, and
grains). Candy, non—diet sodas, cake, cookies, sugar, etc, were considered
junk food.

Cuban women had the best scores on the balanced diet component,

followed by Mexican American women, with Puerto Rican women having the

49

lowest scores. This suggests that Mexican and Cuban women were eating
much better than Puerto Rican women. Regarding junk food, Mexican
American women have the best scores (i.e., they don't eat much junk food)
followed by Mexican American men. Puerto Rican men had the worse score
for junk food intake (i.e., they eat a lot of junk food). it seems that Puerto
Rican men have succumbed to what i call the "McDonald's Factor."

The most important aspect of the Marks et al. (1990) study is that it
supports the notion of subgroup differences in diet. It also suggests that Puerto
Ricans are at greater risk for negative health outcomes because of diet.

The heterogeneous nature of Latino foods is seen in Table 2. Table 2
compares dietary patterns and practices associated with the major food groups
by Latino subgroup.

The data in the Table 2 is based on the work of Romero Gwynn and
Gwynn (1993; 1994). Romero Gwynn and Gwynn, used the Hispanic Health
and Nutrition Examination Survey, 1982—84, and "analyzed data from this
survey . . . to determine differences in food consumption" (199324). Romero
Gwynn and Gwynn (199324) evaluated "the nutritional status and food
consumption of selected traditional and non-traditional foods." in addition,
Romero Gwynn and Gwynn (1993) examined food patterns of immigrant
families from Mexico and Mexicans born in the United States. As part of this
study, interviews were conducted in which "Hispanic food practices and beliefs"

were ascertained (Romero Gwynn and Gwynn 199324). The 1993 study of

50

Romero Gwynn and Gwynn established Latino heterogeneity in food habits. in

a later monograph, Romero Gwynn and Gwynn (1994) expanded on the history

and food types as related to Latinos of Mexican descent and noted differences

In dietary patterns by geographic region in Mexico. Table 2 was developed by

summarizing the major conclusions of the above works of Romero Gwynn and

Gwynn (1993; 1994).

Table 2. Major Food Groups and Food Practices by Latino Subgroups

Food Group

 

 

 

Cubans

 

Puerto Ricans

 

 
 
 

Com/Maize &
‘ Bread

eat more tortillas
than bread

eat more bread
than tortillas

eat more bread
than tortillas

 

 

  

 

  

 

 

Beans eat more beans eat more black eat more black
than other two beans than pinto beans than pinto
groups and beans beans
prefer pinto
beans

~ RicelPasta rice is often fried eat more rice eat more rice
and pasta than and pasta than
Mexicans Mexicans
Dairy Products milk in drinks and milk in drinks and milk in drinks

 

 

Puerto Ricans

cheese 3x week cheese 3x week and cheese 3x
week
Meats more beef than more beef than more poultry

Puerto Ricans

   
 

than Mexicans
and Cubans and
slightly more fish

 

   
 

Vegetables

 

1x week but
more Puerto
Ricans

 

1x week but
more than Puerto

Ricans
jr

 

1x week

 

 

Source: Romero Gwynn and Gwynn 1993; Romero Gwynn and Gwynn 1994.

51

Not only does Table 2 suggest subgroup differences among Latinos in
types of foods consumed, but it also suggests differences in frequency of
consumption. Frequencies were calculated by Romero Gwynn and Gwynn
(1993; 1994) from the data in the Hispanic Health and Nutrition Examination
Survey, 1982-84. These calculations were presented in their work in the form
of percentages. For example, Romero Gwynn and Gwynn (199329-10) write:

In the United States most Cubans (87.7%) and Puerto Ricans (86.4%)

do not eat corn tortillas. in contrast, almost a third (31.6%) of the

persons of Mexican descent living in the United States eat corn tortillas

on a daily basis and at least a third more (37.3%) consume them on a

weekly basis . . . Cubans (62.8%) and Puerto Ricans (51.5%) are more

likely to consume bread on a daily basis than Mexican Americans

(44.4%). »
Romero Gwynn and Gwynn's (1993) discussion on the subgroup differences in
bread and tortilla intake is translated into Table 2 as Mexicans eating more
tortillas than Cubans and Puerto Ricans, and Cubans and Puerto Ricans eating
more bread than Mexicans. This was how Table 2 was developed.
Nut_n'tjog and Low Birthweight

An interesting phenomenon presented itself in reviewing the culture and
foods literature. This phenomenon is the interactive nature of beans and
maize/rice in relation to protein. Beans are a staple of the Latino diet and is
often eaten with tortillas, bread, or rim (Romero Gwynn and Gwynn 1994).
Dewalt and Pelto suggest that "the classic trilogy of the Mesoamerican diet -

maize, beans and squash - is a well known ethnographic fact" (1977279).

Sanjur (19822247) writes:

52

When eaten combined, the lysine deficiency of rice completely
disappears owing to the large contribution of lysine by the bean protein.
Conversely, we can expect the amounts of methionine and cystine to
increase (relative to the amount found in beans alone) owing to the
contribution of the rice protein. in short, rice and beans contain
Complementary Proteins- that is, by combining them, not only do we
increase total amount of protein, we also improve the quality of both
components of the mixture (emphasis in original).

And Dewalt and Pelto find the same phenomenon for beans and maize and

state:

Many anthropologists with limited nutritional knowledge understand the

significance of the protein complementarity of maize and beans. When

lysine—deficient maize is eaten together with tryptophane-deficient beans,
the limiting amino acids in each are "complemented," making the

combination a more nutritionally complete protein (1977279)

What makes this bean and maize/rice protein complementary mix so
interesting, is the role protein plays in pregnancy. Sanjur writes that "the diets
of pregnant women are influenced by the same factors that are important
in determining diets in general" (19822259). This means that pregnant Latinas
who are eating beans and maize/rice would be receiving a good amount of
protein.

Eckstein (1980) conducted an analysis of nutrient intake
recommendations for pregnant women. She found that an increase in energy
caloric intake is necessary "to supply the extra tissues as well as growth and
development needs" (19802451). The recommended increase in caloric intake
is a mean of 300 calories a day above nonpregnant levels. A woman who is
deprived of food will be unable to meet this need. However, women who have

access to food may increase the calorie to a level that results in a negative

53

outcome. This usually occurs when a pregnant woman decreases her activity
level in the last trisemester while maintaining the increased energy intake which
results in unwanted weight gain. Eckstein recommends a mean increase of 30
grams of protein a day above nonpregnant levels “which is used for growth and
development of accessory maternal tissue and fetal tissue" (19802451).
Additionally, to develop bones and teeth, it is recommended that calcium and
phosphate intakes are increased to 400 milligrams per day above nonpregnant
levels. Iron is needed for fetal blood stores and maternal blood loss associated
with delivery. The recommended iron intake increase is 30 to 60 milligrams a
day above nonpregnant levels (Eckstein 19802452).

Iodine and sodium levels are met through "nonnai iodized salt intake"
(Eckstein 19802452). Research has been inconclusive regarding a
recommended increase in magnesium intake. A 5 milligram increase of zinc is
recommended to bolster fetal stores of zinc (Eckstein 19802452).

It is recommended that the intake of vitamin A is increased to
approximately 20-25% above nonpregnant levels. Vitamin A meets needs for
"Cell growth in accessory maternal as well as fetal tissues and tooth and bone
formation" (Eckstein 19802453). Vitamin B supplies added energy and the
recommended increase is .6 milligrams per day above nonpregnant levels. It is
recommended that ascorbic acid (vitamin C) is increased to "support normal
hemoglobin concentrations" and the recommended daily Intake increase is 10

milligrams above nonpregnant levels. Eckstein writes (1980:453):

54

From the above, it is clear that necessary increases for most nutrients
are only modest (emphasis added). A common rule is: lF protein needs
are met, THEN all other nutrient needs are likely to be met, with the
exceptions of vitamins A, C, and D (caps in original).
What Eckstein suggests, is that protein is the most important nutrient in
effecting a positive pregnancy outcome. Other nutrient requirements (except for
A, C, and D) are met through the intake of protein.

I discussed the above passage with Dr. Sharon Hoerr of the Department
of Food Science and Human Nutrition at Michigan State University regarding
the role of the above nutrients in positive pregnancy outcomes (Hoerr 1995).
She concurred that the nutrients of protein and vitamins A, C, and D were
indeed "crucial." However, she also stated that energy, folacin and iron were
also essenth for positive for pregnancy outcomes (Eckstein (1980) viewed
these nutrients as being met in the intake of protein).

Dr. Hoerr stated that curvilinear, not linear, relationships exist among
nutrients and pregnancy outcomes. A woman can intake too many calories or
too much protein. This is consistent with Eckstein's (1980) assertion that
pregnant women who ever consume are at risk for unwanted weight gain.

This chapter has analyzed the relationship between food habits, culture,
and low birthweight. The role of nutrition was addressed regarding positive
pregnancy outcomes. A theory of differential low birthweight and an operational

model to test hypotheses associated with the theory are now proposed in the

next chapter.

CHAPTER 5

THE THEORY AND THE OPERATIONAL MODEL

This chapter proposes a theory of differential low birthweight and
develops an operational model. I first develop a conceptual framework which
identifies and lists the variables of food, acculturation, socioeconomic
conditions, and the experiential histories of people. it is in this framework that
the theory is proposed. The operational model is then discussed. In this
discussion the variables are addressed and specific hypotheses proposed. The
data source is then discussed as is the data analysis.
ancegtugl Framewgrk: Functignal Relationships agd [Decry

Functional Relationships: in the equation below, I propose the following
functional relationships:

LOW BlRTHWEIGHT=liFOOD HABITS AND OTHER)

Furthermore, since food habits are a function of culture:

FOOD HABITS=RACCULTURATION)

And, since culture interacts with social and economic conditions:

CULTURE=RSOCIOECONOMIC CONDITIONS)

Lastly, socioeconomic conditions are a result of historical circumstances:

SOCIOECONOMIC CONDITIONS=fiHlSTORY)

Theory: l propose that differential historical circumstances have created

differential socioeconomic conditions which have resulted in varying degrees of

55

56

acculturation between Mexicans, Cubans, and Puerto Ricans. Mexicans,
Cubans, and Puerto Ricans each have their own distinct cultures. Food habits
are directly related to culture. Food habits create variance in nutritional intake
and it is this variance that results in differential low birthweight outcomes.

Food habits, socioeconomic circumstances, and history have been
discussed in relation to acculturation. "OTHER" refers to variables that may
have an impact on low birthweight but are not the primary focus of this
dissertation. For example, Scribner and Dwyer (1989) used language as a
measure of acculturation in examining the effect of acculturation on low
birthweight. Brooks-Gunn et al. (1988) and Michielutte et al. (1992) have
identified mother's educational level as predictors of low birthweight.

Primgry Hypothesis

i now hypothesize about the effect of the variables on low birthweight.

i expect poor food habits (l.e., undemourishment) to result in high rates of low
birthweight. Since Puerto Ricans have the highest rate of low birthweight, i
expect that Puerto Rican women have poorer nutrition than Mexican and Cuban
women.

i also tested the "OTHER" category in examining the low birthweight
issue. i tested the "OTHER" variables so I would have a basis of comparison
in analyzing low birthweight outcomes. For these reasons, as part of the
"OTHER" variables referred to in the above functional relationship, I will be

looking at the effect of language and education on low birthweight outcomes.

57
rai n l odel

The operational model is now developed. The independent and
dependent variables are discussed and hypotheses proposed. The data source
is addressed as is the data analysis.
lndeppndent Variables

Eckstein (1980), Hoerr (1995), and others discussed in Chapter 4 have
identified the following as important for comparing low birthweight, intake of ( 1)
calories, (2) protein, (3) vitamin A, (4) vitamin C, (5) vitamin D, (6) folacin,
and (7) iron. I have attempted to measure low birthweight in terms of these
factors in a logistic regression model.

To frame the variables consistently, some clarification of nutritional
terminology is in order. Vitamin C is also known as ascorbic acid (Eckstein
19802624). Vitamin D is obtained through the "absorption of calcium and
phosphorous" (Hinton and Kerwin 198127). Since calories are made up of
nutrients, the term "nutrient" will mean calories and nutrients unless specified
otherwise.

The independent variables will be referred to as the "Nutrient intake
Variables." They are measured based on 100 grams of the edible portion of a
food. This is the way nutrient intake is measured in USDA Administrative
Report 378 (US. Department of Agriculture 1994). The data set from which the
independent variables are drawn (the Hispanic Health and Nutrition

Examination Survey 1982-84), uses Administrative Report 378 as its measure

58

of nutrient intake.
Dependent Variables

Low birthweight, the dependent variable, was dichotomized based on
four categories of low birthweight. These categories were <500 grams, 500-
1,499 grams, 1,500-2,499 grams, and >2,500 grams (Rumbaut and Weeks
1989). Creating these categories resulted in 4 dependent variables. These
dependent variables are as follows:
1. Very, Very Low Birthweight = <5oo grams
2. Very Low Birthweight = SOD-1,499 grams
3. High Low Birthweight = 1,500-2,499 grams
4. Normal Birthweight = >2,500 grams

A birth weight of less than 2,500 grams was used as the standard
criterion for low birthweight in all of the studies pertaining to low birthweight
cited in the literature review. I also used this standard as a measure of low
birthweight. The independent and dependent variables are expressed in
hypotheses below.
F n 0 IR 'onshi s V ria les

H1: NUTRIENT lNTAKE=f(ETHNIClTY)

H2: LOW BlRTHWElGHT=fiNUTRlENT INTAKE)

o e s an ub otheses
H1: The nutrient intake of Puerto Rican women is less than the nutrient

intake of Mexican and Cuban women.

59

Subhypothesis 1. The calorie intake of Puerto Rican women is less than
the calorie intake of Mexican and Cuban women.

Subhypothesis 2. The protein intake of Puerto Rican women is less than
the protein intake of Mexican and Cuban women.

Subhypothesis 3. The vitamin A intake of Puerto Rican women is less
than the vitamin A intake of Mexican and Cuban women.

Subhypothesis 4. The vitamin C intake of Puerto Rican women is less
than the vitamin C intake of Mexican and Cuban women.

Subhypothesis 5. The vitamin D intake of Puerto Rican women is less
than the vitamin D intake of Mexican and Cuban women.

Subhypothesis 6. The folacin intake of Puerto Rican women is less than
the folacin intake of Mexican and Cuban women.

Subhypothesis 7. The iron intake of Puerto Rican women is less than
the iron intake of Mexican and Cuban women.

H22 Nutritional intake effects low birthweight outcomes more for Puerto
Ricans than for Mexicans and Cubans.
Was

Nutritional intake is defined as the amount of a specified nutrient as
measured in either calories, grams, milligrams, or micrograms. The intake is
based on 100 grams of the edible portion of a food.

"Less than" is defined as being numerically less (based on 100 grams of

the edible portion of a food) as measured by calories, grams, milligrams, or

60

micrograms.

Low birthweight outcomes are defined as birthweights that are in the
<500 grams, 500-1,499 grams, and 1,500-2,499 grams categories.

page

The Hispanic Health and Nutrition Examination Survey, 1982-84,
(HHANES) was recognized as a useful research tool by many scholars
(Scribner and Dwyer 1989, Mendoza 1991; Romero Gwynn and Gwynn 1993).
These are seen as endorsements of the instrument.

The HHANES data was collected by the National Center for Health
Statistics. Romero Gwynn and Gwynn state that "the best source of data on
the current nutritional status and food practices in the United States is the
national HHANES" (199424). The data contains sixteen components with
particular focus on nutritional practices and physical health (ICPSR 1994).

Hi nic e lth and u ritlon amination rve 198 -84

The sample in the Hispanic Health and Nutrition Examination Survey,
1982-84, (HHANES) consists of three subgroups of Latinos. These groups are
Mexicans (from Texas, Colorado, New Mexico, Arizona, and Caiifomia), Cubans
(from Dade County Florida), and Puerto Ricans (from the New York City area
and parts of New Jersey and Connecticut). The sample is a multistage,
stratified, cluster of the three groups "that yielded an 87% net coverage rate of
the 1980 Mexican-origin population living in the five southwestern states, about

90% of the Puerto Rican-origin population in the defined universe for the New

61
York City area, and about 96% of the Cuban-origin population living in Dade

County, Florida" (Trevino, Moyer, Valdez, and Stroup-Benham 1991).
Limitations

The data was collected in 1980, fifteen years ago. We know little about
how this data compares with today's conditions and this may result in possible
bias. The measure of acculturation is unsophisticated (based on language and
self identity). Also, acculturation is only measured for Mexicans and not for
Cubans and Puerto Ricans. A precise measure of low birthweight is not
offered. The birthweight data is retrospective based on mother’s recollection. It
does not use medical or birth records. A more accurate measure of birthweight
would increase the validity of any findings. Also, gestation is based on the
recollection of mothers. It is measured by mothers' response to a question
which read, "was the child born earlier than expected, when expected, or later
than expected."

The data do not reflect nutrient intake during pregnancy. The Hispanic
Health and Nutrition Examination Survey, 1982—84, only gives data on nutrient
intake in general for the Latino groups. Although the data does not specifically
address nutritional intake during pregnancy, if these needs are not being met in
general, there is little reason to believe they would be met during pregnancy.

Data in the Hispanic Health and Nutrition Examination Survey, 1982-84
(HHANES), does not measure nutrition based on Recommended Daily

Allowances (RDA). This makes it difficult to measure nutrient intake using

62

RDAs as it is used in most literature regarding nutrient intake recommendations
for pregnant women. For example, Eckstein (1980) and Hinton and Kerwin
(1981) suggest that the caloric RDA for pregnant women (19-22 years of age) is
2,400 calories a day. However, this can not be directly tested using the
HHANES because the HHANES uses as its measure of nutrient intake 100
grams of the edible portion of a food.
mm

The Hispanic Health and Nutrition Examination Survey, 1982-84,
(HHANES) is a national sample. Romero Gwynn and Gwynn (199324) write
that "the best source of data on the current nutritional status and food practices
of Hispanics in the United States is the national Hispanic Health and Nutrition
Examination Survey (HHANES)." The characteristic of nationalism will
decrease problems associated with regionalism. The data also accounts for
heterogeneity. It is one of the few data sets that allows the researcher to
conduct systematic cross group comparisons using the same data set The
seven nutrients identified as being most important in effecting positive
pregnancy outcomes are found in the HHANES. The hypotheses are
realistically testable given the data in the HHANES.

W

in a perfect wortd, a scholar‘s data set will enable them to answer all of

their research questions. We do not live In a perfect world and the Hispanic

Health and Nutrition Examination Survey, 1982-84, (HHANES) fell short of

63

allowing me to answer some of my questions as well as i had wanted. The
biggest problem is in combining components. Data for nutrient intake was
drawn from the Dietary Practices, Food Frequency, and Total Nutrient intake
component of the HHANES. Data for low birthweight was drawn from the Child
History Questionnaire component of the HHANES.

The Child History Questionnaire component does not identify adults in a
low birthweight household by gender. Therefore, i had no way of directly
testing the nutrient intake of Latinas in households with low birthweight babies.
The Total Nutrient Intake component did identify adults by gender. I combined
the two sets and matched the Latina cases in the Total Nutrient Intake
component with low birthweight birth cases. By proxy, I inferred that these
Latinas were the mothers of the low birthweight Infants. To best achieve this, i
selected Latinas over 15 years of age. To cross check this, i examined the
relationship of this group is to head of the house. I found that 72.1% of the
group were either the heads of house or married to heads of house (the
remainder of the group were either children of the head of house or
grandchildren of the head of house). Given this, I can reasonably say that at
least 70% cf the Latinas used for measuring nutritional intake in relation to low
birthweight could be mothers of low birthweight infants.

Th e le
In the Hispanic Health and Nutrition Examination Survey, 1982-84

(HHANES), respondents self identify as either "Other, MexicanIMexicano,

64

Mexican American, Chicano, Puerto Rican, Boricuan, Cuban, Cuban American,
Hispano, Latin American/Spanish, Spanish American, or Spanish/Spain." l
eliminated those who self identify as "Other, Hispano, Latin American/Spanish,
Spanish American," and "Spanish/Spain." i did this because Mexicans,
Cubans, and Puerto Ricans are the largest three Latino subgroups in the United
States. Marin and Marin (1991210) write that the term "Other Hispanics is a
label of convenience that designates a wide range of individuals who do not
trace their family background to Mexico, Cuba, or Puerto Rico . . . 'Other
Hispanics’ refers to a very heterogenous group encompassing professional,
white-collar employees, and blue-collar workers who left their countries in
search of better education or employment or in reaction to political changes."
Additionally, "when individuals self-identify as group members . . . they may be
said to have a high ethnic identity" (Phinney 1995258). Vln'lat i am suggesting is
that those who identified as "Other, Hispano, Latin AmericanISpanish, Spanish
American, Spanish/Spain" are not of Mexican, Cuban, or Puerto Rican descent,
which is the focus of my study.

Those who identified as Mexican/Mexicano, Mexican American and
Chicano l collapsed into the group "Mexican.“ Those who identified as Puerto
Rican or Boricuan l collapsed into the group "Puerto Rican." Those who
identified as Cuban or Cuban American I collapsed into the group "Cuban."

The issue of Latino identity is one that I recognize poses some problems

for researchers. It is one that is not easily solved given that Latinos are such a

65

heterogeneous and diverse group. For example, someone who is Mexican may
self-identify as "Spanish" even though they descend from Mexico. As such, my
sample and not perfect and this study is limited by using ethnic self
identification as a method of construting the sample. However, with the data
set, it seemed the most workable and i am comfortable with this approach.

There were 3,472 children in the Child History Questionnaire component
of the HHANES. Of these children, 68.2% were Mexican, 7.8% were Cuban,
and 24.0% were Puerto Rican.

There were 6,548 Latinos over 15 years of age in the Dietary Practices,
Food Frequency, and Total Nutrient intakes component of the HHANES. of
these, 43.3% were male and 56.6% were female. Sixty percent were Mexican,
15.2% were Cuban, and 24.4% were Puerto Rican.

tati 'cal Pr ures

I need to clarify that there were two low birthweight issues that l
addressed. First, I looked at total low birthweight among the Latino population.
i did this to give an overall picture of subgroup differences. The second issue
addressed involved low birthweight in relation to nutrient intake of Latinas.

In my comparison of subgroup differences in total low birthweight, l
separated out preterm and fullterm outcomes of low birthweight. Social and
economic characteristics of households with low birthweight children were then
computed. This was done by Latino subgroup and by preterm and fullterm birth

outcomes.

66

Data for the nutrient intake sample was then computed. This data
consists of the social and economic characteristics of households of the
sample. Crosstabs were then conducted on gestation length by ethnic group
and ethnic group by birthweight category. Chi-square tests of significance were
concluded on the crosstabs. As a test of H1 and subhypotheses 1 through 7, I
concluded an analysis of variance (ANOVA) test on Latinas by Nutrient intake
for Birthweight Category.

For the ANOVA, I created a variable labeled "Latinas." This variable was
composed of the three different Latina subgroups. Therefore, the variable
Latina, was composed of three different measures, Mexicans, Cubans, and
Puerto Ricans. The ANOVA was on Latinas by nutrient intake. A Bonferroni
test of significance was conducted to assess where, if any, between group
differences occurred.

The Bonferroni test of significance is a multiple comparison test. This
test adjusts observed significance levels based on the number of comparisons
being made (NorusislSPSS Inc. 19942185). it makes this adjustment because
the more observations that are made, the more likely that pairs will be
statistically different even it means are equal. NorusislSPSS inc. (19942185)
explain it this way, "if you are making 5 comparisons, the observed significance
level for the original comparison must be less than 0.05/5, or 0.01, for the
difference to be significant at the 0.05 level."

in the ANOVA, the Bonferroni test told me which subgroups were

67

statistically significant in the difference between the means. Put simply,
although an ANOVA may tell me that a significant F exists for my three
measure variable Latina and a certain nutrient, it does not tell me where the
differences are. The Bonferroni test did this for me.

i then tested H2. I used logistic regressions to see what nutrients best
explain the incidence of low birthweight for Latinas by birthweight category. I
used logistic regressions so that my research model was consistent with my
research question. My research question was whether or not nutrition effected
low birthweight outcomes. The best way to test this was to dichotomize the
dependent variable of low birthweight based on birthweight categories and
conduct regression analyses. Birthweight categories were selected based on
the Rumbaut and Weeks (1993) study which was discussed in the literature
review. So, for the first category, birthweights of less than 500 grams were
coded a 1 and all other birthweights coded a 0. For the second category, 500-
1,499 grams, birthweights from coo-1,499 grams were coded a 1 and all other
birthweights coded a 0. For the third category, birthweights from 1,500-2,499
grams were coded a 1 and all other birthweights coded a 0. For the fourth
category, birthweights of over 2,499 were coded a 1 and all other birthweights
coded a 0. However, in the sample, only one birth was less than 500 grams.
Therefore, although i set out to study the effect of nutrition on low birthweight
outcomes, l included births over 2,500 grams to give more opportunity for

comparisons.

68
As a test of the "Other" variables and their effect on low birthweight,

logistic regression analyses were conducted. Again, birthweight categories

were dichotomized as above as the dependent variable and education and
language served as the independent variables. in the Hispanic Health and
Nutrition Examination Survey, 1982-84, (HHANES) education was measured by
"highest grade or year of regular school head of family has ever attended" from
kindergarten through graduate school. Language was measured by Language
preference for completing the interview questions of the HHANES. English was
scored a 1 and Spanish a 2. The language measure was receded as an

indicator variable (i.e., English is coded a 0 and Spanish a 1) for the logistic

regression equation.

Lpgistip Regrpsg’on Models

The first three logistic regressions address the role of nutrition in low
birthweight. The next three regressions examine the role of education on low

birthweight. The last three regressions evaluate the role of language on low

birthweight.
1. Birthweight of 500-1,499 grams (coded as 1, all others coded as 0) =

(3,, + BCalories + B,Protein + B,Vitamin A + B,Vitamin C + B,Calcium +

B,Phosphorous + 87Folacin + B.Iron + e.

2. Birthweight of 1,500-2,499 grams (coded as 1, all others coded as 0)

= (3,, + B,Calories + 8, Protein + 8,Vitamin A + 84Vitamin C + B,Calcium +

B.Phosphorous + (37Folacin + B.Iron + e.

69
3. Birthweight of >2,500 grams (coded as 1, all others coded as 0) =

60+ 6, Calories + [32Protein + (33Vitamin A + 84Vitamin C + 85Calcium +
B,Phosphorous + B,Folacin + Balron + e
4. Birthweight of SOD-1,499 grams (coded as 1, all others coded as 0) =

80+ B,Education + e.

5. Birthweight of 1,500-2,499 grams (coded as 1, all others coded as 0)

= 6,, + B,Education + e.

6. Birthweight of >2,500 grams (coded as 1, all others coded as 0) =

(3,, + B,Education + e.

7. Birthweight of 500.1,499 grams (coded as 1, all others coded as 0) =
(30+ B,Language + e.

8. Birthweight of 1,500-2,499 grams (coded as 1, all others coded as 0)
= [30+ B,Language + e.

9. Birthweight of >2,500 grams (coded as 1, all others coded as 0) =
Bu+ B,Language + 9.
Relation of the Mogel to ELM

I expected Puerto Ricans in the very, very low birthweight (WLBVV)
category to have poor nutrition, low education, and speak Spanish. i also
expected Puerto Ricans in the very low birthweight (VLBW) category to have
poor nutrition, low education, and speak Spanish. I expected Mexicans and
Cubans in the high low birthweight (HLBVV) and normal birthweight (NBW)

categories to have good nutrition, high education, and speak English.

 

 

70

Furthermore, i expected that there would be more Puerto Ricans in the preterm
low birthweight categories than Mexicans and Cubans, and i expected that
there would be more Mexicans and Cubans in the fullterm low birthweight
category. Table 3 presents these expectations.

Table 3. Expectations of Relationships Between independent
and Dependent Variables

Dependent Variable WLBW VLBW HLBW NBW
independent Variables
Nutrition
Mexican Good Good Good Good
Cuban Good Good Good Good
Puerto Ricans Poor Poor Good Good
Education
Mexican High High High High
Cuban High High High High
Puerto Ricans Low Low High High
Language
Mexican Engl. Engl. Engl. Engl.
Cuban Engl. Engl. Engl. Engl.
Puerto Ricans Span. Span. Engl. Engl.

Qallingagty‘

Calci
Cale
Fol
iron
Phos
Prot
Vit A

VitC

"p = < .05. "*p < .01.

71

Table 4. Correlation Matrix of independent Variables

Calci Cals Fol iron Phos Prot Vit A Vit C
1.0

-.51**" 1.0

.66“ -.47""* 1.0

.59"""r -.54*"* 86"“ 1.0

.95""“' -.56*“ .61” 69"“ 1.0

-.04" .44“ .04 .07“ .08” 1.0

-.O4* .44“ -.02 .00 -.08"* .47‘“ 1.0

76”" -.41""" .80‘“ 70”" 74”" .04 .00 1.0

"""p < .001.

Table 4 indicated that high correlations existed between seva of the

independent variables. Many of these correlations were statistically significant.

i checked for colinearity following a method suggested by Dr. Broman.

Dr. Broman stated that variables can be collinear without there being a

problem. However, problems may arise if coefficients are effected by this

collinearlty. To check to see if coefficients were affected, l re-estimated a

series of regression equations that removed one variable at a time. i removed

variables that were strongly correlated with other variables. l removed the

following variables in the following order: calories, phosphorus, calcium, folacin,

vitamin C, and iron. l removed these variables in this order because calories

 

72

was correlated with seven other variables and the rest consisted with five other
variables (all at the .30 level or above).

Using this method, it was found that iron and calcium were collinear.
This is because when calcium was removed from the regression equation, the
coefficient for iron changed dramatically. Voght (19932144) writes that when
colinearity exists, "it makes it difficult if not impossible to determine their [the
collinear variables] separate effects on the dependent variable." This means
that if iron or calcium prove to be significant in my logistic regressions, l have to
take into consideration their collinear relationship in my interpretation of the

results. Chapter 6 now turns to the results.

 

CHAPTER 8

RESULTS

This chapter begins with a descriptive discussion of the sample relating
to low birthweight outcomes. i then present the social and economic
characteristics of the sample. The results of the tests of the hypotheses are
then presented.

Low Birthweight, Gestation Length, and Racial and Ethnic identity

In the literature review, it was seen that subgroup differences exist
among Latinos regarding low birthweight. The question here is "to what degree
did subgroup differences exist among the sample for my study?"

Of the total births, for all terms, one hundred seventy-two (5.2%) of the
Latino children were born low birthweight. Bifurcating for subgroup, 4.4% of the
Mexican children were low birthweight, 5.5% of the Cuban, and 7.7% of the
Puerto Rican. There were four African American Latino children born low
birthweight. Of the African American low birthweight children, one was of
Mexican ethnicity and three of Puerto Rican ethnicity.

Of the low birthweight Latino children, 60.9% were born preterm and
39.0% were born fullterm. Two of the African American children were born
preterm and two were born fullterm. Of the African American children born
preterm, both were of Puerto Rican ethnicity.

For Mexicans, 2.1% of the total births were preterm low birthweight births

73

 

 

74
and 2.2% fullterm low birthweight births. For Cubans, 4.3% of the total births

were preterm low birthweight births and 1.8% fullterm low birthweight births.
For Puerto Ricans, 5.0% of the total births were preterm low birthweight births
and 1.01% were fullterm low birthweight births. Table 5 presents a crosstabs of

low birthweights that bifurcates by ethnic group and gestation length.

Table 5. Crosstabs of Birthweight and Gestation Length by Ethnic Group

 

 

 

 

 

 

 

 

 

 

 

 

M l— T—T
Ethnic Group Preterm Births Fullterm Births Total
Mexican 47 48 95
(45.8%) (71.5%) (55.1%)
Puerto Rican 45 24 61
| (42.9%) (35.7%) (35.5%)
Cuban 13 5 16
(12.4%) (7.5%) (9.2%)
Total 105 67 172
X2 = 12.7

X2 at .01 where df = 2, is 9.21, Significant, p < .01.
X’ at .05 where df = 2, is 5.99, Significant, p < .05.

The crosstabs indicated that the majority of low birthweight births were
preterm births. A Chi-square test found significant differences between the
observed frequencies and expected frequencies at the .01 and .05 significance
levels.

'ai n E o i hara eristi of w B‘ i ht ouse oi
In Chapter 2, i argued that low birthweight can not be examined in

isolation. Additionally, the hypothesis of this dissertation states that the

:I

‘t:

 

75
outcome of low birthweight is a function of nutrition which is an indirect function

social and economic conditions. Therefore, the social and economic
characteristics of the sample from the Hispanic Health and Nutrition
Examination Survey, 1982-84, were calculated to assess if the results are
consistent with data discussed in Appendix H which found that Puerto Ricans
tend to have less income and less education than Mexicans and Cubans.

Table 6. Social and Economic Characteristics of Households
with Low Birthweight Children

 

 

m m
% % % % % Med. %
Aff. Fem. Marr. Frgn. Emp. Fem. 4
Amer. Head Born Inc.“ Yrs.
Hse. Coll.
Year

 

3.0 15.9 81.7 44.9 80.7 13-14 .9 '
0.0 21.4 92.9 92.8 57.1 25-30 7.1
12.9 58.1 40.3 1.5' 30.6 10-11 1.6

 

 

 

 

 

 

 

 

 

 

 

 

 

3935. Data is for reported head of house.
a. 69.4% born in Puerto Rico and are not considered foreign born.
b. Median income is annual for combined family and is in thousands.

There are more Puerto Rican African Americans than Mexican and
Cuban African Americans. More than half of the Puerto Rican households are
headed by women. Less than half of the Puerto Rican heads of households
are married. The majority of Puerto Ricans are unemployed and Puerto Ricans

have the lowest income. Mexicans are undereducated in terms of years of

college.

 

76

Table 7. Social and Economic Characteristics of Households

with Low Birthweight Children Born Preterm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m”: ”~—
% % % % % Med. %
White Aff. Fem. Marr. Frgn. Emp. Fem. 4
Amer. Head Born inc." Yrs.
Hse. Coll
Max. 81.1 3.8 9.4 88.7 34.0 83.0 17-18 0.0
lCub. 100.0 0.0 27.3 90.9 90.9 54.5 25-30 0.0
P.R. 88.1 9.5 57.1 38.1 2.4" 28.6 9.5- 0.0
10.5
Note. Data is for reported head of house.

a. 71.4% born in Puerto Rico and are not considered foreign born.
b. Median income is annual for combined family and is in thousands.

More than half of the Puerto Rican households are headed by women.

Less than half of the Puerto Rican heads of household are married. The

majority of Puerto Ricans are unemployed and Puerto Ricans have the lowest

income.

Low Birthweight Children Born Fullterm

Table 8. Social and Economic Characteristics of Households with

 

 

 

 

 

 

 

 

 

 

 

 

 

% % % % % % Med. %
White Aff. Fem. Marr. Frgn. Emp. Farm. 4
Amer. l-lead Born Inc." Yrs.
l-lse. Coll.
96.2 1.9 21.2 76.9 57.7 78.8 12.5- 3.0
13.5
66.7 0.0 0.0 100.0 100.0 66.7 30-35 0.0
61.5 7.7 61.5 46.2 0.0" 30.8 10-11 0.0

 

 

 

Table 8 (cont'd).

77

Note. Data is for reported head of house.
a. 53.8% born In Puerto Rico and are not considered Foreign born.
b. Median income is annual for combined family and is in thousands.

More than half of the Puerto Rican households are headed by women.

Less than half of the Puerto Rican heads of households are married. The

majority of Puerto Ricans are unemployed and Puerto Ricans have the lowest

income. Mexicans have the most years of college education.

Tests of Hypptheses

Table 9. Crosstabs of Ethnic Group by Birthweight Category

 

 

 

 

 

 

 

 

 

 

 

# I
Ethnic sou-1,499' 1,500-2,499 >2,500 Total
Group
Mexican 4 55 1 ,021 1 ,080
(57.0%) (57.3%) (66.4%) (66.0%)
Cuban 1 4 1 18 123
- (4.2%) (7.7%) (7.4%)
Puerto Rican 2 37 396 435
(28.6%) (38.4%) (25.8%) (26.5%)
Total 7 96 1 ,535 1 ,638
Npte.
a. Only one birth was less than 500 grams and this birth was to a Mexican
woman.
X2 = 8.06

X2 at .001 where df = 4, is 18.46, Not Significant, p > .001.

X2 at .01 where df = 4, is 13.28, Not Significant, p > .01.
X2 at .05 where df = 4, is 9.49, Not Significant, p > .05.

 

 

78

The crosstabs indicated that for all three subgroups, the majority of the
births were not low birthweight (i.e., > 2,500 grams). Of the births under 2,500
grams, the majority of the births occurred in the higher range (1 ,500-2,499
grams). However, a Chi-square test found no significant difference between the
observed frequencies and the expected frequencies. The means of birthweight

category by subgroup are presented in Table 10 below.

Table 10. Means of Birthweight Categories by Ethnic Group

 

    

 

 

 

 

IF— -—_l-—fl
Ethnic Group coo-1,499 1,500-2,499 >2,500
Mexican 1,150 2,223 3,459
Cuban 1,162‘ 2,352 3,436
Puerto Rican 1,431 2,324 3,338

 

 

 

 

 

Npte.
a. This figure is for the one Cuban in this category.

I now turn to the results of the statistical tests of the hypotheses. I begin
with the results of the analysis of variance tests (ANOVA). The logistic
regression analyses results follow for nutrition and low birthweight and

education and language on low birthweight.

79

Table 11. One-way Analysis of Variance of Latinas by Nutrient intake for

Birthweight Category

Means
Nym Mapper; Cgban' Epgrto Ripan F-Ratiob

Very Low Birthweight SOD-1,499 Grams
Calories 7.5 NA. 1.5 7.91*°
(small calories)
Protein 5.5 NA. 2.0 1.84
(grams)

Table 10 (cont'd).

Vitamin A 425.0 N.A. 500.0 2.82
(retinol

equivalents)

Vitamin C 125.6 NA 66.8 1.13
(milligrams)

Calcium 604.5 N.A. 530.6 0.48
(milligrams)

Phosphorus 983.6 N.A. 991.1 0.17
(milligrams)

Folacin 207.5 NA. 92.9 3.11
(micrograms)

iron 11.0 NA. 8.78 1.54
(milligrams)

High Low Birthweight 1,500-2,499 Grams

Calories 8.4 2.6 3.0 0.06
(small calories)

Protein 3.4 4.0 2.8 0.49
(grams)

Table 11 (cont'd).

Vitamin A
(retinol
equivalents)

Vitamin C
(milligrams)

Phosphorus
(milligrams)

Folacin
(micrograms)

Iron
(milligrams)

Calories
(small calories)

Protein
(grams)

Vitamin A
(retinol
equivalents)

wmmmc
(milligrams)

Calcium
(milligrams)

Phosphorus

(milligrams)

Folacin
(micrograms)

Iron
(milligrams)

320.7

332.6

489.1

911.4

56.0

Normal Birthweight >2,500 Grams

2.6

2.6

274.6

223.7

461.7

636.6

247.3

142.4

80

500.1

385.1

426.9

764.5

32.5

2.7

2.6

258.6

395.5

295.9

440.8

959.3

50.1

305.6

259.0

372.8

761.0

40.78

3.7

3.8

356.2

230.4

191.0

358.0

166.0

155.2

0.46

0.40

0.89

0.16

0.59

13.1""

16.6"”

1 1.2“"

0.13

1.71

1.43

0.26

0.13

81
Table 11 (cont'd).

Nata-

a. Only one Cuban in this category.

b. The Host is a test of statistically significant differences between the means
for Mexicans, Cubans, and Puerto Ricans for the nutrient intake variables.

c. A Bonferroni test indicated a significant difference between Mexicans and
Puerto Ricans. There was no significant difference between Mexicans and
Cubans or Cubans and Puerto Ricans.

d. A Bonferroni test indicated significant differences between Mexicans and
Puerto Ricans and Cubans and Puerto Ricans. There was no significant
difference between Mexicans and Cubans.

e. A Bonferroni test indicated significant differences between Mexicans and
Puerto Ricans and Cubans and Puerto Ricans. There was no significant
difference between Mexicans and Cubans.

f. A Bonferroni test indicated significant differences between Mexicans and
Puerto Ricans and Cubans and Puerto Ricans. There was no significant
difference between Mexicans and Cubans.

"p < .05, ”p_< .001

A significant difference exists between the means among Latinas for the
intake of calories at for 500-1,499 grams level. This difference is between
Mexicans and Puerto Ricans with Mexicans taking in more calories than Puerto
Ricans.

At the above 2,500 grams level, significant differences between the
means among Latinas for the intake of calories, protein, and vitamin A exist.
For calories, Puerto Ricans are taking in more calories than Mexicans and
Cubans. For protein, Puerto Ricans are taking in more protein than Mexicans
and Cubans. For vitamin A, Puerto Ricans are taking in more vitamin A than

Mexicans and Cubans.

82

Table 12. Logistic Regression Coefficients for Independent Variables on
Birthweight Category for Latinas

Manama Mpxican Qupgn‘
Very Low Birthweight 500-1,499 Grams

Puerto Rican

Calories 1.68 NA -1.31
Protein .53 NA. -1.64
Vitamin A 0.02 NA. -3.02
Vitamin C 9.86 NA 4.37
Calcium -5.25 NA. -7.34
Phosphorous -3.57 NA. 12.3
Folacin -36.6 NA. -14.5
iron 44.5 NA. -94.3
-2 log likelihood 42.5 NA 11.0
Chi-square 10.0 NA. 14.4
DP. 8 NA 8
Number of cases 1080 NA 435
Education -.08 NA. .22
-2 log likelihood 52.2 NA. 24.3
Chi-Square .57 NA 1 .06
OF. 1 NA. 1
Number of cases 1080 NA 435
Language -8.02 NA. .75"
-2 log likelihood 49.5 NA. 248.2
Chi-square 3.26 NA. 4.70
OF. 1 NA. 1
Number of cases 1080 NA 435

High Low Birthweight 1,500-2,499 Grams

Calories -.00 -.73 -.12
Protein .32 .30 -.29
Vitamin A .21 1.80 -.04
Vitamin C 1.13 6.09 226
Calcium 1.17 3.63 -2.32
Phosphorus -2.09 3.12 .88
Folacin -9.1 1 -2.21 .77

Table 12 (cont'd).

iron

-2 log likelihood
Chi-square

D.F.

Number of cases

Education

-2 log likelihood
Chi-square

D.F.

Number of cases

Language

-2 log likelihood
Chi-square

D.F.

Number of cases

Calories
Protein
wmmmA
wmmmc
Calcium
Phosphorous
Folacin

iron

-2 log likelihood
Chi-square

D.F.

Number of cases

Education

-2 log likelihood
Chi-square

4.7

415.0
12.9

1080
.01

434.5
.187

1080
-.22

433.7
.522
1
1080

-.12
-.33
-.17
-2.38
-1.19
2.08
9.35
-4.8*

436.7
14.1
8
1080

-.01

457.6
.044

83

21.5
6.68

123
-.04

35.3
.100

123
.53

35.0
.272
1
123

.60
.33
~1.12
-4.42
7.31
-3.10
1.66
3.0

30.1
5.10
8
123

-0.01

41.7
.010

9.5

246.6
5.9

8

435

-.06

251.1
1.91
1

435

—8.34

23.5
2.04
1
435

Normal Birthweight >2,500 Grams

.19
.34
.13
2.46
2.43
-.62
-.47
-7.1

254.3
7.55
8

435

.045

261.1
1.28

Table 12 (cont'd).

D.F.
Number of cases

Language
-2 log likelihood
Chi-square

D.F.
Number of cases

Lima-

1 080

.32

456.1
1.17
1
1080

84

1
123

-.95

40.6
1.08
1
123

1
435

-.61

258.8
3.42

1

435

a. Cubans were so few that the regressions did not converge.

"p < .01.

The only significant logistic regression coefficient was iron for Mexicans

at the above 2,500 grams birthweight category. There were no significant

logistic regression coefficients for education variable for any of the three

subgroups at any birthweight category level. Language was the only significant

logistic regression coefficient and this was for Puerto Ricans at the SOD-1,499

grams birthweight category. The next chapter, Chapter 7, now interprets the

l'OSUltS.

CHAPTER 7

INTERPRETATION OF RESULTS AND CONCLUSIONS

This chapter interprets the results of Chapter 6. it begins with a
discussion of the Objectives in the context of the results. Next, I discuss the
relationship between gestation length, Latinos, and low birthweight. This is
followed with a discussion of low birthweight and social and economic
characteristics. The results are then addressed in relation to the hypotheses. I
then discuss the theory I proposed and how this compares with the results of the
study. lfoilow this with a section on the relationship between societal transition
and Latino food habits. I finish this chapter with discussions on the relevance of
the study, the importance of culture in low birthweight research, and
recommendations for further research.
109.091.950.122

My Objectives were to demonstrate between group differences among
Latinos regarding low birthweight, develop an improved measure of cross-
cultural comparison, and to test hypotheses based on correlates of low
birthweight. it was demonstrated that between group differences exist for my
sample. Puerto Ricans have the highest rate of low birthweight followed by
Cubans and then Mexicans. These findings are consistent with those cited in
the literature review (Becerra et al. 1991; Health United States 1990 1992;

Mendoza at. al 1993; US. Census Bureau 1992; CDC 1993).

85

86

Regarding Objective Two, i can not definitively say that my measure of
cross-cultural comparison is an improvement over others previously used. i
would rather think of my measure as an enhancement to measures of cross-
cultural comparison that currently exist. Although i did find nutritional intake
differences among certain nutrients, to say that this completely measures culture
would be in error.

What i have offered in this dissertation is another way to look at culture,
one that has not been used before with this population in such a systematic
manner. My measure is an alternative to others that can be incorporated into
comparisons of culture. Given this, I feel confident saying that a modification of
Objective Two was accomplished. My measure may not be an improvement,
but my measure may improve the way we think about comparing culture.
However, i do think that research that uses nutrient intake as a measure of
culture, should use Recommended Daily Allowances (RDA) as the measure of
nutrient intake. This will allow for a broader range of comparisons since most
studies use RDA as a measure of nutrient intake.

Like Objective Two, Objective Three was accomplished, in part. Did i
develop hypotheses related to research on low birthweight and did test these
hypotheses? i would have to say "yes." Did these hypotheses best explain
differential low birthweight among Latinos? I would have to say "no" because in
the logistic regressions, only iron was found to be significant for Mexicans, and,

this is difficult to assess given the collinear relationship iron had with calcium.

87
Additionally, this was significant only for birthweight above 2,500 grams. in

other words, the effect of iron was on births that were not low birthweight. This
means that iron may effect the incidence of low birthweight rather than having
an effect on low birthweight outcomes (which was my primary research
question).

stjgtion Length, Latinos, and Low Birthweight

In the literature review, I cited Michielutte et al. (1992) who indicated that
the difference between preterm and term low birthweight should be distinguished
when conducting research on low birthweight. This sentiment was also stated
by Dr. Paneth (1995). As such, I dichotomized low birthweight results for Latino
subgroups by gestation length.

Mat i found supported the above. if Puerto Rican women can go
fullterm, they have better results for low birthweight than for Mexicans and
Cubans. If Puerto Rican women go fullterm, they improve low birthweight
outcomes by almost three times. Gestation length is a significant factor in low
birthweight as was indicated by the results of the Chi-square. My study has
reinforced the argument that gestation length has to be controlled for in
conducting low birthweight research.

Puerto Ricans have many more preterm infants followed by Cubans and
then Mexicans. Cubans are much closer to Puerto Ricans in preterm outcomes.
For fullterm low birthweight, Puerto Ricans have the best outcomes followed by

Cubans and then Mexicans. In controlling for gestation, outcomes for low

88
birthweight become asymmetric. What this tells me, is that infant mortality

related to low birthweight is more of an issue of premature birth than an issue of
low birthweight.
ow irthwei ht n ocial and conomic ha cteristics

I believe it is the lower socioeconomic status of Puerto Ricans that
accounts for their higher incidence of preterm low birthweight. in comparing the
differences between household types, it was found that Puerto Ricans generally
tend to be worse off in terms of social and economic characteristics. Puerto
Ricans have less stable environments, meaning that they tend to be single
households headed by women. Additionally, Puerto Ricans have lower rates of
employment and lower incomes. These results are consistent with the findings
of the data in Appendix G.

Puerto Rican households with preterm low birthweight babies fare slightly
worse than Puerto Rican households with fullterm low birthweight babies. More
fullterm households have a head of household that is married, have a higher
rate of employment, and a slightly higher median family income. However,
preterm households tend to have a lower percentage of female headed
households.

Given the results on low birthweight by ethnic group and gestation length,
it seems that the social and economic factors of a stable household have an
impact on the preterm outcome of low birthweight for Puerto Ricans. Puerto

Ricans have the highest percentage of unstable households and the highest

89
percentage of low birthweight preterm infants.

The relationship between socioeconomic status and low birthweight
outcomes is of importance because some researchers have espoused the
"paradox" of positive low birthweight outcomes in the face of poor social and
economic conditions (Rumbaut and Weeks 1993). This paradox may be true,
but it is only true for Mexicans, not Puerto Ricans. This demonstrates the need
to be careful when stating that "Latinos" have positive low birthweight outcomes.
They do, but only some do. Furthermore, type of low birthweight needs to be
considered (i.e., preterm versus fullterm). These issues were made very clear to
me in doing this dissertation.

W02

Hypothesis one proposed that the nutrient intake of Puerto Rican women
would be less than the nutrient intake of Mexican and Cuban women. The
hypothesis one was partially supported. The intake of calories for Puerto Ricans
was significantly less than intake of calories for Mexicans in the 500-1,499
grams category (support for subhypothesis one). However, in the above 2,500
grams category, Puerto Ricans had significantly higher intakes of calories,
protein, and vitamin A than Mexicans and Cubans. This contradicted my
expectations.

What I suggest is occurring, is that the Puerto Rican women who are in
the 500-1,499 gram category and have a low intake of calories, are the ones

having the preterm low birthweight births. However, given the limitations of the

90
Hispanic Health and Examination Survey, 1982-84, (HHANES) as critiqued In

Chapter 5, there is no way to directly test this. This makes it even clearer the
limitations of the HHANES and raises the question of whether it is time to
conduct a "new and improv " HHANES Il. i think it is.

My findings suggest that subgroup differences in dietary practices exist.
These practices do not seem to be related to low birthweight outcomes.
However, these practices may be related to the incidence of low birthweight for
Puerto Ricans. i say this because Puerto Rican women in the above 2,500
gram category have significantly higher intakes of calories, protein, and
vitamin A.

Hymtheg‘g Two

Hypothesis two preposed that nutritional intake would effect low
birthweight outcomes more for Puerto Rican women than Mexican and Cuban
women. None of the logistic regression coefficients in the low birthweight
categories were statistically significant. The hypothesis was not supported.

Additionally, the crosstabs of ethnic group by birthweight category was not
significant. There was no difference between the observed frequencies and
expected frequencies. The null hypothesis could not be rejected. There was no
statistically significant difference between Latinos in birthweight category. This
is interpreted as meaning that nutrition does not have an effect on low

birthweight outcomes, let alone more of an effect for Puerto Ricans.

91

Th E u t'on a Lan us a on w Birthwei h

The only regression coefficient that was statistically significant was
language for Puerto Ricans in the 500.1 ,499 grams category. This means that
Puerto Rican women in this category preferred using Spanish more than
English. Using language as measure of acculturation as did Scribner and Dwyer
(1989), this would suggest that Puerto Ricans are the least acculturated.

if Puerto Ricans are the least Angloized, then this result is contrary to
research that suggests that poor health outcomes are a consequence of
acculturation (Scribner and Dwyer 1989; Balcazar 1993; Borges et al. 1993;
Rumbaut and Weeks 1993). This finding is particularly relevant to the Scribner
and Dwyer (1989) study.

Scribner and Dwyer (1989) found that a higher acculturation index (using
language as one of their measures) resulted in higher rates of low birthweight.
The contrary findings may be due to the fact that language was just one
measure in the Scribner and Dwyer (1989) study. Scribner and Dwyer (1989)
also used nativity and ethnic identification as measures of acculturation. What I
believe may be happening in my study, is that the Puerto Ricans in the sample
who preferred Spanish are the same ones who are taking in less calories.
Incidence of i ei h a

It can not be said that the theory was supported for based on the
research model, there was little statistically significant support for hypotheses

one or two. However, the finding of differential incidence of low birthweight

92

among the Latino subgroups deserves discussion.

Puerto Ricans have the lowest socioeconomic status. Contributing factors
to the lower socioeconomic status of Puerto Ricans are a high rate of female
heads of households, low income, and unemployment. And, as has been seen,
Puerto Ricans are highest in preterm low birthweight. These factors cannot be
taken in isolation. I see it as no mere coincidence that Puerto Ricans have a
high rate of preterm low birthweight. The interdependent nature of correlates
that effect the incidence of low birthweight is now discussed in the context of the
theory that was proposed.

Regarding Puerto Ricans, the roots of the functional relationship as
pr0posed in the theory begins with the historical social and economic oppression
that Puerto Ricans have experienced. These conditions evolved into poor social
and economic circumstances. it is this lack of economic resources that may
contribute to the high incidence of Puerto Rican children born preterm low
birthweight.

Puerto Ricans are mired in poverty. This translates into poor living
conditions and little hope of social advancement. The situation of Puerto Ricans
is similar to African Americans. African Americans have a high degree of single
female headed households and many African American children live in poverty
(Duncan et al. 1994; Huston et al. 1994). African Americans also have a high
incidence of low birthweight and infant mortality (Health United States 1990

1992; US. Bureau of the Census 1992; From the MMWR 1993).

93

The Puerto Rican situation is due in large part to the patriarchal and
dependent relationship Puerto Rico has with the US. In Appendix E, l
discussed the history and consequences of this history on Puerto Ricans.
Puerto Ricans have a "second class citizenship" and have been relegated to the
lowest echelons of society. African Americans have had a similar experience.
Where African Americans were enslaved in chains, Puerto Ricans were
psychologically enslaved through the loss of their independence and the
subsequent reliance on the US. for more subsistence. Until this relationship
changes, Puerto Ricans will continue to be stymied by the oppressive specter of
poverty.

Like Puerto Ricans, Cubans also had a high incidence of preterm low
birthweight. It is my speculation that these Cubans are recent immigrants. Not
all Cubans came in the first "wave." As reflected in the study of Oropesa and
Landale (1995), the socioeconomic status of second generation Cubans ("native-
bom children with at least one foreign-bom parent") falls significantly from the
first generation. I suggest that it is these Cubans who are the ones experiencing
poorer social and economic conditions. These are not the "Golden Wavers" of
the early sixties. These women are the maids, nannies, and seamstresses of
the Cuban p0pulation. i call them the "Lumpencuban." They are the minority of
the Cuban population (as evidenced by the socioeconomic data), but they exist
nonetheless. i would venture that as the status of "glamour refugee" continues

to decline for Cubans, the number of Cuban women in this situation will

94

increase.

Mexicans pose an unique case. They have almost the same number of
preterm low birthweight infants as fullterm low birthweight infants. Given the low
level of Mexican preterm low birthweight infants, i am led to believe that
Mexican women are doing something "right" in terms of issues related to
pregnancy and childbearing. These issues may be part and parcel of the overall
Mexican experience.

I believe that the low incidence of preterm low birthweight for Mexicans is
a result of a sense of stability that is a manifestation of their history in the United
States. This history is related to the geographical proximity of Mexico to the
United States. This sense of stability and history has been critical in the
development of the Mexican life-world. it is this life-world that accounts for the
low incidence of preterm low birthweight.

e' 'e-Worl an w ouTst i?

I draw upon Habermas' (1976) concept of life-worlds in my discussion of
acculturation. In explaining between group differences in low birthweight
outcomes among Latinos, I propose that the life-world of Mexicans is more
durable than that of Cubans and Puerto Ricans. i believe this because of the
historical and geographical experience of Mexicans. it is in this Mexican life-
world that a sense of stability and continuity exists. This continuity is conducive
to a low incidence of preterm low birthweight.

The best way I can define what I believe to be the Mexican life-world is

95

by contextualizing it in history and geography. In discussing the differential
historical circumstances of the three Latino subgroups (Appendix E), l stated that
because much of the United States was once Mexico and because of the
contiguous border, a sense of history and tradition is associated with the land.
Psychologically, for many Mexicans, a border does not exist. This sense gives
Mexicans a deeper feeling of security and this makes it easier to survive in a
hostile environment. It is this historical, social psychological phenomenon that
results in the Mexican life-world. This is not to say that Mexicans have not
suffered at the hands of oppression and racism, rather, because of geographical
and historical conditions, Mexicans were in a position to better adapt.

Neither Cubans or Puerto Ricans have this relationship with the land that
do Mexicans. Cubans and Puerto Ricans who immigrate to the United States
immigrate to a land that was never once their country. I believe that this
connection with the land and history has been instrumental in developing the
Mexican life-world.

One major problem. How does one measure this life-world? This
question is a dissertation in itself and an area that I may pursue in future
research. At present, it seems like an unsurmountable task to try and measure
the psychology of an entire people, people who are quite varied by region,
socioeconomic status, acculturation, etc. However, I am convinced it is out

there.

A na 9 estion

in looking at the differential outcomes of low birthweight among Latinos, it
is reasonable that one may ask, "if preterm low birthweight is being controlled
for and Mexicans have a higher incidence of fullterm low birthweight, doesn‘t
that mean Mexicans have negative pregnancy outcomes?" I want to address
this question because to do otherwise would be a glaring oversight as a
researcher.

Yes, this does mean that Mexicans have a higher incidence of fullterm
low birthweight than Cubans and Puerto Ricans which can be interpreted as a
negative outcome. However, it is more likely that an infant will die if it is born
preterm low birthweight than fullterm low birthweight. More than 70% of the fetal
and neonatal deaths in the US. occur among preterm Infants (Adams
19952739). This is why i interpret Mexican low birthweight outcomes as better
than Cubans and Puerto Ricans. Furthermore, the Mexican fullterm low
birthweight outcome is still low when compared to the low birthweight outcome
for Non-Latino Whites.

Why do Mexicans have a higher rate of fullterm low birthweight? It may
be that Mexicans are more similar to Non-Latino Whites than the acculturation
argument that was made in Chapter 3 would have suggested. Although I
argued that the Mexican life-world may be conducive to positive low birthweight
outcomes, there may be a portion of the Mexican population that has

incorporated the comparatively unhealthy habits of Non-Latino Whites (i.e., have

97
become Angloized).

This is consistent with the research by Marks et al. (1990) which
compared the degree of a balanced diet among Mexicans, Cubans, and Puerto
Ricans. Mexican American women had the second best scores on the balanced
diet component. It could be that the women who contributed to the middle range
scores on this measure (i.e., the ones who were not eating a balanced diet) are
the ones having the fullterm low birthweight infants.

The results for Mexicans are meaningful for they can serve to guide
health researchers in examining low birthweight. A more intensive study of what
exactly these women are doing, or not doing, seems to be in order. Early on in
this dissertation, I stated that this study was valuable because the knowledge
gained about Latinos could be beneficial for the society as a whole. I believe
that the findings on Mexicans opens the door for this learning.
mate) Transition and Latinp Fpog Hapits

One of my criticisms of the Hispanic Health and Nutrition Examination
Survey, 1982~84, (HHANES), is that it is dated. This poses the question of
relevancy to 1996. In other words, it has to be considered if the dietary patterns
of Latinos is the same now as it was in 1982-84.

i would argue that society has been in transition since the early 1980's in
relation to work patterns. An uncertain economic climate and higher costs of
living are requiring more families to have dual incomes and more women to

work. For example, in 1980, 51.5% of the females aged 16 years and over were

98

in the civilian labor force. in 1990, this figure increased to 67.0%. It is projected
that by the year 2000, 69.3% will be in the civilian labor force (US. Bureau of
the Census 19922381). The trend similar for Latinas with 47.4% in the civilian
labor force in 1980, 53.0% in 1990, and 69.3% projected for the year 2000 (US.
Bureau of the Census 19922381).

With more families requiring both parents to work and with many single
mothers working, the practice of having formal, sit~down "family meals" may be
impractical. Preparing a meal takes time and effort, a luxury that many families
do not have. This may lead to more families eating at restaurants and
consuming more "fast foods" and not eating home prepared meals.

The effect of more families eating restaurant and fast foods has led to a
proliferation of these type of establishments. As an example, I was listening to a
report on National Public Radio last week and heard that 3,000 new
"McDonald's" restaurants open every day. Additionally, it seems that when ever
a new "McDonald's" is opened, soon after a new "Burger King" or "Taco Bell"
are built on the same street block, if not next door.

Furthermore, with the number of fast food "ethnic" restaurants like "Taco
Bell" increasing at an phenomenal rate, it has to be wondered what this means
to the whole concept of eating "traditional foods." Has traditional ethnic food
been bastardized in modern society? In other words, do people think they are
really eating "Mexican food" when they eat at "Taco Bell?" Do Mexicans feel

they are eating traditional Mexican food when they are eating at "Taco Bell?"

99
i think that Non-Latino Whites could interpret "Taco Bell" as "real"

Mexican food for they just have not had the experience of authentic traditional
Mexican food. I further think that Mexicans who are more Angloized could
interpret "Taco Bell" as traditional. i say this about Mexicans in the context of
the argument that was made in this dissertation that food is a cultural indicator
and the more one is removed from their culture of origin, the less they would
practice tractional food habits.

Given that eating traditional Latino foods is healthy (Romero Gwynn and
Gwynn 1993;1994), if the "Taco Bells" of the world are seen as traditional, then
there may be some people who feel that eating at "Taco Bell" is healthy. This
can be problematic for I can not see eating fast food as being healthy. This is
something that researchers who sepouse the virtues of eating traditional foods
need to be aware of and clarity in their work "Traditional" does not mean "Taco
Ball" or "Chi-Chis." "Chi-Chi‘s is more formal restaurant than "Taco Bell," but it
is an "Americanized" version of Mexican food. In fact, they even seem to invent
foods under the guise of being "Mexican" that are not recognized as traditional
in Mexican culture (based on the work of Romero Gwynn and Gwynn 1994). As
an example of this in my own life, growing up i never even heard of something
called a "Chimmi-Chonga" (one of Chi-Chi's specialties).

The Latino population is increasing in the United States. This increase
means that there will be more Latinos with disposable incomes. This is a trend

that has not gone unrecognized by the fast food industry and their advertising

100

departments. In fact, I was recently watching "Univlsion" (a Spanish station
broadcast on cable television) and a commercial for "McDonald's" was aired that
portrayed a Latino family in which both parents came home from work and took
the family to "McDonald's."

My point here is that due to the societal transition that has occurred since
the early 1980's, the dietary practices of Latinos have changed. if this is the
case, then any impact that culture has on dietary practices would be diluted. it
has been suggested by Romero Gwynn and Gwynn (1993: 1994) that the diet of
Latinos is healthy and that the more acculturated Latinos become, the more
unhealthy that diet becomes. If a diet is unhealthy then there can be
consequences for health outcomes (e.g., obesity, high blood pressure, heart
disease, etc).

Clearly, this is a situation that health practitioners should be aware of and
monitor, especially those that work with Latinos. if a health care practitioner
who works with Latinos should observe a health change in one of their patients
for the worse, they would be wise to inquire about any changes in diet. This
inquiry should pay particular attention to any changes as related to a move away
from traditional foods to more Angloized foods.

Given that society Is in transition and Latinos in the United States are
Included in this transition, perhaps there is a better way to use food habits as a
cultural indicator than focusing on nutritional intake. One method that I suggest

is to examine the role of food in celebrations and rituals. I believe that food

101

takes on very unique meanings in the context of rituals and celebrations. For
example, the symbolic water in Catholic masses. This approach is consistent
with the symbolic interactionist (Cooley 1962; Mead 1969) and Habennasian
(1976) approaches with its emphasis on the importance of symbols in the
definition of reality and social environment.

Furthermore, this approach is consistent with the research of James
(1993) who noted the importance of symbols in cultural identity and Magana and
Clark (1995) who discussed the role of the Lady of Guadalupe as a defining
symbol for Mexicans. The foods of "enchiladas," "females," "menudo," and
"mole" may not have the same importance for Mexicans as the Lady of
Guadalupe, but I would argue that as symbols, they have a special role in
defining cultural identity.

To examine this, a research study could be designed that specifically
examined the role of food in rituals and celebrations. This study would consist
of interviews of Latino families which asked questions about types and
frequencies of traditional foods in rituals and celebrations. A ethnic identity type
of instrument could also be administered. The results of the interviews could be
compared with the results of the ethnic identity instrument to assess whether
those who used traditional foods in celebrations and rituals had a stronger sense

of ethnic identification. I believe that they would.

102

Releva t d

i now address the issue of where the study has taken the field and where
the study may lead the field. Although the hypotheses were not supported,
support for the concept of Latino heterogeneity was found. This was found in
comparisons of history, social and economic conditions, and differences in
birthweight. It seems clear that Latino heterogeneity must be considered in
doing research on Latinos and this study has promoted this approach.

The study also points out that the Puerto Rican situation Is in dire need of
attention. Puerto Rican infants are at great risk of dying due to the high rate of
preterm low birthweight. The rate is alarming. Some proactive measures need
to be taken to address this situation. i believe that improving the social and
economic situation of Puerto Ricans would help alleviate the situation. This task
falls on the government that contributed to their current conditions, the United
States. In the long run, resources that are committed to improving the plight of
Puerto Ricans will decrease the costs associated with health problems related to
preterm low birthweight.

in my interpretation of the low birthweight outcomes for Mexicans, I
suggested that Mexican culture may be beneficial for positive pregnancy
outcomes. This is something that health care professionals really should take
note of. More needs to be learned about this culture in terms attitudes toward
pregnancy and childbirth. Mexican culture is a recourse waiting to be tapped.

Doing so would benefit the population as a whole.

103

The question still remains, "what causes low birthweight?" I do not have
an answer. I can say that among Latinos, in this study, nutritional differences do
not mm to effect low birthweight outcomes. This little bit of knowledge can be
added to the store of information on low birthweight research. I have eliminated
just one explanation. This topic and research on this t0pic has by no means
reached Its pinnacle. As mentioned previously, studies can be conducted that
investigate symbols as related to food habits and studies can be conducted on
attitudes and behaviors of Mexican women during pregnancy. However, i do
believe that the Hispanic Health and Nutrition Examination, 1982-84, (HHANES)
has reached its usefulness as a data source.

Along with the methodological issues that were addressed regarding the
HHANES, because of the transitional nature of society, dietary practices may
have changed since the HHANES was conducted. Furthermore, the total Latino
population has increased from 1980 to 1990 (US. Bureau of the Census 1992).
An increase in numbers may have the effect of changing majority/minority
dynamics in some geographical regions. The reader may wonder, "if this is the
case, why did i use the HHANES in the first place?" i used the HHANES
because I discovered these problems and considered these issues as part of
doing this research. Perhaps by addressing them in this dissertation and
commenting on my experience with the limitations and shortfalls of the

HHANES, i may spare some researcher some frustration in the future.

104
The [mppmce pf leture in Low Birthweight Researpp

I now answer the question of "to what degree is culture important in low
birthweight outcomes?" In Chapter 2, the biomedical risk factors of low
birthweight were discussed. To be sure, the physiological can not be removed
from the sociological. As I have argued throughout this dissertation, an
interdependent relationship may exist between variables that effect low
birthweight. However, it seems that the role of culture in effecting low
birthweight has not been given its full due in the scientific community.

Although no statistical support was found to support this notion, I do
believe that culture is just as important as biomedical factors in considering low
birthweight outcomes. i believe this because of the important role that culture
has (and the role symbols play In culture) in effecting how people interact in their
environment (Cooley 1962; Mead 1969; Haben'nas 1976). Knowing the
biomedical effects upon low birthweight is only half the story. To fully
understand the phenomenon of low birthweight, culture should be considered. if
only the biomedical aspects of a group's behavior are addressed, then
researchers are only focusing on the machinery of being human, not what
makes humans human.

For me, at this point in my research, I see culture as the avenue to
emlore as an explanation for differential low birthweight outcomes. l was just
not able to demonstrate the impact of culture in this particular study. However, it

has given me a path to follow for further research.

105
W

No research problem can be completely explained in one short work. I
think that good research often ends with more questions to pursue. One
question that i was left with is "what are Mexican women doing that results in
positive low birthweight outcomes?" The key word here is "doing." I do not
believe that this can be discovered using purely quantitative methods. However,
one way to get at what Mexican women are doing regarding pregnancy and
pregnancy outcomes, is to ask them. This would mean a more qualitative
approach.

The everyday behaviors of pregnantMexican women should be studied. 1
think that in depth case studies would be the means to accomplish this type of
study. If I had the means, i would design an ethnographic study to assess how
pregnant Mexican women behave and what they believe during pregnancy. I
would have interviewers talk to a sample of pregnant Mexican women on a daily
basis. They would be asked about what they did or did not do in terms of their
daily activities. Furthermore, questions would be asked about the role of
religion, the family, values, and beliefs in pregnancy. The special emphasis of
this study would be on the role symbols play for these women and their families
as related to family, children, and childbirth. It is my hope that soon In my

career, i will be able to pursue this type of research.

106

sio

This study began with the argument that nutrition, as a cultural indicator,
is the best explanation of low birthweight among Latino subgroups. i can not
say that food habits are the best predictor of low birthweight outcomes, but I can
say that this study offers another way to look at culture that has not been
previously considered in such a systematic manner. Differences were found in
nutrient intake among Latino subgroups which can be seen as a cultural
indicator. i believe that my study has given the scholarly community another
tool in which to consider culture.

What the research especially points out, is that a complicated issue like
low birthweight can not be examined In isolation. It appears to be an
interdependent phenomenon. This lends support for interdisciplinary approaches
to research. Vlfith so much information out there in so many different fields and
subfields, it is almost impossible for one researcher alone to completely explain
phenomenon. I foresee a future of more "team" type approaches to research. I
know that this study would have been lacking In its conceptual development

without the input received from those in other fields.

APPENDICES

1 07
APPENDIX A: KEY CONCEPTS

Acculturation - A process of learning of a new culture while evaluating the old
culture. In evaluating the old culture, some things are kept and some are
discarded (Garcia Coll and Meyer 1993).

African Americans - African Americans living In the mainland United States.
Cubans - Cuban Americans living In the mainland United States.

Cubanos - Same as Cubans.

Cubanas - Cuban women.

Culture - A complete way of life.

Infant Deaths/Infant Mortality - Deaths occurring within the first year of life.
Infant Mortality Rate - Infant deaths per 1,000 live births in a population.
Latinas - Cuban, Mexican, and Puerto Rican women.

Latinos - Cubans, Mexicans, and Puerto Ricans.

Low Birthweight - Weight of less than 2,500 grams (5 pounds, 5 ounces) at
birth.

Mexicans - Mexican Americans living in the mainland United States.
Mexicanas - Mexican women.

Non-Latino Whites - Those outside the groups of Cuban, Mexican, and Puerto
Rican, except for African Americans. Has commonly been known as "White."

Nutrition - Level of nutrient intake.
Nutritional Practices - Food habits.
Prenatal - Previous to birth.

Preterm/Premature - Births occurring prior to 37 weeks gestation.

108

Puerto Ricans - Puerto Ricans living In the United States mainland.
Puerto Ricanas - Puerto Rican women.

Sociocultural Context - Nationality, occupational, attitudinal, personal Identity,
food, activity preference, and social group preference characteristics (Garcia
Coll and Meyer 1993211).

Socioeconimc Status - Social and economic characteristics. Comprised of the
variables of education, income, occupation, and having a single parent
household status (US. Bureau of the Census 1992; Oropesa and Landale
1995; Rumbaut 1995). Also used as measures of social and economic
characteristics are marital status, being foreign born, and having a female head
of household.

109

APPENDIX B: SIGNIFICANT RISK FACTORS FOR LOW BIRTHWEIGHT
PRETERM AND FULL TERM INFANTS (MICHIELUTTE ET AL. 1993)

Age
(16-17 years old)
(15 or younger)

Education
(9 years or less)

Race (African American)
Smoking

No Previous Live Birth
Under 5 feet tall

Weight under 100 lbs.
<5 ft. 8‘ > 100 lbs.

>5 ft. & < 100 lbs.

<5 ft. & <1OO lbs.

Previous premature
LBW birth

Two or more 1st
trimester abortions

One or more 2nd
trimester abortions

Last birth < 1 year
ago

Uterine anomaly]
DES Exposure

All LBW AOR
1.58
1.58

2.10

2.10
2.16
1.79
1.54
2.45
N.A.
N.A.

N.A.

3.78

1.42

1.80

1.63

2.80

P-LBW AOR

1.43
2.34

1.96

1.96

1.72

1.61

N.A.

2.41

N.A.

N.A.

N.A.

4.14

1.58

1.94

1.78

3.81

T-LBW AOR
NS.
0.38

2.24

2.24
2.61
1.94
N.A.
N.A.
2.53
2.53

2.12

2.41

N.A.

N.A.

N.A.

N.A.

110

APPENDIX C: SUMMARY OF SELECTED LOW BIRTHWEIGHT

(LBW) STUDIES

Authofls) Data Results Conggsioglgritigue

Brooks-Gum Literature Possible factors in Access to prenatal care

at al. 1988 review LBW are and matemel education
demographics, are needed to reduce
medical risks the Incidence of LBW.

during conception
and pregnancy,
behavioral risks,
environment,
physical factors,
lack of social
support, work
patterns, health
habits,and a lack

of prenatal

care.
Hogue et NIMS 97.6% of all infant
al. 1987 deaths for cohort

were LBW infants.

Kramer Meta- Weight gain is
1987 analysis associated with
of French lUGR.
and English

publications.

A hospital in Harlem was
used as an example of
attempts made to
accomplish the above.
However, no data was
given on the success of
the attempts.

Factors identified as
increasing the risk of
Infant mortality were male
gender, short gestation,
for or third (or later) birth
order for LBW Infants,
younger or older mothers
less maternal education,
and lack of prenatal care
received the first trimester.
National sample.

Women who are under-
nourished are more at risk
for lMR and LBW infants.
lntamational sample.

Michielutte
et al. 1992

Rumbaut and
Weeks 1993

Taffel
1 990

Physician
Records

San Diego
Compre-
hensive
Perinatal
Program

US. Dept.
Health &
Human
Services

111

Differences exist
between P-LBW
and T-LBW

Previous live
birth was the
best predictor
of "Infant" health
outcomes.

Women who
gained less than
21 pounds
during pregnancy
were 2.1 times

Age, education, race,
smoking, weight, previous
premature/LBW birth and
and uterine anomaly/DES
are most associated with
LBW. The main
differences between P-
LBW and T-LBW are in
the age, weight, and
previous premature/LBW
and uterine anomaly/DES
exposure factors. Study
did not address chemical
use or Latinos. Study was
limited to North Carolina.

LBW was one measure of
outcomes. Study limited
to San Diego. Latinos not

subgrouped.

Proper nourishment Is
essential for expectant
mothers. Lack of proper
nourishment can result in
negative outcomes.

more likely to have National sample.

an LBW infant and
1.5 times more

likely to experience

Infant mortality

112

APPENDIX D: SUMMARY OF STUDIES ON LATINOS AND LOW

Authgr(s)

Becerra
at al.
1 991

Borges
et al.
1 993

CDC
1993

Data

 

*1: ._.__NLW M. Q.

Linked Birth 4.6 4.0
and Infant

Deaths Data

Set

NAS - -

Monthly 6.1 5.6
Vital

Statistics

Report

4.1 4.0
3.12 -
(Mexico)
5.5 5.7

EB.

6.6

9.0

BIRTHWEIGHT (LBW), PREGNANCY, AND CHILDBIRTH

LBW Results 66 of Births LB\_M Qonclusignl
Qn’tigue

Results contrary to
SES expectations.
Latinos are a
heterogeneous
group. Puerto
Ricans have a
higher Incidence of
LBW. National
sample.

Heavy drinking
during pregnancy
increases the risk
of LBW and
preterm delivery.
Study was a survey
study that used
mother's
recollection of
LBW. Mexican
National sample.

Statistical
document. No
interpretation or
conclusions
discussed.

m

Dowling
and Fisher

Fenster
and
Coye 1990

Gaviria
et al.
1 982

Hayes-
Bautista

113

Data LBW Results

Hospital -
Birth
Records

Hospital 1.8 -
Birth
Records

Interviews - -
of 89 Pregnant
Women in

Chicago

County 5.32 -
Govt.
Statistics

.L. MM.

angusionl
Qritigue

Results contrary to
SES expectations.
May be due to
"protective
sociocultural effect"
and selective
immigration. Study
was limited to
Chicago. Study did
not subgroup
Lafinos.

Agricultural wok
did not increase
incidence of LBW.
May be due to
selective migration.
Study did not
subgroup Latino
sample. Study was
limited to two
agricultural counties
in Caiifomia.

Recent arrivals to
US. and Mexican
American women
sought prenatal
care later than did
long term arrivals.
Study limited to
Chicago area.

Article did not
subgroup Latinos.
Data limited to
Los Angeles
County.

114

MM W

L
Health Nat. Ctr. -
U. S. 1990, Health
1992 Statistics

Mendoza Nat. Vital 6.2
at al. Statistics
1991

Rumbaut 8. Linked Birth/ 5.2
Weeks Infant Death
1989 Records

Rumbaut 8. San Diego 4.1

Weeks Comprehen- (U.S. born)

1 993 Siva
Perinatal 1 .9

M M $2
5.7 5.6 5.7
to to

6.0 6.0

5.6 5.7 5.9
5.1 - -

Program (foreign born)

9.3

Conclusion!
Qg'tigue

Results contrary to
SES expectations.
Latinos are a
heterogeneous
group. Calls for a
more research to
explain differential
LBW. National
sample.

Results contrary
to SES
expectations.
Latinos are a
heterogeneous
group. Calls for
more research to
explain differential
LBW. National
sample.

Results contrary

to SES
expectations.
Latinos not
subgrouped. Study
limited to San
Diego area.

Latinos not
subgrouped.
Study limited to
San Diego area.

Author(s)

Scribner
1989

US.
Census
Bureau
1992

Ventura
8. Martin
1 991

(in
Mendoza
1994)

Weinman
8. Smith
1994

Data LBW as its

HHANES - -

US. 6.2 5.7
Census
Data

Dept. 6.1 5.7
Health

& Human

Services

Medical - -
Records

of

Adolescent

Lafinas

Presenting

for Delivery

5.6

IO

5.8

4.8

BR

9.0

7.9

Qon_c_lusionl
Qritigge

As acculturation
increases,
incidence of LBW
increases.
Measures of
acculturation do not
include nutrition
variables. National

sample.

Statistical
document. No
interpretations or
conclusions
discussed.
National sample.

Mexicans not
discussed. Puerto
Ricans have a
higher incidence of
LBW than Cubans.
National sample.

U.S. born and
Mexican born
Latinas both did not
follow-up with
postpartum visits.
Suggests groups
are similar and rely
on alternative,
culturally related
care regarding
pregnancy and
childbearing. Study
limited to Houston.

*L=Latino; NLW=Non-Latino White; M=Mexican; C=Cuban; PR=Puerto Rican

117

APPENDIX E: HISTORICAL AND GEOGRAPHIC CIRCUMSTANCES OF
MEXICANS, CUBANS, AND PUERTO RICANS

Hernandez (1994) discusses five stages of Anglo-American policy that
had ramifications for Latinos in the US. These stages are "(1) exclusive
occupation of conquered lands, (2) Internal colonization as a conquered
minority, (3) restricted citizenship through statehood, (4) external colonization
by political dependence, [and] (5) subordination in a world economic system"
(Hernandez 1994:18). Prior to when the first Anglo immigrants arrived In the
1600s, the Latino presence had already been a part of the New World for about
one hundred years.

Spain had colonies in what is present day Texas, California, Colorado,
Utah, Nevada, New Mexico, Arizona (Identified as the southwest and
Caiifomia), Mexico, Louisiana, Florida, Guam, the Philippines, Cuba, Puerto
Rico, the Dominion Republic, and Central and South America. Table 13 below
shows that a considerable number of Latinos were living in areas that were

later to be conquered by the US.

Table 13. Population Estimates of New Spain Colonies Prior to 1776

 

 

 

 

Geographic Area Estimated Populatlon

Southwestern US. 8. Caiifomia 5,200,000 1
Cuba 160,000 I
Puerto Rico 50,000 I

 

 

Source: Hernandez 1994:18

118

in Stage 1 (1776-1834), exclusive occupation, the US. sought to
increase Its wealth by expanding its agricultural base in the south. This was
accomplished by land purchases from Spain in 1819 which prompted the
takeover of Florida. Anglos then began to move Into the entire southeastem
region of the US. This influx of Anglos resulted in many Latinos returning to
Spain or migrating to Cuba (Hernandez 1994:19).

lntemal colonization (1835-1859), Stage 2, came soon after Mexico won
its independence from Spain in 1821. The new country encouraged northern
migration Into its territories. However, the need for Anglos to have more land
precipitated Texas' ”independence” and the War with Mexico. The end of the
War In 1848 resulted in Mexico losing its northern territories (the southwest and
Caiifomia).

Almost overnight, Mexicans were relegated to a lower class status. The
Influx of Anglos Into these territories due primarily to the gold rushes of 1849
(Caiifomia) and 1858 (Colorado) reduced Mexicans to a numerical minority. By
1858, in Texas and Caiifomia, Latinos made up little more than 20% of the
population (Hernandez 1994:20). Due to carrupt legal decisions,
disproportionate numbers, and racism, many Latinos lost property, and
ranching, mining, and farming rights. Not only had Latinos become the
numerical minority, they had become an ethnic minority with a subordinate
status.

Between 1860 and 1897, Stage 3 occurred. During this stage of

119

restricted citizenship, many of the lands that were Spain's and Mexico's became
states of the US. Statehood increased political control and coercion that
further isolated and distanced Latinos. Agribusiness in California called for
many laborers to perform uniform tasks during certain seasons. This created a
large pool of menial laborers who were mired in poverty and low status.

Hernandez writes that "the stigma of defeat combined economic poverty
with cultural alienation" (1994221). Being Isolated economically and politically,
Latinos maintained a culture distinct to themselves. Latinos may have been
citizens, but they were not allowed to be “Americans" and fully participate In all
of the benefits of citizenry.

After the Spanish-American War In 1898, the US. gained control of
Puerto Rico and occupied Cuba until 1902. During this fourth stage (1898-
1956) of external colonization, the US. sought to increase its power
internationally. The islands of Puerto Rico and Cuba were seen as places to
expand economic markets, particularly the crop of sugar cane. Additionally, the
location of the islands were seen as militarily important, places for naval ports
and a first line of defense In the case of a southern Atlantic sea attack.

Acquiring Puerto Rico as a territory and Cuba as a protectorate
increased the Latino population of the US. The Latino population of Puerto
Rico was 1 million and the Latino population of Cuba 1.7 million In 1898
(Hernandez 1994222). Large US. companies went into the islands and

proceeded to monopolize the sugar cane crops. Latino Inhabitants of the

120
Islands were expected to provide cheap labor while becoming Anglozied and

”Americanization was imposed to acculturate the external colonies to their
functions as outposts of the United States" (Hernandez 1994:22). However, as
with Latinos on the mainland, cultural traditions and values "defined their
[Puerto Ricans and Cubans] Identity as distinct from the Anglo culture they
learned in school" (Hernandez 1994223).

During this stage, the number of Latinos increased In the US. in 1910,
Latinos comprised 2 percent of the total US. population. Most of these Latinos
lived in the geographic areas that was New Spain. However, groups began to
migrate at this time. For example, In 1910, Puerto Ricans and Cubans formed
their first communities In New York City. Most came to work in cigar factories
and "garment sweatshops" (Hernandez 1994223). immigration also Increased
from Mexico. Between 1910 and 1930, approximately 1 million Mexicans came
and settled In Caiifomia or the north central states (Hernandez 1994224).

Most of the immigrants came for economic reasons. The attitude of the
Non-Latino White dominate culture toward Immigrants was one of acceptance
of the immigrant as a worker, but non-acceptance as a person. When they
were needed, they were welcome. When they did not serve the needs of the
economy, they were deported. For example, the Los Braceros program began
during World War II and was “designed to supply labor for US. agnculturalists,
underwrote Mexicans‘ travel costs, insured a minimum wage, and guaranteed

jobs and equitable treatment” (Alicea 1994238). The program continued until

121

1964. However, during the depression many Mexicans were "repatriated" by
force to Mexico (many were US. citizens). From 1924-30 "more than 400,000
Mexicans were forced to leave the country" (Alicea 1994:38).

Puerto Ricans came In large numbers "when American political and
economic influences displaced plantation workers In Puerto Rico and
encouraged one third of the island's workers to move to New York City and
other northern cities during the 19505" (Hemandez 1994225). Due to racism,
most immigrants were only offered unskilled labor occupations and had to live
in segregated, dangerous, deteriorated neighborhoods.

Cubans who came after the revolution in 1959 were, by comparison,
more fortunate than Mexicans or Puerto Ricans. Since the revolution 800,000
Cubans have immigrated to the US. (Alicea 1994250). The majority of these
Immigrants settled in the Miami area. The strategy of the US. was to claim a
political victory over Cuba by offering capitalist America as a haven from
communism. in 1960, The Cuban Refugee Program helped emigres find jobs,
provided financial assistance, health care, and funds for relocation outside of
Miami (Alicea 1994252).

Stage 5, economic subordination began in 1960 and continues. in
economic subordination, Latinos play the role of the exploited. They are
overrepresented In low paying, low skill occupations with little room for
advancement. Their status In the dominate society Is to fill a need In

occupations that are undesirable. They fill this need due to economics. Recent

122

Cuban immigrants are finding themselves in the same situation as Mexicans
and Puerto Ricans. The atmosphere of the Cold War has warmed and their
status as a symbol has diminished. Thus, they are finding themselves as
laborers and service people.

Although many Latinos have dispersed throughout the US,
"demographic data indicate that Hispanics remained in their historical places
and in central city neighborhoods with limited economic potential, segregated
from Anglo and African Americans, regardless of US. region" (Hernandez
1994227). The Latino population of the southwest and Caiifomia Is dominated
by Mexicans, Florida by Cubans, and the New York City area by Puerto Ricans.
These geographical concentrations can be traced to historical trends. As i
addressed in the above discussion, these trends are related to Issues of
conquest and occupation, economic exploitation, and immigration. Table 14

below Indicates this dispersion.

Table 14. Latinos by Region (Selected Cities)

       

 

 

 

 

Region % Latino Largest Latino Group :

RepresentedIClty (ill. of Latinos that are
of that group)

Southwest! 34.5 Mexican

Albuquerque 53.0

Southwest! 23.0 Mexican

Denver 70.0

Southwest! 69.0 Mexican

El Paso 96.0

 

 

 

1 2 3
Table 14 (cont'd).

 

    
   
  
   
   

 

 

 

 

Caiifomia] 20.0 Mexican
Watsonville 86.0
Caiifomia! 39.9 Mexican

Los Angeles 76.0

North Central States! 19.6 Mexican
Chicago 93.0

New York City! 20.1 Puerto Rican
Brooklyn 83.0

Florida! 62.5 Cuban
Miami 59.0

 

 

 

 

Source: COSSMHO 1994 and 1990 US. Census data.

The contemporary Cuban socioeconomic status may be attributed to the
governmental assistance offered during the early 19603 (see Appendix H). This
allowed first "wave” Cubans to become established and be in a position to
better assist subsequent Cuban immigrants. Mexicans, although a "conquered"
people, have the distinguishing characteristic of having much of what is the
US. ones being Mexico. This allows for a sense of cohesion and uniformity,
especially since Mexico is contiguous with the US. A deep sense of history
and tradition may be associated with the land and psychologically, borders are
nonexistent. This may make it easier for Mexicans to survive In a hostile
environment (in other words, they are not really ”alien"). Lastly, Puerto Ricans
have the unfortunate status of being paternally dependent on the US. When
the island was industrialized, many were forced to become dependent on the

US. and forced into low status occupations or public assistance.

123

APPENDIX F: MARIN AND MARIN'S (1991) DISCUSSION OF CULTURAL
SIMILARITIES AMONG LATINOS

Marin and Marin (1991) argue that Issues associated with allocentrism,
simpatia, familialism, power distance, personal space, time orientation, and
gender roles, reflect values that are common to all Latinos.

According to Marin and Marin (1991) allocentrism is a form of
collectivism in which the needs of the "in group" of are primary concern. Marin
and Marin define allocentic societies as emphasizing "the needs, objectives,
and points of view of an ingroup while individualistic cultures determine their
social behavior primarily in terms of personal objectives, attitudes, and values
that resemble little If at all those of the ingroup" (1991211). Yep (1995) defines
allocentrisrn as ”a cultural trait associated with the preference for interpersonal
relationships in Ingroups that are nurturing, empathetic, loving, intimate,
respectful, and willing to sacrifice for the welfare of the group" (Yep 19952201).
Allocentric societies can be thought of being similar to Durkheim's (1984)
concept of a society that operates on the basis of mechanical solidarity. A
mechanical society moves and operates as one. its primary allegiance is to
societal needs and not individual needs.

Marin and Marin (1991) suggest that simpatia is related to allocentrism.
Simpatia "emphasizes the need for behaviors that promote smooth and
pleasant social relationships" (Marin and Marin 1991:12). Empathy, dignity, and

respect for others are behaviors characteristic of simpafia. Simpafia seeks to

124

reduce conflict and allow for harmonious Interactions.

Familialism has been identified by Marin and Marin (1991) as one of the
strongest values shared by Latinos. Baca-Zinn (19942167) defines familialism
as "an assortment of beliefs and behaviors associated with family solidarity and
the extended family." Marin and Marin define familialism as an "Individuals
strong identification with and attachment to their nuclear and extended families,
and strong feelings of loyalty, reciprocity, and solidarity among members of the
same family" (1991213). Three "value orientations" are associated with
familialism. These orientations Involve giving emotional and financial support to
extended family, relying on relatives for support, and viewing the family as a
reference for attitudes and behavior (Marin and Marin 1991213).

The concept of "fictive kin" is also an element of familialism. Fictive kin
are friends of the family who assume a relational status. This status as a family
member Is Imbued due to the close relationship between the parents and the
friend or the friends Involvement In the raising of children. These fictive kin are
referred to as oompadres (Marin and Marin 1991214).

Power distance "supports the notion that societies have powerful
Individuals as a result of inherent traits (e.g., intelligence) or of Inherited or
acquired characteristics (e.g., money, education)" (Marin and Marin 1991214).
This Is similar to Weber‘s (1946) concept of traditional authority. in traditional
authority, the authority figure rules due to some type of "divine right." Power is

ascribed and not achieved. in power distance societies, those In power seek to

125

maintain the status quo and the society tends to "support these power
differentials" (Marin and Marin 1991215). in these societies, obedience,
conformity, and deference to autocratic authority Is valued. Marin and Marin
(1991) suggest that the notion of "power distance" Is common to Latino culture.

Regarding physical proximity to others, Marin and Marin argue that
Latinos "have been shown to prefer shorter distances than non-Hispanic
Whites. Hispanics, like other 'contact cultures,‘ feel comfortable when
physically close to others and are less likely to feel that their personal space
has been invaded when a stranger comes close to them" (1991215). This value
is of importance because It may cause for awkwardness In situations where
Latinos are Interacting with Non-Latino Whites. The Non-Latino White may feel
infringed upon and the Latino may feel insulted.

A society that is present time oriented Is "often described as unable to
delay gratification or to plan for the future and as inefficient and not punctual"
(Marin and Marin 1991216). A present time orientation is one in which the
concept of time is malleable, it is an orientation that lacks structure and
consistency. Marin and Marin (1991) suggest that researchers have shown that
Latinos are present time oriented and are more flexible in their attitude toward
time. Like physical proximity, this value has potential for problems when
interacting with the Non-Latino White population which tends to be present time
oriented. For example, late arrival at meetings and social events may be

perceived as negligence. However, given the values of simpafia, the quality of

126

the time spent may be of more importance to Latinos.

Lastly, regarding gender roles, Marin and Marin suggest that much has
been written about the Latino male "being strong, In control, and the providers
for their families ('machismo')" (1991216). Conversely, Latino women "are
described as submissive and lacking in power and influence" (Marin and Marin
1991216). However, the issue of gender roles is in flux given the ever changing
social and economic circumstances of Latino families. Some researchers are
convinced that machismo exists and others view it as a cultural stereotype
(Cases, Wagenheim, Banchero, and Mendoza-Romero 1995:235-6).

Marin and Mann's (1991) discussion concerning "power distance"
suggests that Latino society Is highly stratified. Almost every man Is
subordinated to an other. if a man lives in a world in which he is constantly at
the boot heels of an other, the home may be the only place where he may
occupy the highest status. Machismo keeps women subordinated and is a
microcosm of the larger society.

Although Marin and Marin's (1991) discussion on Latino values appears
to over generalize, it provides a basis in which to begin to compare differences
among Latino groups. Issues involving the interactive nature of gender roles,

family, and culture are paramount to a discussion of acculturation.

127

APPENDIX G: SOCIAL AND ECONOMIC CHARACTERISTICS OF
LATINO SUBGROUPS AND THE ACCULTURATION PROCESS

Differences exist among the Latinos subgroups for social and economic
variables. The higher level of socioeconomic status for Cuban immigrant women
is demonstrated in Table 15 below. Social and economic variables may Impact
gender roles and the family. This impact may be related to acculturation.

Table 15. Social and Economic Characteristics of Cuban and Mexican Women
Who immigrated to the US. Between 1960 and 1970

Ethnic
Group

 

 

Cuban 22.2 5.4 0.1 16.8 35.08
Mexican 7.4 0.8 17.3 9.4 22.36

 

 

 

 

 

 

 

Source: Pedraza-Bailey 1985:89-112

Ortiz (1994) using 1980 US. Census data, also shows a differential
socioeconomic status between Cuban, Mexican, and Puerto Rican women
(Puerto Ricans will be discussed later in this subsection). Although the data for
employment is only reflective of professional occupations, it is useful for it

bifurcates these measures of socioeconomic status by ethnic group.

128

Table 16. Educational Attainment and Those in Professional Occupations:
Cuban, Mexican, and Puerto Rican Women

 

     
   

 

 

 

 

Ethnic Group 96 High School 96 College 96 Professional
- Graduate Graduate Occupation
; Cuban 53.3 13.2 15.7
Mexican 36.3 3.7 10.8
Puerto Rican 39.1 4.8 13.4

 

 

 

 

 

Source: Ortiz 1994:26-30

Tables 15 and 16 show that Cuban Latinas are better educated than
Mexican Latinas. Higher education increases earning power. Differential

economic characteristics of Cubans and Mexicans is seen in Table 17.

Table 17. Economic Characteristics of Cubans and Mexicans

   
   
 

it. Below

     
   

as Receiving Per Capita

 

Ethnic Group

  

Poverty Line Public income
Assistance

 

 

 

 

 

Cuban 14.6 15.2 $13,786
Mexican 26.3 12.5 $7,447

 

Source: Rumbaut 199526

Cubans have a higher per capita income and have less people below the
poverty line than Mexicans. Cubans do have a slightly higher number of people
on public assistance, but this may be due to Mexicans having a higher number
of undocumented workers in the US. Although this data is not bifurcated for

gender, Tables 16 and 17 suggest that Cuban women are Important contributors

129

to the comparatively higher level of socioeconomic status of the Cuban
population. Cuban women are more educated, earn more, and have higher
status occupations than Mexican women. Additionally, Mexican women have a
higher unemployment rate than Cuban women. in 1991, 10.0% of Mexican
women (aged 16 years or older) were unemployed as compared to 8.6% of
Cuban women (US. Bureau of the Census 19922382).

As Mexicans can be compared to Cubans by looking at education and
income, Puerto Ricans be compared to these two other Latino subgroups. Table
18 below compares the percent of college graduates for Cubans, Mexicans, and

Puerto Ricans.

Table 18. Percent of Cubans, Mexicans, and Puerto Ricans Who are
College Graduates (persons aged 25 years or older)

 

 

 

 

 

Ethnic Group % College Graduate

Cuban 16.5

Mexican 6.3

Puerto Rican 9.5 i

 

 

Source: Rumbaut 199526

Cubans are more educated than Mexicans and Puerto Ricans, but Puerto
Ricans are more educated than Mexicans. As difference In educational levels

exist, so do differences in economic characteristics as seen in Table 19.

Table 19. Economic Characteristics of Cubans, Mexicans, and Puerto Ricans

130

 

 

 

 

T
Ethnic Group % Below % Receiving Per Capita
Poverty Line Public income
Assistance
Cuban 14.6 15.2 $13,786
Mexican 26.3 12.5 $7,447
Puerto Rican 31.7 26.9 $8,403

 

 

 

 

 

Source: Rumbaut199526

Cubans are better off than Mexicans and Puerto Ricans In terms of
having fewer people below the poverty line and a higher per capita income.
Mexicans have fewer people below the poverty line than Puerto Ricans but have
a lower per capita Income. Puerto Ricans may have a higher proportion of
people on public assistance because of US. citizenship. Puerto Ricans may be
more inclined to apply for govemmentai aid because the fear of deportation does
not exist as it may for undocumented Cubans and Mexicans. The slightly higher
per capita income for Puerto Ricans may be because Mexicans tend to have
slightly larger families (COSSMHO 1994:66-170). As another basis of economic

comparison, family income in 1990 is presented In Table 20.

131

Table 20. Total Family Annual income for Cubans, Mexicans,

and Puerto Ricans

 

 

  

 

 

 

 

 

 

 

 

 

  

 

 

 

:- JI—Ial
income Range Cuban Mexican Puerto Rican
(% per 1,000)
< $5,000 5.7 5.7 11.0
$5,000-$9,999 8.1 11.5 22.7
$10,000-$14,000 9.0 13.7 10.1
$15,000—$24,999 17.6 22.4 18.8
325,000-334,999 19.7 17.7 11.3
$35,000-$49,999 16.1 16.9 12.1
$50,000 + 23.9 12.2 14.1
Median income $31,439 $23,240 $18,008
(dollars) -

 

 

 

 

 

 

Source: US. Bureau of the Census 1992241

Table 20 makes the economic picture of the three groups a little clearer.

it is seen that for most Income ranges, Cubans are better off than Mexicans and

Mexicans are better off than Puerto Ricans. The only exceptions are in the less

than $5,000 range where Cubans and Mexicans are even, the $10,000-$14,000

range where there are less Puerto Ricans than Mexicans, and the $35,000-

349,000 range where there are only slightly more Mexicans than Cubans.

"Better off" means that there are a smaller proportion of the group in the lower

ranges and a higher proportion of the group in the higher ranges. Also, median

Income reflects the gradation of Cuban, Mexican, and Puerto Rican from better

off to worse off.

Lastly, occupational status is considered in comparing social and

132

economic forces that may effect gender, family, and culture. The result of this
comparison Is not surprising given the findings of educational level and income
for each group. Based on data from the US. Census for 1990, the occupational

distribution for each Latino subgroup is given in Table 21 below.

Table 21. Occupational Distribution for Cubans, Mexicans, and Puerto Ricans
(% per 1 ,000 in population for persons aged 16 years and older)

; Occupation Mexican Puerto Rican
Managerial and 25.1 10.4 17.9
' Professional

Tech, Sales, 33.1 22.5 32.0
f and Admin.

‘ Support
Services 12.0 18.9 17.5

I Precision 12.8 13.9 10.7
. Production,
Craft, and Repair

9 Operators, 15.6 25.9 20.7
' Fabricators, and
; Laborers

Farming, 1.4 8.4 1.2
. Forestry, and
Fishing

 

        
   
 
    
   
    
  

 

 

 

 

 

 

 

 

 

 

 

Source: US. Bureau of the Census 19922382

Cubans tend to be In the higher status occupations (managerial and
professional) while Mexicans tend to occupy the lower status occupations
(operators, fabricators, laborers, farming, forestry, and fishing). Puerto Ricans

do better than Mexicans in the higher status occupations and have slightly less

133

people In the lower status occupations. However, they do not fare as well as
Cubans.

When considering group comparisons, generational differences need to
be considered. This is because "generational comparisons are central to most
efforts to understand the prospects for the assimilation and acculturation of
immigrant groups into the mainstream of American society" (Oropesa and
Landale 199524).

l have argued thus far that Cubans fare better than Mexicans and Puerto
Ricans, and Mexicans better than Puerto Ricans. Oropesa and Landale (1995)
examined socioeconomic indicators for 3. generations of Latino children (17
years of age and younger) using 1990 US. Census data. First generation
children were defined as "foreign-bom children of foreign-Dom parents," second
as "native-Dom children with at least one foreign-00m parent," and third as
"native born children with native-bom parents" (Oropesa and Landale 1995211).
Table 22 below compares the socioeconomic Indicators of median household
Income, poverty rate, educational attainment, and "white collar" occupational
status for Cubans, Mexicans, and Puerto Ricans. This comparison is important
for it puts Into context the effect of immigration and acculturation on the

socioeconomic conditions of Latino subgroups.

134

Table 22. Socioeconomic Status Measures for Latino Children by Generation

           

 

 

 

. Ethnic Median % Below % % Head % Head
Group! Househld. Poverty Receiving Househld. Househld
j Gener. income Line Public College White
' A (100s) Assist. Grad. Collar
i Cubans
First 23.0 25.8 11.7 10.8 8.9
Second , 38.0 13.1 7.5 21.7 25.2
Third 32.6 22.2 9.2 20.2 25.5
Mexicans
First 19.0 44.2 16.7 3.2 4.0
Second 22.0 31.6 10.0 3.6 6.2
Third 25.0 28.1 12.5 8.2 13.3
Puerto
Ricans '
First 14.1 51.5 23.1 10.1 10.5
Second 21.0 38.0 28.4 7.0 10.8
Third . 19.0 40.5 21.4 1.91 d 13.9

 

 

 

 

 

 

 

Source: Oropesa and Landale 1995:19

it is seen that first, second, and third generation Cubans have higher
median Incomes than Mexicans and Puerto Ricans of the same generations, and
first, second and third generation Mexicans have higher median incomes than
Puerto Ricans of the same generations. More first, second, and third generation
Puerto Ricans are below the poverty line and on public assistance than Cubans
and Mexicans of the same generations, and more first second and third
generation Mexicans are below the poverty line and on public assistance than
Cubans of the same generations.

More first, second, and third generation Cuban heads of households have

135

college degrees than Puerto Ricans. More second generation Puerto Ricans
have college degrees than Mexicans of the same generation. However, by the
third generation, Mexicans and Puerto Ricans are about equal In the percentage
of heads of households with college degrees. The increase in second to third
generation heads of households with college degrees Is greater for Mexicans
than Puerto Ricans.

More first generation Puerto Ricans are in white collar occupations than
Cubans and Mexicans of the same generation and more first generation Cubans
than Mexicans are in white collar occupations. However, by the second
generation, Cubans overwhelmingly overtake Puerto Ricans in heads of
households In white collar occupations. Mexicans slightly improve but still
remain behind Cubans and Puerto Ricans. By the third generation, Cubans
minimally Improve but are still ahead of Puerto Ricans and Mexicans, Puerto
Ricans slightly improve, and Mexicans improve almost twice from the second
generation but are much closer to Puerto Ricans.

Except for percent heads of households in white collar occupations, and
percent heads of households with college educations, Cubans are better off than
Mexicans and Puerto Ricans, and Mexicans are better off than Puerto Ricans.
This is consistent with other data that has been presented in this study in
comparing socioeconomic status indicators among Latinos. Regarding
percentage heads of households in white collar occupations, it could be that

Puerto Ricans who Immigrate to the US. are unable to replicate their status on

136

the island In the US. Cubans who immigrate may have more extended family
who have been successful (due to govemmentai assistance programs) to help
the second generation In accessing white collar occupations. Most Mexican
immigrants come into menial positions and by the third generation may have
been "promot " to a white collar position over newly arrived immigrants.
Educationally, second and third generation Mexicans may have learned the
language and thus been able to better succeed at college. Puerto Ricans may
be more familiar with the language due to the territory’s commonwealth status
(i.e., English is taught in the schools and is more often used as a "second
language").

The above discussion demonstrates that differences exist among Cubans,
Mexicans, and Puerto Ricans pertaining to socioeconomic conditions. These
differential socioeconomic conditions may effect gender roles and the family
which may effect acculturation. Social and economic conditions are related to
history.

Bean and Trends (1987) argue that because Mexicans have a longer
history than the other Latino subgroups, they are more "generationaiiy diverse."
This diversity has resulted in some of the urban "upwardly mobile" Mexican
Americans forming a cultural allegiance with the United States. This occurred
because the transition from a rural, agricultural worker to an urban industrial
worker was seen by some Mexican Americans as an improvement of their life

circumstances. This perceived improvement facilitated a move toward cultural

137

assimilation with the dominant culture. Hope (although misplaced) was put by
many in the "American Dream." Bean and Tienda write that "cultural
manifestations of changes associated with the urbanization experience include
the trend toward a language shift away from Spanish, the declining isolation of
the barrio, and indicators pointing to a greater degree of assimilation into Anglo
society" (1987:22).

Cubans had the benefit of economic assistance from the United States.
However, many came thinking that some day they would return to Cuba. When
it became clear that Castro was not going to be overthrown by any counter-
revolution, Cubans quickly turned their energies toward becoming successful in
the United States. This promoted assimilation into the dominate culture where
"Cubans aggressively sought to learn English and new skills necessary for the
economic rewards that would eventually signal their social integration" (Bean
and Tienda 1987231).

Puerto Ricans have a dependent, patriarchal relationship with the United
States. Although Puerto Ricans are citizens, the country of their citizenship has
done little to promote economic and social mobility. Puerto Ricans are second
class citizens who fare little better than African Americans in their socioeconomic
status. Many Puerto Ricans go back and forth between the island and the
mainland (circular migration) which results in a shared culture between the two
geographic locales. This circular migration and "relegation to the lowest level of

the socioeconomic ladder are two Important defining features of the ethnic

138

structuring process for Puerto Ricans" (Bean and Tienda 1987226). Because of
this lower status and perpetual life of poverty, Puerto Ricans have maintained
strong ethnic communities and rejected a "quick transfer of cultural identity"
(Bean and Tienda 1987227).

Given the role that gender and family plays in culture (Baca-Zinn 1994), it
seems that a change in gender roles (as prompted by economic conditions) may
result in an acculturation effect. The chart below demonstrates the acculturation
process as i theorize it to operate. However, the only variable that is
measurable In this chart Is socioeconomic conditions. Whether or not these

relationships truly exist, remains to be determined.

Figure 1. Path Chart of Acculturation Processes

Historical
& Geographical
Circumstances
l
l
v
Contemporary
Social &
Economic
Conditions—«~--> Mexicans Cubans Puerto Ricans
l i l
I l l
v v v
Possible transformation of family and culture due to
changing role of women. This may result In varying degrees
of acculturation.

LIST OF REFERENCES

LIST OF REFERENCES

Adair, Linda S. 1987. "Nutrition in the Reproductive Years." Pp. 119-54 In
Nutritional Anthropology edited by Francis E. Johnston. New York: Liss.

Adams, Melissa M. 1995. "The Continuing Challenge of Preterm Delivery."
Journal of the American Medical Association 273:739-40.

Alicea, Marisa. 1994. "T he Latino immigration Experience: The Case of
Mexicanos, Puertorriquenos, and Cubanos." Pp. 35-56 in Handbook of
Hispanic Cultures in the United States: Sociology edited by F eiix
Padilla. Houston: Arte Publico Press.

Baca-Zinn, Maxine. 1994. "Mexican-Heritage Families in the
United States.” Pp. 161-72 in Handbook of Hispanic Cultures in
the United States: Sociology edited by Felix Padilla. Houston:
Arte Publico Press.

Back, Kurt W. 1977. “Food, Sex and Theory." Pp. 24-34 in Nutrition and
Anthropology in Action edited by Thomas K. Fitzgerald. Assen:
van Gorcum.

Balcazar, Hector. 1993. "Mexican Americans' intrauterine Growth Retardation
and Maternal Risk Factors." Ethnicity and Disease 32169-75.

Bean, Frank D. and Marta Trends. 1987. The Hispanic Population of the
United States. New York: Sage.

Becerra, Jose E., Carol J. R. Hegue, Hani K. Atrash, and NIisa Perez. 1991.
"infant Mortality Among Hispanics: A Portrait of Heterogeneity." Journal of
the American Medical Association 2652217-21.

Borges, Guilherrne, Malaquias Lopez-Cervantes, Ma Elena Medina-Mora,
Roberto Tapia-Conyer and Francisco Garrido. 1993. "Alcohol Consumption,
Low Birth Weight, and Preterm Delivery In the National Addiction Survey
(Mexico)." lntamational Journal of the Addictions 282355-68.

139

140

Brooks-Gum, J., Marie C. McCormick, and Margaret C. Heagarty. 1989.
"Preventing infant Mortality and Morbidity: Developmental Perspectives."
American Journal Orthopsychiatry 582288-96.

Cases, Manuel J., Burl R. Wagenheim, Robert Banchero, and
Juan Mendoza-Romero. 1995. "Hispanic Masculinity: Myth or
Psychological Schema Meriting Clinical Consideration." Pp. 231-44 in
Hispanic Psychology: Critical Issues in Theory and Research edited
by Amado M. Padilla. Thousand Oaks: Sage.

CDC. 1993. Monthly Vital Statistics Report 9:41.

CDC. 1994. "increasing incidence of Low Birthweight--United States,
1981-1991. MMWR 432335-39.

Cookerham, William C. 1986. Medical Sociology. Third Edition. New Jersey:
Prentice Hall.

Cole, Leslie. 1991. "Dietary Diplomat." UC Davis Magazine 9:28-31.

Collins, James W. and Richard J. David. 1993. "Differences in Neonatal
Mortality by Race, income, and Prenatal Care." Ethnicity and Disease
2:18-26.

Cooley, Charles H. 1962. Social Organization: A Study of the Larger Mind.
New York: Schoken.

COSSMHO (National Coalition of Hispanic Health and Human Services
Organization. 1994. Profiles of Health: Documenting Health and
Establishing Priorities. Washington, DC: COSSMHO.

Dewalt, Kathleen M. and Gretel H. Pelto. 1977. "Food Use and Household
Ecology In a Mexican Community." Pp. 79-93 In Nutrition and Anthropology
In Action edited by Thomas K. Fitzgerald. Assen: van Gorcum.

Diaz-Guerrero, Rogelio. 1977. "A Sociocultural Psychology?" Pp. 17-27 in
Chicano Psychology edited by Joe L. Martinez, Jr. New York: Academic
Press.

Dowling, Patrick T. and Michael Fisher. 1987. "Maternal Factors and Low
Birthweight infants: A Comparison of Blacks with Mexican Americans."
Journal of Family Practice 252153-58.

141

Duncan, Greg J., Brooks-Gunn, Jeanne, and Pamela Kato Klebanov. 1994.
"Economic Deprivation and Early Childhood Development." Child
Development 65:269-318.

Durkheim, Emile. [1893] 1984. The Division of Labor in Society.
New York: Free Press.

Eckstein, Eleanor F. 1980. Food, People, and Nutrition. Westport: AVl.

Fenster, Laura and Molly J. Coye. 1990. "Birthweight of infants Born to
Hispanic Women Employed In Agriculture." Archives of Environmental
Health 45:46-52.

Fitzpatrick, Joseph P. 1971. Puerto Rican Americans: The Meaning of
Migration to the Mainland. New York: Prentice Hall.

Freedman, Robert L. 1977. "Nutritional Anthropology: An Overview." Pp. 1-
23 in Nutrition and Anthropology in Action edited by Thomas K. Fitzgerald.
Assen: van Gorcum.

Fracasso, Maria P. Nancy A. Busch-Rossnagel, and Celia B. Fisher. 1994.
"The Relationship of Maternal Behavior and Acculturation to the Quality of
Attachment in Hispanic infants Living in New York City." Hispanic Journal of
Behavioral Sciences 162143-54.

From the MMWR. 1993. "Infant Mortality-United States, 1990." Ethnicity and
Disease 32427-31

Gabbe, S. G. 1991. Obstetrics: Normal and Problem Pregnancies.
New York: Churchill Livingston.

Garcia, Maria Cristina. 1994. "Cuban Women in the United States." Pp. 203-
18 In Handbook of Hispanic Cultures in the United States: Sociology edited
by Felix Padilla. Houston: Arte Publico Press.

Gerda Coil, Cynthia T. 1988. "The Consequences of Teenage Childbearing in
Traditional Puerto Rican Culture." Pp. 111-132 in Cultural Context of Infancy
edited by K Nugent and T. B. Brazelton. New Jersey: Abiex.

Garcia Coll, Cynthia T. 1990. "Developmental Outcome of Minority infants" A
Process-oriented Look Into Our Beginnings." Child Development 612270-89.

142

Garcia Coll, Cynthia T. and Elaine C. Meyer. 1993. "The Sociocultural
Context of infant Mortality." Pp. 1-35 in Handbook of Infant Mental Health
edited by C. Zeanah. New York: Guliford.

Gaviria, Moises, Gwen Stern, and Stephen L. Schensul. 1982. "Sociocultural
Factors and Perinatal Health in a Mexican American Community." Journal of
the National Medical Association 742983-89.

Habermas, Jurgen. 1976. Legitimation Crisis. London: Helnemann.

Hayes-Bautista, David E. 1992. "Latino Health indicators and the Underclass
Model: From Paradox to New Policy Models." Pp. 32-47 in Health Policy
and the Hispanic edited by Antonio Furino. Boulder. Westview.

Health United States 1990. 1992 Public Health Service, Centers for Disease
Control, National Center for Health Statistics. Maryland: US. Department of
Health and Human Services.

Hernandez, Jose. 1994. "Hispanics Blend Diversity." Pp. 17-34 in Handbook
of Hispanic Cultures In the United States: Sociology edited by Felix Padilla.
Houston: Arte Publico Press.

Hinton, Sarah M. and Diane R. Kerwin. 1981. Maternal Infant and Child
Nutrition: A Resource Book for Health Professionals. North Carolina:
Health Services\Consortium, inc.

Hoerr, Sharon (Associate Professor, Department of Food Science and Human
Nutrition, Michigan State University) 1995. Discussion with David A. Lopez.
July 6, 1995.

Hogue, Carol J. R., James W. Buehler, Lilo T. Strauss, and Jack C. Smith.
1987. "Overview of the National infant Mortality Surveillance (NIMS)
Project-Designs, Methods, Results." Public Health Reports 102:126-37.

Huston, Aletha C., Vonnie C. McLoyd, and Cynthia Garcia Coll. 1994.
"Children and Poverty: issues In Contemporary Research." Child
Development 652275-82.

lCPSR. 1994. Guide to Resources and Services 1994-1995. Ann Arbor:
inter-university Consortium for Political and Social Research.

James, Sherman A. 1993. "Racial and Ethnic Differences In infant Mortality
and Low Birthweight: A Psychosocial Critique." American Journal of
Preventive Medicine 92130-6.

143
Johnston, Francis E. (editor). 1987. Nutritional Anthropology. New York: Liss.

Kramer, Michael. 1987. "Determinants of Low Birth Weight: Methodological
Methodological Assessment and Meta-Analysis." Bulletin of World Health
Organization 52663-737.

Krasocvec, Katherine. 1991. "Background issues." Pp. 15-28 in Maternal
Nutrition and Pregnancy Outcomes: Anthropometric Assessment edited by
Katherine Krasovec and May Ann Anderson. Washington DC.

Lenski, Gerhard E. 1966. Power and Privilege: A Theory of Social
Stratification. New York: McGrew-Hill.

Levin, Jeffrey S. (1991) "Religious involvement Among Hispanic and Black
Mothers of Newborns." Hispanic Journal of Behavioral Sciences 13:436-447.

Lowenberg, Miriam E., E. Neige Todhunter, Eva D. Wilson, Moria C. Feeney,
and Jane R. Savage. 1968. Food and Men. New York: John Wiley and
Sons.

Magana, Aizita and Noreen M. Clark. 1995. "Examining a Paradox: Does
Religiosity Contribute to Positive Birth Outcomes in Mexican American
Populations?" Health Education Quarteriy 22:96-109.

Marin, Gerado and Barbara VanOss Marin. 1991. Research with Hispanic
Populations. Newbury Park: Sage.

Marks, Gary, Melinda Garcia, and Julie M. Solis. 1990. "Health Risk Behaviors
of Hispanics In the United States: Findings from HHANES, 1982-84.
American Journal of Public Health (Supplement) 80:29-26.

Martinez, Joe L. 1977. "Cross-Cultural Comparison of Chicanos and Anglos
on the Semantic Differential: Some implications for Psychology."
Pp. 29-43 in Chicano Psychology edited by Joe L. Martinez. New York:
Academic Press.

Mead, George H. 1969. On Social Psychology: Selected Papers edited by
Anselm Strauss. Chicago: University of Chicago Press.

Mendoza, Fernando S. 1994. "The Health of Latino Children in the United
States." Critical Issues for Children and Youth 4243-72.

144

Mendoza, Fernando 8., Stephanie J. Venture, R. Burciaga Valdez, Ricardo O.
Castillo, Laure Escoto Saidivar, Katherine Baisden, and Reynaldo Martorell.
1991. "Selected Measures of Health Status for Mexican American, Mainland
Puerto Rican, and Cuban American Children." Journal of the American
Medical Association 265:227- 33.

Michielutte, Robert, J. M. Ernest, Mary Lou Moore, Paul J. Meis, Penny C.
Sharp H. Bradley Wells, and Paul A. Buescher. 1992. "A Comparison of
Risk Assessment Models for Term and Preterm Low Birthweight."
Preventive Medicine 21298-109.

NorusislSPSS Inc., Marija. 1994. SPSS: SPSS 6/1 Base System User's
Guide, Part 2. Chicago: SPSS inc.

Oropesa, R. S. and Nancy S. Landale. 1995. "immigrant Legacies: The
Socioeconomic Circumstances of Children by Ethnicity and Generation in
the United States (Revised Edition)" Working Paper 95-01. Population
Research institute. The Pennsylvania State University.

Ortiz, Vilma. 1994. "Women of Color. A Demographic Overview." Pp. 13-40
In Women of Color in U. S. Society edited by Maxine Baca Zinn and
Bonnie Thornton Dill. Philadelphia: Temple University Press.

Padilla, Elena. 1958. Up From Puerto Rico. New York: Columbia University
Press.

Paneth, Nigel (Director, Program In Epidemiology, Michigan State University)
1994. Discussion with David A. Lopez. January 25, 1995.

Pedraza-Baiiey, Silvia. 1985. Political and Economic Migrants in America:
Cubans and Mexicans. Austin: University of Texas Press.

Perez, Lisandro. 1992. "Cuban Miami." Pp. 83-108 in Miami Nowl:
Immigration, Ethnicity, and Social Change edited by Guillermo J. Grenier
and Alex Stepick lll. Gainesville: University Press of Florida.

Perkins, Daniel F. 1995. "An Examination of the Organismic, Behavioral, and
Contextual Covariates Among Risk Behaviors of Adolescents." Ph.D.
dissertation, Department of Family and Child Ecology, Michigan State
University, East Lansing, Mi.

Phinney, Jean 8. 1995. "Ethnic identity and Self-Esteem: A Review and
integration" Pp. 57-70 in Hispanic Psychology: Critical Issues in Theory and
Research edited by Amando M. Padilla. Thousand Oaks: Sage.

145

Romero Gwynn, Eunice and Douglas Gwynn. 1993. "Foods and Dietary
Patterns of Latinos of Mexican Descent: Monograph." University of
Caiifomia Cooperative Extension Department of Nutrition Davis: Pilot issue.

Romero Gwynn, Eunice and Douglas Gwynn. 1994. "Foods and Dietary
Acculturation Among Hispanics" (in press) in Hispanic American
Encyclopedia. Davis: Arte Publico Press.

Rueschenberg, Erich J. and Raymond Burial. 1995. "Mexican American
Family Functioning and Acculturation: A Family Systems Perspective."
Pp. 15-25 in Hispanic Psychology: Critical Issues in Theory and Research
edited by Amado M. Padilla. Thousand Oaks: Sage.

Rumbaut, Ruben G. 1992. "A Problem in Search of an Explanation: Some
Research Questions on Assimilation and Immigrant Health, Mental Health,
and Education. Workshop on U. 8. Immigration Research: An Assessment
of Data Needs for Future Research National Academy of Sciences.
Washington, DC.

Rumbaut, Ruben G. 1995. "Immigrants from Latin America and the Caribbean:
A Socioeconomic Profile." Statistical Brief No. 6. East Lansing:
Julian Samora Research Institute.

Rumbaut, Ruben G. and John R. Weeks. 1989. "infant Mortality Among
lndochinese Refugees: Patterns of infant Mortality, Birthweight and Prenatal
Care in Comparative Perspective." Research in the Sociology of Health
Care 8:137-96.

Rumbaut, Ruben G. and John R. Weeks. 1993. "Ethnicity, Nativity, and the
Paradox of Perinatal Health and Morbidity: An Analysis of Sociocultural and
Biomedical Causal Factors." Paper Presented at the Annual Meeting of the
American Sociological Association. Miami, Florida. August 15, 1993.

Rumbaut, Ruben, 6., Leo R. Chavez, Robert J. Moser, Shelia M. Pickwell, and
Samuel M. Wishik. 1988. "The Politics of Migrant Health Care." Research
in the Sociology of Health Care 7:143-202.

Scribner, Richard and James H. Dwyer. 1989. "Acculturation and Low
Birthweight among Latinos in the Hispanic HANES." American Journal of
Public Health 79:1263-67.

Seitz, Victoria and Nancy H. Apfel. 1994. "Effects of a School for Pregnant
Students on the incidence of Low-Birthweight Deliveries." Child
Development 65:666-76.

146

Splegel, Claire. 1991. "Medical Care Problems, Not Health Habits, Pose Risk
to Latino Population, Poll Finds." Los Angeles Times March 29 1991.

Sung, John F., Gene A. McGrady, Diane L. Rowley, Carol J. R. Hogue,
Ernest Alema-Mensah, and Monica L. Lypson. 1993. "interactive Effect of
Race and Marital Status in Low BirthWeight." Ethnicity and Disease
3"129-36.

Swingewood, Alan. 1991. A Short History of Sociological Thought. New York:
St. Martin's Press.

Taffel, S. 1980. "Matemai Weight Gain and the Outcome of Pregnancy."
Vital and Health Statistics US. Department of Health and Human Services.
Publication (PHS) 86-1922. Washington, DC.

Trevino, Fernando M., Eugene Moyer, Burciaga Valdez, and
Christine A. Stroup-Benham. 1991. "Health insurance Coverage and
Utilization of Health Services by Mexican Americans, Mainland Puerto
Ricans, and Cuban Americans." Journal of the American Medical
Association 265:238-37.

US. Bureau of the Census. 1992. Statistical Abstract of the United States:
1992: (112th edition). Washington DC: US. Government Printing Office.

US. Department of Agriculture. 1994. Administrative Report 378.
Hyattsville, MD: Agricultural Research Service, Human Nutrition information
Service.

Voght, Paul W. 1993. Dictionary of Statistics and MethodoIogy: A
Nontechnical Guide for the Social Sciences. Newbury Park: Sage.

Weinman, Maxine L. and Peggy B. Smith. 1994. "US. and Mexico-Bom
Hispanic Teen Mothers: A Descriptive Study of Factors that Relate to
Postpartum Compliance." Hispanic Journal of Behavioral Sciences 16:
186-94.

Weber, Max. [1915] 1946. From Max Weber: Essays in Sociology.
New York: Oxford University Press.

Yep, Gustavo A. 1995. "Communicating the HIV/AIDS Risk to Hispanic
Populations: A Review and Integration." Pp. 196-212 in Hispanic
Psychology: Critical Issues in Theory and Research edited by Amado M.
Padilla. Thousand Oaks: Sage.

147

Yerushalmy, J. 1967. "The Classification of Newborn Infants by Birth Weight
and Gestational Age." The Journal of Pediatrics 71 :164-72.