Citation
A comparison study bewteen a low income sample of European Americans and Latin Americans in terms of stress and social support

Material Information

Title:
A comparison study bewteen a low income sample of European Americans and Latin Americans in terms of stress and social support
Creator:
Samper, Paula
Publication Date:
Language:
English
Physical Description:
xiii, 154 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Stress (Psychology) -- Cross-cultural studies ( lcsh )
European Americans -- Social networks ( lcsh )
Hispanic Americans -- Social networks ( lcsh )
Poor -- Social networks -- United States ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 140-154).
General Note:
Department of Psychology
Statement of Responsibility:
by Paula Samper.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
747099565 ( OCLC )
ocn747099565
Classification:
LD1193.L645 2010m S35 ( lcc )

Full Text
tl
A COMPARISON STUDY BETWEEN A LOW INCOME SAMPLE OF EUROPEAN
AMERICANS AND LATIN AMERICANS IN TERMS OF STRESS AND SOCIAL
SUPPORT
By
Paula Samper
B.A., Pontificia Universidad Javeriana 1995
M.A., Universidad Autonoma de Manizales 1997
A thesis submitted to the
University of Colorado Denver
In partial fulfillment
Of the requirements for the degree of
Master in Clinical Psychology
2010


This thesis for the Master in Clinical
Psychology degree by
Paula Samper
Has been approved
Allison Bashe
<0/7/ZLQd
Date


Samper, Paula (MA., Clinical Psychology)
A Comparison study between a Low Income Sample of European Americans
and Latin Americans in terms of Stress and Social Support
Thesis directed by Senior Professor Kevin Everhart
ABSTRACT
Extensive research has shown the positive effects that social support has on
stress. However, little research has been conducted to understand variations
among ethnic groups. This study looks at the differences in terms of stress
and social support (SS) between European Americans (EA) and Latin
Americans (LA) with high (HALA) and low degree of acculturation (LALA).
The study drew data from 1285 mothers from the Early Head Start (EHS)
program of young children. The mothers participating were living at or
below poverty level and included 819 EA and 466 LA. The measures of stress,
social support and acculturation were obtained from the mothers by
interviews at the EHS program at 31 sites in 30 states throughout the United
States, when the children were 14,15,24 and 26 months old respectively.
Dunst's Family Resource Scale (FRS) and Abidin's Parenting Distress Index
were used to measure SS and stress respectively and English proficiency was


used to measure acculturation. SS was measured as a whole with the total
score for the FRS and, the scores for its three subscales were used to measure
instrumental, emotional and informational support. An analysis of variance
(ANOVA) was conducted to look for significant differences in stress and SS
among groups. A multiple regression analysis was used to explore the
variables that better explain stress, and Spearman' rho and Pearson
correlations were used to examine the nature and direction of the
relationship between SS and acculturation. I found statistically significant
differences between LA and EA for both SS and stress. A statistically
significant difference was found between IAIA and EA in stress and in SS
and no significant difference between HALA and EA or between LALA was
found for any of the variables when using SS as a whole. When using the
types of support significant differences were found for both instrumental and
informational support for all groups at all intervals, but no difference was
found for emotional support. As expected higher levels of social support were
correlated with lower levels of stress. HALA and EA show higher levels of
social support and lower levels of stress than low acculturated LA. I also
found a positive correlation between social support and acculturation. Some
possible explanations for the results and limitations are discussed in the


study. In future studies, it would be of significance to include ethnic groups in
their natural environment.
This abstract accurately represents the content of the candidate's thesis. I
recommend its publication
Signei
Kevin Everhart


TABLE OF CONTENTS
Figures ....................................................... IX
Tables ......................................................... X
CHAPTER
1. INTRODUCTION.............................................14
2. REVIEW OF THE LITERATURE
Ethnicity..........................................18
Definitions..................................18
Construct of Collectivism- Individualism.....19
Latin Americans a collectivistic society and
European Americans an Individualistic .......21
Social Support
What is social support?......................23
Social Support, health and well being........24
Social Support, Ethnicity and individualistic
and collectivistic societies.................25
Stress
VI


Definitions
29
Low Socio-economical Status and Stress .31
Social Support and Stress...............32
Direct and buffering effect models......33
Acculturation
Understanding the concept of acculturation..........35
How to measure acculturation........................37
3. METHODS SECTION
Participants........................................41
Measures ...........................................42
Procedure ..........................................45
4. RESULTS SECTION...........................................48
5. DISCUSSION................................................56
APPENDIX
A. FIGURES ...............................................66
B. TABLES ................................................80
C. HUMAN SUBJECTS APPROVAL...............................140
REFERENCES..................................................141
VII


LIST OF FIGURES
Figure
A1 Stress at 14 months for EA and LA...............................66
A2 Stress at 14 months for all groups..............................67
A3 Stress at 24 months for EA and LA...............................68
A4 Stress at 24 months for all groups..............................69
A5 Social support at 15 months for EA and LA......................70
A6 Social support at 15 months for all groups......................71
A7 Social support at 26 months for EA and LA......................72
A8 Social support at 26 months for all groups......................73
A9 Instrumental support at 15 months for all groups................74
A10 Emotional support at 15 months for all groups...................75
All Informational and special needs support at 15 months for all
groups........................................................76
A12 Instrumental support at 26 months for all groups................77
A13 Emotional support at 26 months for all groups...................78
A14 Informational and special needs SS at 26 months for all groups...79
VIII


LIST OF TABLES
Table
B1 Case processing data for stress at 14 months for LA and EA ...80
B2 Descriptive Statistics for stress at 14 months for EA and LA ...81
B3 Normality tests for stress 14 for LA and EA..............82
B4 Case processing data for stress at 14 months for all groups ...83
B5 Descriptive statistics for stress at 14 months for all groups.... 84
B6 Test of Normality for stress at 14 months for all groups..85
B7 Case processing data for stress at 24 months for EA and LA.86
B8 Descriptive statistics for stress at 24 months for EA and LA ...87
B9 Normality tests for stress at 24 months for EA and LA....88
BIO Case processing data for stress at 24 months for all groups.... 89
Bll Descriptive statistics for stress at 24 months for all groups.... 90
B12 Normality tests for stress at 24 months for all groups...91
B13 Case processing data for SS at 15 months for EA and LA..92
B14 Descriptive statistics for SS at 15 months for EZ and LA..93
B15 Normality tests for social support at 15 m for EA and LA.94
B16 Case processing data for SS at 15 months for all groups.95
B17 Descriptive statistics for SS at 15 months for all groups.96
IX


B18 Normality tests for social support at 15 m for all groups..97
B19 Case processing data at 26 months for EA and LA............98
B20 Descriptive statistics for SS at 26 months for EA and LA...99
B21 Normality tests for SS at 26 months for EA and LA.........100
B22 Case processing data for SS at 26 months for all groups...101
B23 Descriptive statistics for SS at 26 months for all groups.102
B24 Normality tests for SS at 26 months for all groups........103
B25 A Comparison analysis of the variables for LA and EA......104
B26 Analysis of variance for stress 14 and SS 15 among all groups
......................................................105
B27 Analysis of variance for stress 24 and SS 26 among all groups
......................................................106
B28 Levene test for stress 14 and SS 15.......................107
B29 Multiple comparisons for stress at 14 among all groups.108
B30 Multiple comparisons for SS at 15 for all groups..........109
B31 Tukey and Duncan for SS 15 for all groups.................109
B32 Levene test for SS 26.....................................110
B33 Multiple comparisons for SS 26 for all groups.............110
B34 Tukey and Duncan for SS 26 for all groups.................Ill
X


B35 Multiple regression analysis for stress at 14 months.....112
B36 Multiple regression analysis for stress at 24 months.....113
B37 Pearson Correlations for SS 15 for all LA groups..........114
B38 Spearman's rho correlations for SS 15 for all LA groups..114
B39 Pearson Correlations for SS 15 for all LA groups and EA..115
B40 Spearman's rho Correlations for SS 15 for all LA groups and EA
..........................................................115
B41 Pearson Correlations for SS 26 for all LA groups..........116
B42 Spearman's rho Correlations for SS 15 for all LA groups..116
B43 Pearson Correlations for SS 15 for all LA groups and EA..117
B44 Spearman's rho Correlations for SS 15 for all LA groups and EA
..........................................................117
B45 Descriptive statistics for all demographic at 15 months..118
B46 Descriptive statistics for all demographic at 26 months..119
B47 Case processing data for SS subscales at 15 months........120
B48 Normality Tests for SS subscales at 15 months.............121
B49 Analysis of variance of the three SS subscales at 15 months .122
B50 Homogeneity of variance for instrumental and informational SS
at 15 months..............................................123
XI


B51 Multiple comparison for instrumental and informational SS at
15 months............................................124
B52 Case processing data for the three SS scales at 26 months.... 125
B53 Normality tests for the three scales at 26 months....126
B54 Analyses of variance for the three SS scales at 26 months.127
B55 Homogeneity of variances for the three scales at 26 months
.........................................................128
B56 Multiple comparisons for instrumental and informational SS at
26 months............................................129
B57 Multiple regression analyses including the SS scales for stress
at 14 months.........................................130
B58 Multiple regression analyse including the SS scales for stress 24
.....................................................131
B59 Pearson correlation for Instrumental support at 15 months.132
B60 Spearman's rho correlation for Instrumental SS at 15 months
.........................................................132
B61 Pearson correlation for emotional SS at 15 months..........133
B62 Spearman's rho correlation for emotional SS at 15 m .......133
B63 Pearson correlation for informational SS at 15 m...........134
XII


B64 Spearman's rho correlation for informational SS at 15 m..134
B65 Pearson correlation for Instrumental SS at 26 months.....135
B66 Spearman's rho correlation for Instrumental SS at 26 m...135
B67 Pearson correlation for emotional support at 26 months...136
B68 Spearman's rho correlation for emotional SS at 26 months...l36
B69 Pearson correlation for Informational SS at 26 months....137
B70 Spearman's rho correlation for informational SS at 26 m 137
B71 Descriptive statistics with SS subscales at 14 months....138
B72 Descriptive statistics with SS subscales at 24 months ...139
XIII


CHAPTER 1
INTRODUCTION
The role of social support (SS) as a moderator of stress has been
well documented in the literature, and emotional support in particular
has been clearly linked to health outcomes (Alferi, Caver, Antoni & Duran,
2001; Aneshensel & Frerichs, 1982; Bal, Crombez, Van Oost &
Debourdeaudhuij, 2003; Billings and Moos, 1982 & Logsdon, 2000).
However, little is known about differences in SS and its effect on
stress among different ethnic groups. There seems to be no agreement
among researchers regarding ethnic differences in SS, and those reported
vary among studies and are somewhat controversial. Research suggests
that collectivistic values dominated the history and cultural tradition of
Latinas/os and other ethnics groups (Allik & Realo, 2004; Marin &
Triandis, 1985) while individualistic values dominated Americans of
European descents (Allik & Realo, 2004; Coon & Kemmelmeier 2001;
Gomez, 2003; Hofstede, 1980,1991).
Individuals who subscribe to individualistic cultural values rely and
value support from others to a lesser extent than individuals who have a
14


collectivistic value orientation (Kim, Triandis, Kagitcibasi & Yoon, 1994;
Triandis 1995).
Acculturation also seems to have a role on the way and degree in
which SS is perceived and sought by Latinos (Dressier & Bernal, 1982;
Griffith, 1983; Lopez & Contreras, 2005 & Orshan, 1996,1999).
In an attempt to address the aforementioned issues the present
study drew on data from 1,285 Mothers of young children living in the
United States. The sample included 819 European Americans (EA) and
466 Latino Americans (LA). To account for the effects that acculturation
can have on the way and degree in which SS is perceived and sought by
Latinos, the LA group was divided into low and highly acculturated
according to their English skills, according to the parameters described in
the procedure section.
The first issue addressed by the study is the relationship between
stress and SS. Levels of stress are expected to decrease as the SS level
increases independently of the ethnic group and/or degree of
acculturation.
Based on the characteristics and classifications of collectivistic and
individualistic societies described above LA are expected to have higher
15


levels of social support than EA and therefore LA are also expected to
have lower levels of stress than EA.
As described above, acculturation can influence the relationship
between stress and SS. SS may not be as important for EA and highly
acculturated Latinos as it is for low acculturated LA, therefore I will
expect to find a negative correlation between SS and acculturation. In
other words, highly acculturated LA will be expected to score lower levels
of SS than low acculturated LA, and EA will be expected to score lower
levels of SS than both low and highly acculturated LA on SS as a whole as
well as on the different types of support: instrumental, emotional and
informational- special needs.
A better understanding of the relationship among stress, SS,
acculturation and ethnicity can be a highly relevant issue for stress
treatment, research and prevention, and it may provide new perspectives
on how to approach and manage stress among groups with different
cultural backgrounds and degrees of acculturation.
Stress management and stress research have been targeted equally
for different cultures, which could lead to lost opportunities. Just as heart
attacks among women were initially treated from a male perspective,
16


(Schwarzer,R., & Leppin, A., 1991), cultural differences in response to
stress could represent an important issue enabling us to look at stress
from a multicultural perspective and providing improved tools for stress
research and management There may be resources available for certain
populations that have not been used for research or treatment These
differences are of concern to the present study.
The definitions as well as the relationship among the concepts of
stress, ethnicity, acculturation, collectivism, individualism and SS are key
issues in the foregoing study. Therefore, the following pages are
dedicated to defining and showing the research as well as the relevance
and relationship of these concepts to the hypothesis of this study.
Knowing how to approach stress prevention and treatment for different
cultures and how SS works for them constitutes an important
contribution to the field.
17


CHAPTER 2
REVIEW OF THE LITERATURE
Ethnicity
Ethnicity has been defined in many ways by different authors and
there seems to be no agreement as to one universal definition.
The term ethnicity has often been used interchangeably with ethnic
origin. Ethnicity has been used to designate national or geographic origin.
In fact, the word ethnicity is derived from the Greek word ethnos that
means "nation" (Kato, P.M., 1996). McGoldrick (1982) defines ethnicity
as "a sense of commonality transmitted over generations by the family
and reinforced by the surrounding community. It is more than skin color,
religion, or national and geographic origin" (p.222). In the present study,
ethnicity will be considered as the unique characteristics that have been
transmitted over generations through a common ancestry. These
characteristics are more than skin color, religion, history, language or
geographic origin; they are deep psychological processes that determine
our thinking, feeling and behaviors in a common way. Individuals who
share ancestors also share similar patterns, even if they are not aware of
18


it. These patterns are different across ethnic groups (Triandis, 1994 &
McGoldrick et al., 1996).
The construct of collectivism- individualism (C-I), or the tendency of
cultures to think and behave primarily in terms of the in-group or in
terms of the individual (Hofstede, 1980,1991,2001; Triandis, 1995) has
been described as one of the most unique attributes of cultural variation
and one of the most important ways in which ethnic groups differ
(Fijneman, Willemsen, & Poortinga, 1996; Hofstede, 1980; Phinney, 1996;
Shkodriani & Gibbons, 1995; Triandis, 1995).
From a cultural perspective, collectivism and individualism are
considered to represent opposite sides of a continuum, and cultures are
usually described as being either individualistic or collectivistic in their
orientation (Triandis, 1994). The C-I construct can be used to explain
cross-cultural differences over a broad range of situations, including but
not limited to SS network. (Fijneman et al., 1996).
Collectivist societies stress group solidarity, duties and obligations,
stable friendships, (Kim, et al., 1994) and they give priority to the goals of
their in-groups (Triandis, 2001). According to Hofstede (1991),
Collectivistic societies are those in which "people from birth onwards are
19


integrated into strong, cohesive in- groups which throughout peoples
lifetime continue to protect them in exchange for unquestioning loyalty"
(p. 51). Individuals from collectivistic societies are characterized by a
reliance on others for help and advice and a high degree of social
responsibility and sharing (Briley & Wyer, 2001). In collectivistic
societies interdependence among family members, which often includes
the extended family is also very important Collectivism, according to Hui
(1988) is related to solidarity, concern for others and integration with
other people.
Individualistic societies on the other hand, have been characterized
by focus on oneself as an independent being, lack of attention to the views
of others, relatively little concern for family and relatives (Triandis,
1995), emphasis in autonomy, emotional independence, privacy,
individual initiative, competitiveness, universalism (Kim et al., 1994) and
emotional independence from "groups", organizations and other
collectives (Hofstede, 1980). Individuals from individualistic societies
give priority to their personal goals over the goals of their in-groups and
they are autonomous and independent from their in-groups (Triadis,
2001). The individual is the single most important unit in any social
20


setting, independence is stressed, individual achievement rewarded and
the loyalty to a given group is very weak (Montaigne & Bowen. 2000).
Although I acknowledge that each individual can have components
of both individualism and collectivism, I do agree with Triandis and
Gelfand (1998) that cultures have a unique impact on individuals and
such an impact leads to the differentiation between individualistic and
collectivistic cultures or societies. In fact a broad body of research
supports that Latin- Americans cultures can be characterized as
collectivistic while Americans and European-American cultures can be
characterized as individualistic societies.
Hofstede (1980,1991) found the United States (US) to be the most
individualistic country among 53 countries surveyed while LA countries
fall on the opposite side of the spectrum as very collectivistic cultures.
Hispanics in the US also displayed the same collectivistic values as LAs
outside the US (Marin & Triandis, 1985). In addition, Gomez (2003)
found that acculturation was not related to individualism in a population
of Hispanic MBAs. Allik & Realo (2004) studied 37 countries and also
found that the US was the country with the highest level of individualism,
followed by Canada and European countries such as Norway, Denmark,
21


Finland and Germany. LA countries (such as Argentina, Brazil, Chile &
Mexico) on the other hand have shown very low levels of individualism.
Coon and Kemmelmeier (2001) studied the four largest ethnic groups in
the United States and found that the three largest minority groups in the
United States (African Americans, Asian Americans and LA) score higher
in collectivism than EA. And, EA scored higher in individualism than all
the minority groups.
Shkodriani and Gibbons, (1995) also reported than Mexican
students were significantly more collectivistic than US students.
All these empirical cross-cultural studies seem to support the
present study's classification of LA as a collectivistic culture and EA as an
individualistic culture (Triandis, 1991; McGoldrick, Giordano and Pearce,
1996).
According to the definitions and values held by individualistic and
collectivistic societies, individuals from collectivistic societies will be
expected to have and seek more SS than individuals from individualistic
societies independent of their degree of acculturation.
Having established the role of ethnicity and how it affects different
cultures with different degrees of individualism and collectivism, I will
22


explore the concept of SS and how it is related to cultural backgrounds.
Social Support
Many definitions of social support (SS) have been given and many
terms such as: "social network", "social relationships", "social activity",
"social resources" and "social ties", have been used as equivalents of SS.
However, there seems to be an agreement among researchers that SS
makes reference to receiving and/or perceiving emotional or
instrumental help from others. Caplan (1974) defined SS as feedback
given to a distressed person with the focus on "action", "identity", and
"guidance" as the supporter tries to help the supportee understand the
stressor and /or identify ways to cope with it
Ullman & Siegel (1994) defined SS as information from others that
one is cared for and loved, esteemed and valued; a sense of belonging and
being part of a network of communication and mutual obligations that
includes a spouse, parents, relatives, friends, and social and community
contacts among others. Cohen & Syme (1985) defined SS more broadly as
the resources that one person provides to another. Other authors such us
House, Umberson & Landis (1988) use the term social relationships as an
23


umbrella term that includes: SS, social integration, and social networks.
They describe social relationships as the existence, number and
frequency of social contacts. In the present study, SS will be understood
as the resources available and/or perceived (Cohen et al.,2003) by an
individual including informational, instrumental tangible (provision of
material support such as services, financial assistance or goods) and
emotional support that help the individual feel that he/she has a social
network to rely on.
For more than three decades, researchers have been studying the
construct of SS and its effects on health and well-being. Numerous
studies have shown that SS is linked to psychological and physical health
outcomes and it may also play a role in the etiology and maintenance of
diseases and illnesses. Most important, prospective studies such as those
conducted by Berkman & Syme (1979) and House, Robbins & Metzger
(1982) have shown a link between SS and mortality. They found that
mortality from all causes was greater among individuals with low SS.
Several studies have also reported a positive relationship between
mental health and SS (Alferi, Caver, Antoni & Duran, 2001; Aneshensel &
Frerichs, 1982; Bal, Crombez, Van Oost & Debourdeaudhuij, 2003; Billings
24


and Moos, 1982; & Logsdon, 2000).
Some examples include: the findings reported by Pengilly & Dowd
(2000) who found that individuals with relatively high levels of SS were
less prone to depression in the face of stress. And those of Aranda,
Castaneda, Lee & Sobel (2001) who reported that lower levels of SS from
spouse or partner predicted depressive symptoms for pregnant women.
The role of SS as a stress moderator has also been well documented,
and it will be discussed later on.
While many aspects of SS have been widely studied, little research
has been conducted to look at differences among ethnic groups. Some of
the studies that have been conducted however, have reported finding that
not only the amount but also the source and kind of SS can vary among
groups.
The literature seems to suggest the existence of cultural differences
among ethnic groups being SS usually higher but almost more important
or necessary for individuals from collectivistic societies such as African
Americans, Asian Americans and LA compared to EA (Perez, 1997;
French, Rianasari, Pidala, Nelwan & Buhrmester 2001).
Perez (1997) for example, found that individuals from Collectivistic
25


societies have a higher level of perceived SS from family than friends
compared to individuals from individualistic societies. Veron & Roberts
(1985) found that Mexican Americans reported having more friends and
being more likely to rely on their relatives for different needs such as a
caretaker or borrowing money than EA. Sabogal, Marin, Otero-Sabogal,
VanOss-Marin & Perez-Stable (1987) also found that Hispanics, even
highly acculturated ones, score higher in perceived support from their
family than non-Hispanic whites. According to their findings the strength
of family values tends to decrease with the increase in the level of
acculturation, but does not disappear. First generation Hispanics, and
those who were born or had spent their first 15 years of life in Latin
America, reported higher scores on the Familial obligations and family as
referent factors than second generation Hispanics and those who were
born or grew up in the U.S. The respondent's perceived support from the
family however, was not affected by acculturation, place of birth,
generation, or place of growing up. Hispanics from Mexico, Cuba and
Central America reported similar levels of agreement with the items of
perceived family support. Overall, LA reported a higher mean in the
Familial obligation, the perceived support from the family and the family
26


as referent dimensions than non-Hispanic whites (WNH). Campos,
Dunkel & Abdou, (2008) also found significantly higher levels of familism
in LA than EA. These findings seem to support that individuals from
collectivistic societies do have more perceived and available SS, at least
from their families, than EA or at least LA seem to need more SS than EA.
Sagrestano, Feldman, Killingsworth, Woo & Dunkel-Shetter (1999)
found significant differences in terms of SS between LA and EA women
during pregnancy. EA women reported more perceived support from the
baby's father, having a greater number of friends and relatives in their
social network and having more friends and receiving significantly more
support from them than LA, but LA reported more frequent interactions,
higher quality relationships and more support from family members than
EA regardless of demographic differences. They also found that the
amount of support varies according to the socio-economical level, being
higher for women from high SES than those from low SES. These findings
suggest that ethnicity is an important component to understanding
support process fully and that differences among ethnic groups not only
exist in terms of quantity but also of kind and quality.
Krzysztof and Norris (2000) found that LA had the greatest
27


preference for seeking help from their family compared to EA.
Latinos were equally comfortable seeking the three types of support
(informational, tangible and emotional) while EA were more comfortable
seeking informational support followed by emotional support and
tangible support last. The level of comfort seeking help was not context
dependent for the LA. On the contrary, for the EA was always influenced
by the context, especially related to seeking tangible help. A clear
preference to seek help from the family than from any other source was
found in the Latino families.
The findings reported by Krzysztof and Norris (2000) suggests that
differences among ethnic groups are not only related to the amount of SS
available but also, to the perception, access and willingness to ask for it as
well as the kinds of SS available.
The kind of SS perceived or available seems to vary among ethnic
groups and appears to be very important. It can make a difference in
terms of the effectiveness of SS to help reduce stress. According to Taylor
(1999) of the different kinds of SS, support from the family appears to
have particularly strong effects on physiological functioning.
Lack of SS during times of need can also be very stressful in itself,
28


especially for people with high needs of SS but insufficient opportunities
to obtain it (Taylor 1999]. For example, the lack of SS may not be as
stressful or as important for an EA (individualistic society) or for a highly
acculturated LA (collectivistic society) as it may be for a low acculturated
LA. This phenomena is what Cohen & Wills (1985) describe as the
matching hypothesis which sustains that effective SS may depend on a
match between one's needs and what one receives from others in ones
social network.
Learning about the way (s) in which SS works for LA and EA and the
role that acculturation plays on it, can be very important for a better
design of prevention and treatment programs for these populations.
Let us now look at how this is related to stress and the implications
thereof.
Stress
Stress can be defined as "a negative emotional experience
accompanied by predictable biochemical, physiological, cognitive and
behavioral changes that are directed either toward altering the stressful
event or accommodating to its effects" (Taylor, 1999, p.168). Stress arises
29


when the individual appraises a situation as threatening or demanding
and does not have the appropriate coping strategies to respond (Lazarus
& Launier, 1978; Cohen & Wills, 1985). The biochemical, behavioral, and
physiological changes that take place in the organism in response to a
stressor or stressful event have an adaptive purpose. These changes are
designed to help the individual to react, think and act better under those
stressful circumstances. However, the response should disappear after
the stressor is gone and the organism should return to balance. When the
stress is constant, too frequent or long lasting, the organism continues to
produce this physical activation, which eventually goes from being
adaptive to being detrimental and can cause damage to the system. This
constant activation of the system produces an alteration of the
individual's functioning and can over time affect the individual's physical
and/or psychological wellbeing (Lazarus & Launier, 1978; Crespo et al,
1998). Even the cumulative impact of small daily stressors can wear
down an individual, predisposing him or her to become ill (Taylor, 1999).
Mechanisms linking stress to illness include serious and constant
disruptions of the immune or endocrine system and the elevation of
physiological responses and behavioral adaptations accompanied by a
30


deficit of the individual's ability to cope (Baum, Singer & Baum, 1981;
Cohen & Wills, 1985).
There are many factors that can cause stress and these are called
stressors. Poverty, also called low SES, has been described as one of them.
The lack of access to basic resources such as food, housing, heat, child
care, education and/or money can generate stress on an individual and
therefore inherent consequences on health and well being of said
individual (Adler, Marmot, McEwen & Stewart, 1999; Gallo & Matthews,
2003; Lynch & Kaplan, 2000; Taylor, 1999). Individuals with low SES
showed higher levels of stress than individuals from high SES and this
constant life stress predicted poor health outcomes and was related to
depression (Cohen, Kaplan & Salonen, 1999). Adler & Stewart (2010) also
consider that low SES exposes the individual to multiple risk factors and
these risk factors increase the levels of stress and/or predispose the
individual to stress. Low SES measured through poverty level will be
considered as a stressor for the present study.
As discussed previously, the positive association between SS and
well- being has been well documented. Numerous studies indicate that
people who have partners, friends and families that provide them with
31


psychological, emotional, and tangible support are in better health and
under less stress than those with lower SS (Broadhead et al., 1983., Leavy,
1983., Mitchell, Billings & Moos, 1982).
The protective effects of SS in prevention and/or reduction of
psychological and physiological disorders related to stressful life events
have been found in numerous studies (Pengilly & Dowd, 2000; Fiore,
Coppel, Becker & Cox, 1986). SS has been reported to effectively reduce
psychological distress, such as anxiety or depression, during times of
stress (Fleming, Baum, Gisriel & Gatchel., 1982; Haines, Hurlbert & Beggs,
1996; Pengilly & Dowd, 2000). People with high levels of SS may
experience less stress when confronted with a stressful situation, and
may cope with it more successfully (Taylor, 1999). According to Pengilly
& Dowds (2000) findings, SS appears to buffer the effects of stress on
depression. Cohen, Hammen, Risha & Daley (2003) also found that both
higher levels of stress and lower levels of SS from the total network
independently predicted depressive recurrence over a 1-year follow-up
in a population of individuals with bipolar disorder, after controlling for
clinical history and compliance.
In addition SS appears to lower the likelihood of illness, help people
32


hold off or minimize complications from more serious medical conditions
and disorders, speed recovery from illness (Kulik & Mahler, 1989) and
reduce the risk of mortality (Berkman, 1982; House, Umberson, & Landis,
1988, Taylor, 1999).
Alferi, Caver, Antoni, Weiss & Duran (2001) for example, found that
emotional support pre-surgery predicted less distress post-surgery and
instrumental support from the spouse pre-surgery was also related to
less distress post-surgery. Perceptions of spousal support were
consistently related to lower distress.
SS also appears to help individuals to remove themselves from a
stressful situation. Matud-Aznar, Aguilera, Marrero & Moraza (2003)
found that battered women currently living with the aggressor have less
perceived and structural SS than women who had been battered and are
no longer living with the abusive partner, and the women in the control
group.
Two models have been used to explain the mechanisms through
which SS moderates stress and brings positive health outcomes. The main
effect model proposes that there is an overall beneficial effect of SS
because having social networks provide people with a regular positive
33


experience and a socially rewarded role in the community as well as a
sense of predictability and stability. There is in this case a direct
relationship between SS and stress. SS from this model is also viewed as a
protective factor rather than a healing factor meaning that SS reduces the
physiological impact of stress (Cohen & Wills, 1985; Schwarzer & Leppin,
1991).
The buffering model on the other hand, suggests that SS buffers the
effects of stress, such as: weakened immunity, depression and anxiety, by
reducing the impact of the stressful situation (La Rocca, House & French,
1980). Sheldon & Wills (1985) describe two ways in which SS buffers
stress: one, by attenuating or preventing a stress appraisal response and
two, by intervening between the stressful event and the onset of
pathological outcomes through reducing or eliminating the stress
reaction or directly influencing physiological processes (House, 1981;
Cohen & Wills, 1985).
Support has been provided for both models. However, there seems
to be no agreement on which of the two models more accurately explain
the mechanisms through which SS moderates stress. Indeed, what has
been well documented is that SS buffers or moderates stress irrespective
34


of the mechanism, which best explains this process (Alferi et al. 2001.,
Pengilly & Dowd, 2000., Perez, 1999.,Baum, Singer & Baum, 1981,
Lazarus, 1966 among others).
I can argue that each model could better explain how SS mediates
stress in each particular ethnic group. The main effect model for example
may explain better the effects that SS has on stress in LA individuals.
Since LA have a more consistent, permanent and long term SS from their
families, SS acts as a protective and preventing factor rather than a
healing factor. The buffering effect model on the other hand could better
explain the effect of SS on stress in EA. In this case SS if available helps
reduce the impact of stress rather than prevents it.
In sum, SS "reduces the experience of stress, enhances the ability to
cope, can reduces the prospect of mental and physical distress or illness,
and speeds recovery when illness does occur" (Taylor, 1999. p.228).
Acculturation
Throughout the years acculturation has been defined in different
but similar ways. Redfield, Litnon, and Herskovits, (1936) and Graves
(1967) defined acculturation as the continuous and first hand contact
35


between cultures that produces an effect on the subject with consequent
changes in language preference, attitudes, identity, values, behaviors and
schemas. Acculturation can also be defined as the process of altering a
society and the modification of a primitive culture by contact with an
advanced culture. And Rogler, Cortes and Malgady (1991) defined it as
the process through which immigrants change their behavior and
attitudes toward those of the host society.
The degree in which acculturation affects Latin American's
psychosocial processes, values, behaviors and schemas are not quite
clear. Findings related to the effect that the process of acculturation has
on individuals, include the following. Sabogal et al. (1987) found that the
strength of familism values tends to decrease with the increased level of
acculturation. First generation LA, and those who were born or had spent
their first 15 years in Latin America, reported higher scores on the
familial obligations and family as referent factors than second generation
LA or those who were born or grew up in the United States. The only
factor that was not affected by acculturation was the perceived support
from the family. Highly acculturated LA score higher in perceived
support from the family than EA. On the other hand, Reynoso, Felice, &
36


Shragg (1993} & Zambrana, Scrimshaw, Collins & Dunkel-Schetter (1997)
found that highly acculturated adolescents are more likely to experience
prenatal stress and receive less support from their childs father than
their less acculturated peers. Another variable that is usually included
when looking at the effects of acculturation is the role of language. On
this topic Lopez & Contreras (2005) reported a strong association
between linguistic balance (ability and desire to communicate both in
English and Spanish in multiple arenas) and adjustment. They found that
the relationship between language, acculturation and psychological
adjustment was different for the different language groups. In the English
group, mother's greater levels of language acculturation were related to
higher levels of symptomatology; in the Spanish group, greater levels of
language acculturation were related to lower levels of symptomatology.
Orshan (1996; 1999) on the other hand reported that language
acculturation was unrelated to self-esteem and SS in both Puerto Rican
and Dominican pregnant adolescents. Other studies that have looked at
the relationship between acculturation and psychological adjustment
include: Dressier & Bernal (1982) who found that Puerto Ricans with the
longest length of residence and the fewest psychosocial resources
37


[including SS) for coping reported the worst stress outcomes. Griffith
(1983), found that Spanish-speaking subjects had lower scores of
psychosocial dysfunctions but higher scores of anxiety than the subjects
on the English speaking and bilingual groups. In addition he found that
first generation individuals scored lower in psychosocial dysfunctions
than the individuals from third-generation. Lack of verbal proficiency in
English was significantly associated with higher levels of dysfunctional
symptomatology (Warheit et al., 1985; Salgado de Snyder, 1987). Lang et
al. (1982) reported that the best adjusted Hispanics were those who were
bilingual, but oriented more toward Hispanic than Anglo culture Caetano
(1987) found that more acculturated women had five times more the
probability of being drinkers than low acculturated women. And Burman
et al. (1997) reported that highly acculturated US born Mexican
Americans had higher lifetime prevalence scores of dysthymia, phobia
and major depression.
Some researchers have argued that the reason for these conflicting
and in some cases contradictory results is the inappropriate
conceptualization and measurement of the acculturation constructs.
(Cabassa, 2003; Chun, Balls-Organista, & Marin, 2003; Rogler, Cortes &
38


Malgady, 1991).
Language use and preference however, has been described as an
essential component of the construct of acculturation. This component
has been used to measure acculturation and has proven to be valid and
reliable. It has also been widely used by many researchers (Lopez &
Contreras, 2005; Cuellar, Arnold, & Maldonado, 1995; Salgado de Snyder
1987; Neff, 1986; Warheit, Vega, Auth, & Meinhardt, 1985; Ortiz & Arce,
1984; Griffith, 1983; Melville, 1978 (as cited in Rogler, Cortes & Malgady,
1991)). And this variable will be used here to measure level of
acculturation.
SS seems to help reduce and/or prevent stress. I can also say that an
individuals degree of acculturation could play an important role in the
amount or kind of SS received and-or perceived by that individual as well
as the role that this concept plays in their lives. It has been shown that LA
have collectivistic values while EA are governed by individualistic values.
The role that individualistic and collectivistic values play on stress and SS
however has not been well documented.
Achieving a better understanding of how SS moderates stress on
individuals from different cultural backgrounds such as LA and EA could
39


be very helpful for stress research, treatment and prevention.
Considering all of the above, it would be reasonable to hypothesize
that:
1 . Levels of stress are expected to decrease as SS levels increase
independently of the ethnic group and/or degree of acculturation.
2. SS total scores will be expected to be higher for individuals
from collectivistic societies such as LA than for individuals from
individualistic societies such as EA.
3. SS scores for the three types of social support: instrumental,
emotional and informational-special needs are expected to be higher for
LA than for EA.
4. LA will be expected to have lower levels of parental stress than
EA.
5. A negative correlation will be expected between SS and
acculturation. Highly acculturated LA will be expected to score lower
levels of SS than low acculturated LA, and EA will be expected to score
lower levels of SS than both high and low acculturated LA.
40


CHAPTER 3
METHODS
Participants
The participants were part of the Head Start / Public School
Early Childhood Transition Demonstration Study 1991-1999
program. This study was established over 30 years ago, and it was
designed to provide services to low-income children and their
families. The Head Start program was designed from its earliest days
as a multi-pronged, two-generation program, providing and securing
services in the areas of education, health, social services, and parent
involvement for children and families across the county. The study
included a total of 10,829 participants. Participants included mothers
between the ages of 18 and 59 and their toddlers. The subjects pertain
to six ethnic groups and they all comply with the EHS requirements
and have been in the program as controls or participants since the
mothers were pregnant or the babies were newborns; all the
participants' incomes were at or below the federal poverty level. The
subjects were enrolled between the fall of 1992 and the spring of
41


1993. The present study drew on data from 1,285 families including
819 EA families and 466 LA families (in its majority Mexican
Americans) from whom I had data on the variables I am studying.
Because gender differences in terms of stress, SS and coping
styles have been well documented in the literature (Matud, Ibanez,
Bethencourt, Marrero & Carballeira 2002; Dwyer & Cummings 2001;
Gonzales, Tein, Sandler & Friedman 2001) I only included women on
the study.
For a full description of all the demographics of the sample refer
to tables 2,5,8,11,14,17,20,23, 24,45,46,71 and 72 on appendix B.
Measures
Demographics: The following demographics were included:
Mother's annual income, Mother's educational level (measured in a
scale of 1 to 13 going from no education at all to doctoral degree),
Occupation or job (measured on a scale of 1-6 being 1 not at all to 6
full time), Number of children in the house, Average family size, Help
from Father (0= never, 1= some time, 2= most of the time). Father
presence (0= no, 1= Yes)
42


Social Support:
Social support was measured using The Family Resource Scale
(FRS). The FRS is designed to assess the types of resources and needs
which have been identified as major components of both intra-
personal and inter-personal support (Dunst & Leet, 1987). The
original scale is a 30-item self-report questionnaire. Each item is rated
on a five-point scale ranging from not at all adequate to almost always
adequate. Sample items include: "To what extent are the following
resources adequate for your family": "Food for two meals a day",
Money to buy necessities", "Time to get enough sleep/rest", "baby
sitting for your children", "child care for your children" etc. The scale
ranges from 0 to 150 with higher scores indicating higher levels of
support. Cronbach's coefficient alpha reported is .97, the split-half is
.95 and test retest reliability for the total scale is .70.
The total score of the scale can be used to measure social
support as a whole dimension. The scale can also be divided into three
subscales that measure the three different kinds of SS. 1) The
Instrumental or Tangible Support Subscale, which includes the
43


following items:
1,2,3,4,5,6,7,8,9,10,11,12,14,20,23,24,28,29,30,31,32,37; 2) The
Emotional Support subscale, which includes items 13,
15,16,17,18,19,21,22,25,26,27 and 35; And 3) The Informational and
Special Needs Subscale, which includes items: 33,34,36,38 and 39.
Stress:
Levels of stress were measured with the Parenting Distress
Index (Abidin, 1990). This 10-question self-report questionnaire
assesses perceived stress. The scale is based on a 5-point Likert-Type
scale that ranges from "Never" to "Very Often" Sample items include:
"In the last month how often have you felt nervous and "stressed""
and "In the last month how often have you been able to control
irritations in your life?" Scores range from 0 to 50 with higher scores
indicating higher levels of stress. Cronbachs coefficient alpha
reported is .90 and test re-test reliability for the total scale is .82.
Acculturation English Proficiency
To measure acculturation the host language (English) skills and
44


proficiency was measured with one question "to what extent are your
needs met for your English speaking skills" the item was rated on a
five-point scale ranging from 1 to 5 being 1 not at all adequate, 2
inadequate, 3 somewhat adequate, 4 adequate and 5 special strength.
The answers were then coded 1,2 & 3 low acculturated and 4&5
highly acculturated. Measures of reading skills are also available but
will not be included given that the data can be affected as many
subjects are analphabets. This measure of language proficiency has
some limitations that will be discussed later on in the discussion
section.
For the early head start program all the instruments were
adapted and standardized in Spanish.
Procedure
As stated earlier, the data for the present study is part of the data set
of the Head Start / Public School Early Childhood Transition
Demonstration Study 1991-1999 program. The data were collected from
1992 until 1997. The data were collected at 31 sites throughout 30 states
in the United States including the Navajo country. The measures were
45


taken through interviews given by the research assistants working at the
study. These interviews were conducted in the mothers native language.
Demographic data were collected at entry level and when the
children were 14 and 24 months old. Parental stress scores and level
of English were both measured when the children were 14 and 24
months old and SS level was obtained when the children were 15 and
26 months old. Half of the participants at the study were part of the
program and the other half part of the control group.
For the statistical analysis the LA families were subdivided into
two groups low acculturated and highly acculturated according to
their level of English. Therefore to compare stress at 14 months or SS
at 16 months the division of the LA group was done using the
participant's English level also measured when the children were 16
months old. In the same manner to compare the levels of stress at 24
months old and SS at 26 months old the participants' English level
score used was the one collected when the children were 26 months
old. For the statistical analysis an ANOVA was used to look at variance
among the groups. A multiple regression analysis was conducted to
control for demographic variables and try to explain the relation
46


between stress and SS. Finally Spearmans rho & Pearson correlation
analyses were calculated to look at the correlation between SS and
acculturation. For the correlations the five levels of acculturation
were included.
For the statistical analysis of social support both the total score
and the scores for the three subscales were used.
47


CHAPTER4
RESULTS
To find whether significant differences in terms of stress and
social support exist between EA and LA in general, and among low
acculturated LA (LALA), highly acculturated LA (HALA) and EA a
graphical analysis was conducted. First, I compared stress at 14 and
24 months and SS at 15 and 26 months using the FRS total score (See
figures 1-8 on Appendix A).
The information obtained from the graphical analysis, which uses
the median, shows that there seems to be statistically significant
differences among the medians of most of the groups being compared
with the least notable differences for stress levels at 24 months old. (For
descriptive statistics and processing data of the samples represented in
the figures, see tables on appendix B).
To corroborate if this was true, formal statistical tests were
conducted.
First I assessed if the samples had a normal distribution using the
Kolmogorov-Smirnov and the Shapiro- Wilk tests. I found that the null
48


hypothesis of normality of the distribution was rejected in almost all of
the cases at a significant level of 0.05. The only sample with a normal
distribution was the low acculturated LA sample for the variable of stress
at 24 months old (for test results see tables 3,9,12,15 & 18 of Appendix
B).
Since the majority of the samples did not have a normal distribution,
which is common for large samples, (Field, A., 2005) I used non-
parametric tests for the statistical analyses.
To observe if there were statistically significant differences between
the two main groups, EA and LA, a Mann-Whitney U and a Kolmogorov-
Smirnov tests, for independent samples were conducted. The results
from both tests showed statistically significant differences between EA
and LA for stress at 14 months old, and for Social Support at 15 and 26
months old but no statistically significant differences were found for
stress at 24 months old (see table 25 on Appendix B).
Afterwards an ANOVA was used to look for statistically significant
differences including degree of acculturation for the LA sample. The tests
were conducted for each of the four variables studied for the three
49


groups: low acculturated LA, Highly acculturated LA and EA. See Tables
26 & 27 on Appendix B.
Statistically significant differences were found for Stress at 14, SS at
15 and SS at 26 among the three groups. However, no statistically
significant differences were found for stress at 24 months.
Since there were no statistically significant differences in terms of
stress at 24 months among any of the groups, this variable was not
considered any further except for the correlational analysis.
Post-hoc test were conducted to look at which pairs were different
from one another. In order to use the appropriated tests a Levene
contrast was used. The Levene contrast showed no significant differences
of the variances for SS 15 and SS26. The variance was only significantly
different for Stressl4. (For detailed test results see Tables 28 & 32 on
Appendix B).
The multiple comparisons were done for stress 14, SS15 and SS26
for the 3 groups: LALA, HALA and EA. (For results see tables
29,30,31,33,34 on Appendix B).
The results from the above tests showed a statistically significant
difference between the LALA and EA, being the levels of stress higher for
50


the LALA than for EA. No significant differences were found between
LA LA and HA LA or between HALA and EA.
Statistically significant differences also were found in the levels of
SS between the LALA and the EA. SS was significantly higher at .032 and
.025 (with a significance level set at p < 0.05) for EA than for LALA. No
statistically significant differences were found between the other groups
neither at 15 months old nor at 26.
Next I conducted a multiple regression analysis to establish if some
of the variables explained more than others the differences in levels of
stress. The following variables were included: SS (FRS), mother's
educational level, mother's annual income, mother's occupation, number
of children living in the household, Average size of the family, help from
father (support from dad), presence of the father and acculturation
(English Skills).
The methodology used was to include all the variables and to
eliminate in each step the variables that were not statistically significant.
The multiple regression analysis was done for levels of Stress at 14m and
24m.
51


The results from the multiple regression analysis (for results, see
tables 35 & 36 on Appendix B) show that in both cases SS (FRS) is
statistically significant. For the levels of stress at 14 months old there are
also two other variables, mother's level of education and average size of
the family, that together with SS account for stress. For levels of stress 24,
SS (FRS26 and support from dad) does account for all of the variance and
was the only one statistically significant at the end of the regression. (For
the descriptive statistics see tables 45 & 46 of Appendix B).
Finally I ran a Pearson and a Spearman' rho correlation to explain
the relationship between SS and acculturation. The correlations
conducted for FRS15 and FRS26 using the full scale of acculturation 1-5.1
found a positive and statistically significant correlation between social
support and acculturation. I found a Pearson correlation coefficient of
r=.133 and a Spearman rho of r = .208 between FRS15 and acculturation
1LA (14 months), and a Pearson coefficient of r =.130 and a Spearman of
r= .152 between FRS26 and acculturation 2LA (24 months). The positive
correlation shows that as the level of acculturation increases the levels of
SS also increase. See the results on tables 37 through 44 of the Appendix
B.
52


Since different types of support seem to act differently for different
ethnic groups, the second part of the analysis was to look for significant
differences among the groups but using the three SS subscales: 1)
Instrumental Tangible, 2) Emotional and 3) Informational- Special
Needs rather that the FRS total score.
The graphical analysis of the subscales showed that there seems to
be statistically significant differences among the groups for instrumental
and informational support but not for emotional. (See Figures 9 trough 14
of the Appendix A, and tables 47 and 52 for the processing data and 71 &
72 for the descriptive statistics). From these analysis there seems to be
statistically significant differences between the median of instrumental
and informational support both at 14 and 24 months being the median
significantly higher for EA followed by HALA with the LALA at last.
The results from the analysis of variance both at 14 and 24
months showed statistically significant differences among the groups:
LALA, HALA and EA for instrumental- Tangible and Informational-
Special needs support. There were no statistically significant
differences among the groups for emotional support. (See tables 45 &
54 on Appendix B).
53


On the multiple comparison analysis by pairs at 15 months, I
found that the level of Instrumental -Tangible support was higher for
HALA than for LALA and EA; there was no significant difference
between LALA and EA. The Informational- Special Needs Support level
was also higher for HALA followed by EA being the lowest for LALA
(See table 51 on Appendix B for results).
On the multiple comparison analysis at 26 months (see table 56
on Appendix B) the results showed statistically significant differences
among groups. Instrumental support was higher for HALA followed by
LALA and EA at last. And, on the Informational- Special Needs scale
the level was higher for the HALA followed by the EA and the LALA at
last.
Finally the Pearson and Spearman rho correlation analyses
showed a positive and significant correlation between Instrumental-
Tangible and Informational- Special Needs supports and acculturation
at both 15 and 26 months. No significant correlation was found
between Emotional support and acculturation neither at 15 nor at 26
months. These results seem to suggest that instrumental and
informational support seem to increase as the acculturation level
54


increases.


CHAPTER 5
DISCUSSION AND CONCLUSIONS
As I assumed in the hypothesis I found that the higher the level of SS
the lower the level of stress in all individuals irrespective of their
ethnicity, demographics and degree of acculturation, which is consistent
with the findings described in the literature (Broadhead et al., 1983;
Fleming et al., 1982; Haines et al., 1996; Leavy, 1983; Mitchell et al., 1982;
Pengilly & Dowd, 2000).
1 expected to find higher levels of SS and lower levels of stress for all
LA than for EA. I also expected Low acculturated LA to have higher levels
of SS than highly acculturated LA and EA. Interestingly enough, when SS
was measured as a whole I found that: 1) SS was higher for the EA and the
HALA than for the LALA and 2) there was no significant difference
between EA and HALA or between LALA and HALA in terms of SS or
stress. When the three SS subscales were used for the analyses on the
other hand, significant differences were found among the groups. In this
case I found: 1) No statistically significant differences in terms of
emotional support among the three groups, 2) Statistically significant
56


differences for both Instrumental and informational support among the
groups, 3) Instrumental support levels were higher for the HALA at both
intervals, followed by the EA at 15 months, and the LALA at 26 months. 4)
The Informational- Special Needs support level was also higher for the
HALA than for the EA and for the EA than for the LALA both at 15 and 26
months.
Individuals from collectivistic societies did not show higher levels of
SS and lower levels of stress than individuals from individualistic
societies when SS was measured as a whole. When SS was divided by
types though, levels of support on the types that were significant:
instrumental and informational were higher for HALA at all times
followed by EA in three cases and LALA in one case. The importance and
weight of emotional support for SS as a unique dimension, and the lack of
a significant difference in emotional support among the groups could
explain the initial results. From this perspective dividing SS in types
proved to be very relevant for the study.
The fact that there are no significant differences in terms of stress at
24 months but only at 14 months can be explained by the LALA initial
lack of access to resources. As the LALA learned to get access to other
57


types of support, as we can see by the increase level on the instrumental
support. At 26 months LALA's level of SS as a whole increases and
therefore their level of stress decreases, not being significantly different
among the groups at 24 months.
These results are similar to those reported by Dressier & Bernal
(1982) who found that Puerto Ricans with the fewest psychosocial
resources (including SS) for coping reported the worst stress outcomes.
Griffith (1983), also found that Spanish-Speaking subjects had higher
scores of anxiety than the subjects on the English Speaking and bilingual
groups.
I can argue that the lower levels of SS as a whole and therefore
higher level of stress found on the LALA sample can be explained by the
lack of access to original resources. The lack of significant differences
found in terms of emotional support among groups can explain the lower
levels of support found in the LALA group. The LALA in particular are
probably far away from their extended and/or immediate families, which
constitutes their preferred source of support (Krzysztof & Norris, 2000).
As LALA get more acculturated they start to access other types of support
such as instrumental and informational to compensate for the lack of
58


access to emotional support and therefore their SS increases and their
stress decreases as seen with the levels reported by the HALA group.
The fact that the data were collected about 15 years ago also makes
the context different and can influence the results. A sample in a Latin-
American country will really help to look at LA and therefore collectivistic
societies in its original environment. If 1 take into consideration that
collectivistic societies in this case represented by the LALA and the HALA
groups tend to stress group solidarity, give priority to the group and rely
on family members (including the extended family) for help and advice,
(Briley & Wyer, 2001; Hui 1988; Hofstede 1991; Kim, et al, 1994. &
Triandis, 2001) then not having access to these resources can not only
limit the amount of SS they will endure but also increase the levels of
stress, even above those experienced by an individual who is not used to
having them. Individuals from individualistic societies like the EA do not
rely on family and relatives for help. On the contrary, they look at
themselves as independent beings, the emphasis is on autonomy,
emotional independence and privacy. Individual initiative,
competitiveness and independence are stressed and individual
achievement rewarded. Individualistic societies also tend to rely more on
59


friends than family. (Hofstede, 1980; Kim et al., 1994; Montaigne &
Bowen, 2000;Triandis, 1995 & Triadis, 2001). If I take this into
consideration SS is not needed by EA but if available is beneficial and
does result in lower levels of stress. In other words under similar
circumstances not having SS can be more detrimental to an individual
that needs it and is used to having it, than for an individual who does not
need it and is not counting on it.
The results obtained by the HALA group, which showed the highest
level of instrumental and informational support at all times and no
difference in terms of stress with the EA is in agreement with the
premises explained on the study. It shows that individuals from
collectivistic societies with access to SS have at least the same amount of
stress than individuals from individualistic societies such as the EA. It will
be interesting to replicate the study with a LA group from a different
socio-economical level and therefore exposed to a different stressor. A
reported by Sagrestano et al. (1999) the amount of SS varies according to
the individual's SES and is higher for individuals with a higher socio-
economical level when compared to individuals from a low SES.
If I look at the results from the stress models I could argue that SS
60


for the LALA works as the main effect model proposes as an overall
beneficial effect because having social networks provides them with a
regular positive experience and a socially rewarded role in the
community as well as a sense of predictability and stability. There is in
this case a direct relationship between SS and stress. SS acts as a
protective factor and reduces the physiological impact of stress (Cohen &
Wills, 1985; Schwarzer & Leppin, 1991). Not having that protective factor
for a low acculturated Latino can be stressful in itself. The effect of SS on
stress for the EA on the other hand maybe better explained by the
buffering effect model. In this case EA does not have a cultural need for SS
but if available it does reduce the impact of the stressful situation by
attenuating or preventing a stress appraisal response and by intervening
between the stressful event and the onset of pathological outcomes
(Cohen & Wills, 1985; House, 1981; La Rocca, House & French, 1980 &
Sheldon & Wills, 1985).
I found a positive rather than negative correlation between SS and
acculturation, meaning that as the acculturation increases the levels of SS
also increase. This could be explained by the lack of access to original
resources and the fact that the more acculturated the individual becomes
61


the more similar to the receiving culture he/she becomes. The individuals
that have just arrived to the United States have less access to
instrumental and informational support such as medical care, education,
access to federal resources etc. As seen on the results from the subscales,
the more acculturated the individual gets the easier it is to get
information and therefore access to instrumental and informational and
special needs support The LALA also left their social networks (family,
friends, contacts etc.) behind. As the individual becomes more
acculturated he/she increases his/her friends, contacts and access to
available resources.
There are some limitations to the present study that should be taken
in to consideration for future research. The first one makes reference to
the Latin-American sample. The low acculturated sample available was
small compared to the other two and therefore limits the spectrum of the
conclusions. Using a Latino sample in a foreign environment is also a
limitation. The access to their natural resources may be affected by the
geographical location and not being in their native countries might be
producing an additional stress that was not accounted for. Comparing a
sample of native EA in the United States with a sample of LA in their
62


original country will be very valuable to the field and will also help with
the limitations of what measuring acculturation brings, which is
discussed below.
First, a more extensive measure of English proficiency such as a
standardized test or other measure of English abilities will be ideal or
desirable.
Second, even though the ability to speak the language of the
receiving culture is considered the most accurate measure of
acculturation (Lopez & Contreras, 2005; Cuellar, Arnold, & Maldonado,
1995; Salgado de Snyder 1987; Neff, 1986; Warheit, Vega, Auth, &
Meinhardt, 1985; Ortiz & Arce, 1984; Griffith, 1983; Melville, 1978 (as
cited in Rogler, Cortes & Malgady, 1991) two other variables could be
taken into consideration. The first one is neighborhood composition,
which makes reference to the unequal distribution of groups across
space. People are residentially grouped across a number of dimensions,
including age, socioeconomic status, andmost relevant for this study
by ethnicity (White, 1987). Franco, Pottick & Huang (2010) found that
neighborhood context was significantly associated with parenting stress
and minority parents experience less stress than White parents in higher-
63


disordered neighborhoods.
It would be expected that less acculturated Latinos live in a
predominantly Latino neighborhood while more acculturated Latinos
would tend to live in a more diversely composed neighborhood. In
addition, living with extended family would also be considered as a
variable of acculturation. The Latino families place a very important role
on the extended family. The extended family is considered the primary
source of social contact and support for the nuclear family and it is
included and taken into consideration in child-upbringing (Levendecker
& Lamb 1999). It would be expected that less acculturated Latinos would
tend to live with extended family while more acculturated Latinos would
not. This is of course when possible many Latinos actually have to
immigrate to the United States on their own while leaving not only
extended but also nuclear family behind.
Another variables that should be taken into consideration, that were
not available in the data set, are place of birth and years that the
participant has been in the country. As mentioned earlier a new
environment can add stress that it is not being accounted for nor
measured.
64


A limitation of the sample is that the LA sample is formed in its
majority by Mexican Americans and therefore the Mexican Americans are
being considered as representative of LA but do not represent the entire
LA population.
This study constitutes a step forward in understanding ethnic
differences in terms of stress and SS between individualistic and
collectivistic societies and does show the existence of variations among
groups. However, more research needs to be conducted before further
recommendations for prevention and/or treatment can be inferred.
65


Stress14
APPENDIX A
Figure 1. Stress at 14 months for EA and LA
mt; nrst udi represents uie enure sample
66


Stress14
Figure 2. Stress at 14 months for all groups
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA


Figure 3. Stress at 24 months for EA and LA
50.00-
40.00-
s
8
2
Si
30.00-
20.00-
10.00-
0.00-
r~
EA
I
LA
Ethnicityl
The first bar represents the entire sample
68


Figure 4. Stress at 24 months for all groups
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample
69


Social_Support_15
Figure 5. Social support at 15 months for EA and LA
EA
Ethnicityl
LA
The first bar represents the entire sample
70


Social_Support_15
Figure 6. Social support at 15 months for all groups
140.00-
120.00
100.00-
80.00-
60.00-
40.00
20.00-
0.00-
T----------------1----------------1
E.4 HALA
English_Skills15
The first bar represents the entire sample
i
LALA
71


Social_Support_26
Figure 7. Social support at 26 months for EA and LA
The first bar represents the entire sample
72


Social_Support_26
Figure 8. Social support at 26 months for all groups
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample and the numbers the atypical
cases.
73


Figure 9. Instrumental support at 15 months for all groups
English_Skills15
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample and the numbers the atypical
cases.
74


Figure 10. Emotional support at 15 months for all groups
English_Skills15
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample and the numbers the atypical
cases.
75


Figure 11. Informational and special needs support at 15 months for all
groups
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample and the numbers the atypical
cases.
76


Figure 12. Instrumental support at 26 months for all groups
120.00-
w 100.00-
t:
o
a.
a.
3
CO

n
u
c
ca
3
80.00-
60.00-
40.00-
~T7l 14
1^7;
1,223
1,108
O
20.00-
o.oo-
1,218
1,198
1,262
*1.261
2,0772,060
*
2,075
I----------------1
EA HALA
English_Skills26
LA LA
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample and the numbers the atypical
cases.
77


Figure 13. Emotional support at 26 months for all groups
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample and the numbers the atypical
cases.
78


Figure 14. Informational and special needs support at 26 months for all
groups
25.00"
5
s.
a.
5
o
0
20.00-
15.00-
u
*
wl ^
o 10-00-
c
"l
0
c
o
£
c
5.00-
0.00-
1,218
01.262
1,261 1,198
------1------
-j1,592
2,086
2,0602,075

2,084
------1----
EA
818923
PB11
O
965 878
* 1,108
T
HALA
English_Skills26
LALA
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample and the numbers the atypical
cases.
79


APPENDIX B
Table 1. Case processing data for stress at 14 months for LA and EA
Case Processing Summary
Elhnicilyl Cases
Valid Mssing Total
N Percent N Percent N Percent
Stress 14 697 06.2% 112 13.8% 809 100.0%
EA 725 88.5% 94 11.5% 819 100.0%
LA 397 85.2% 69 14.8% 466 100.0%
80


Table 2. Descriptive Statistics for stress at 14 months for EA and LA
Descriptive*
Ethnicityl Stafsfc StJ. Error
Stress 14 Mean 24.2541 .42935
95% Confidence Interval Lower Bound 23.4111
for Mean
Upper Bound 25.0971
5% Trimmed Mean 24.2870
Median 24.0000
Variance 128.487
Sid. Deviation 11.33520
Minimum 2.00
Maximum 49.00
Range 47.00
Interquartle Range 15.00
Skewness -.163 .093
Kurins is -.361 .185
EA Mean 23.9792 .38031
95% Confidence Interval Lower Bound 23.2326
for Mean
Upper Bound 24.7258
5% Trimmed Mean 23.9953
Median 24.0000
Variance 104.858
Sid. Deviation 10.24004
Minimum 2.00
Maximum 49.00
Range 47.00
Interquartle Range 1300
Skewness -.103 .091
Kudos is 058 .181
LA Mean 25.2658 .57915
95% Confidence Interval Lower Bound 241272
lor Mean
Upper Bound 26.4044
5% Trimmed Mean 25.3477
Median 25.0000
Variance 133.158
Std. Deviation 11.53940
Minimum 2.00
Maximum 49.00
Range 47 00
Interquartle Range 16.00
Skewness -.175 .122
Kudos is -.371 .244
81


Table 3. Test of normality for stress 14 for LA and EA
Tests of Normality
Elhnicityl Kolmogorov-Sm irnoV Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Stress 14 .071 697 .000 .974 697 .000
EA .055 725 .000 .979 725 .000
LA .061 307 .001 .974 397 .000
a. Lilliefors Significance Correction
82


Tabla 4. Case processing data for stress at 14 months for all groups
Case Processing Summary
English_Skills15 Cases
Valid Missing Total
N Percent N Percent N Percent
Stress 14 740 85.1% 130 14.9% 870 100.0%
EA 725 88.5% 94 11.5% 819 100.0%
HALA 258 88.7% 33 11.3% 291 100.0%
LALA 96 84.2% 18 15.8% 114 100.0%
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
83


Table 5. Descriptive statistics for stress at 14 months for all groups
Dhct^i
English SMIIs15 Stelsflc Stl.Error
Stress 14 Mean 24.4107 .41762
95% Contdence Interval Lower Bound 235908

Upper Bound 252306
5% Trimmed Mean 24.4505
Median 24.5000
Variance 129.062
SU.Deviaion 11.36052
Mnimum 2.00
Ma4mum 49-00
Range 47.00
Intorqua/ile Range 15.00
Skewness -.162 .090
Kurtosis -.357 .179
EA Mean 23.9792 .38031
95% Confdence htervel Lower Bound 23-2326

Upper Bound 24.7258
5% Trimmed Mean 23.9953
Median 24.0000
Wriance 104.858
Ski. Deviafon 1024004
Mnimum 2.00
Mumum 49.00
Range 47.00
taterquarfle Range 13.00
Skewness -.103 .091
Kurtoeie .058 .181
MALA Mean 24.2741 .70469
95% Con fidence In larval Lower Bound 22.8864

Upper Bound 25.6618
5% Trimmed Mean 24.3080
Median 24.0000
Variance 128.120
Sto. DeviAfon 11.31899
Mnimum 2.00
Maitmum 49.00
Range 47.00
Inlerquarfle Range 14.00
Skewneee -.163 .152
Kurtosis -.315 .302
LALA Mean 27.1771 121314
95% Contdence kiterval Lower Bound 24.7687

Upper Bound 29-6866
5% Trimmed Mean 27.4074
Median 275000
Viriance 141.284
Sid. Deviaion 11.88630
Mnimum 2.00
Mnimum 49.00
Range 47.00
Intorquartle Range 17.00
Skewness -.271 246
Kurtosis -.454 .488
84


Table 6. Test of Normality for stress at 14 months for all groups
Tests of Mormality
English Skills15 Kolmogorov-Sm irnoV Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Stress 14 .069 740 .000 .974 740 .000
EA .055 725 .000 .979 725 .000
HALA .072 258 .003 .973 258 .000
LALA .063 96 200 .971 96 .032
a. Lillicfbrs Significance Correction
*. This is a lower bound of the true significance.
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
85


Table 7. Case processing data for stress at 24 months for EA and LA
Case Processing Summary
Elhnicityl Cases
Valid Mssing Total
N Percent N Percent N Percent
Stress24 663 82.0% 146 18.0% 809 100.0%
EA 713 87.1% 106 12.9% 819 100.0%
LA 400 85.8% 66 14.2% 466 100.0%
86


Table 8. Descriptive statistics for stress at 24 months for EA and LA
Descriptive*
Ethnicityl Statistic Std. Error
Stress24 Mean 23.4963 .39761
95% Confidence Interval Lower Bound 22.7156
for Mean
Upper Bound 24.2771
5% Trimmed Mean 23.5604
Median 24.0000
Variance 104.818
Std. Delation 10.23805
Mnimum 2.00
Maximum 48.00
Range 46.00
Interquartile Range 15.00
Skewness -.093 .095
Kurtosis -.352 .190
EA Mean 22.4591 .37391
95% Confidence Interval Lower Bound 21.7250
for Mean
Upper Bound 23.1932
5% Trimmed Mean 22.4313
Median 23.0000
Variance 99.683
Std. Delation 9.98412
Mnimum 2.00
Maidmum 48.00
Range 46.00
Interquartile Range 13.00
Skewness .012 .092
Kurtosis -.123 .183
LA Mean 23.4254 .55012
95% Confidence Interval Lower Bound 22.3439
for Mean
Upper Bound 24.5069
5% Trimmed Mean 23.3365
Median 24.0000
Variance 121.053
Std. Delation 11.00241
Mnimum 2.00
Maxmum 49.00
Range 47.00
Interquartile Range 15.00
Skewness .052 .122
Kurtosis -.352 243
87


Table 9. Normality tests for stress at 24 months for EA and LA
Tests of Normality
Ethnidtyl Kolmogorov-Sm imov Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Stress24 .054 663 .000 .982 663 .000
EA .067 713 .000 .980 713 .000
LA .062 400 .001 .982 400 .000
a. Lilliefors Significance Correction
88


Table 10. Case processing data for stress at 24 months for all groups
Case Processing Summary
English_Skills26 Cases
Valid Missing Total
N Percent N Percent N Percent
Stress24 706 81.7% 158 18.3% 864 100.0%
EA 713 87.1% 106 12.9% 819 100.0%
HALA 269 87.1% 40 12.9% 309 100.0%
LALA 88 86.3% 14 13.7% 102 100.0%
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
89


Table 11. Descriptive statistics for stress at 24 months for all groups
Englteh_SkMs26 Stefsfc Sid. Error
S*ess24 Mot 23.5963 .38737
96% ConMence MotbI Lower Bound 223358

Upper Bound 243568
5% Trimmed Nban 23.6493
Median 243000
Variance 105339
Ski. Deviation 1029267
Mnimum 2.00
Iteimum 46.00
Range 46.00
InjerquarWe Range 15.00
Skewness -.067 .092
Kurtoet* -332 .1B4
EA Mean 22.4591 37391
95% ConMenoe interval Lower Bound 21.7250

UpperBound 23.1932
5% Trimmed Men 22.4313
Median 23.0000
Variance 99.663
Ski Deviation 936412
Mnimum 2.00
Merimum 46.00
Range 46.00
interquarfle Range 13.00
Skewness 312 .092
Kurtosks -.123 .163
HALA Mot 23.0635 37446
95% ConMenoe Marvel Lower Bound 21.7355

UpperBound 243914
5% Trimmed Mean 223710
Median 24.0000
Variance 122373
Std. Deviatkin 1136225
Mnhnum 2.00
Me4mum 49.00
Range 47.00
MerquariMe Range 17.00
Skewness 354 .149
Kurtosis -.445 .296
LALA ktear 23.6952 1.15127
95% ConMenoe interval Lower Bound 214070

Upper Bound 25.9635
5% Trimmed Mean 233210
tfedlan 243000
Wrtance 116337
Sid. Deviation 10.79986
Mnimum 2.00
Merimum 46.00
Range 46.00
Merquarfle Range 11.75
Skewness -.020 357
Kurtosis -342 306
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
90


Table 12. Normality tests for stress at 24 months for all groups
Tests of Normality
English_Skills26 Kolmogorov-Smimov Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Stress24 .055 706 .000 .983 706 .000
EA .067 713 .000 .980 713 .000
HALA .075 269 .001 .978 269 .000
LALA .096 88 .045 .973 88 .065
a. Lilliefbrs Significance Correction
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
91


Table 13. Case processing data for Social Support at 15 months for EA and
LA
Case Processing Summary
Ethnidtyl Cases
Valid Missing Total
N Percent N Percent N Percent
Social_Support_15 704 87.0% 105 13.0% 809 100.0%
EA 721 88.0% 98 12.0% 819 100.0%
LA 405 86.9% 61 13.1% 466 100.0%
92


Table 14. Descriptive statistics for social support at 15 months for EA and
LA
Descripthres
Elhnici VI Statistic Sid. Error
Sodal_Support_15 Mean 90.7273 1.82686
95% Confidence Interval Lower Bound 87.1405
for Mean
Upper Bound 94.3140
5% Trimmed Mean 92.4905
Median 115.5000
Variance 2349.549
Std. Delation 48.47215
Minimum 1.00
Maamum 147.00
Range 146.00
Interquartile Range 106.00
Skewness -.660 .092
Kudos is -1.014 .184
EA Mean 94.2150 1.79053
95% Confidence Interval Lower Bound 90.6997
for Mean
Upper Bound 97.7303
5% Trimmed Mean 96.2737
MBdian 117.5000
Variance 2311.529
Std. Deviation 48.07837
Mnimum 1.00
Maamum 149.00
Range 148.00
Interquartile Range 86.00
Skewness -.970 .091
Kurtosis -.826 .182
LA Mean 87.9988 2.31047
95% Confidence Interval Lower Bound 83.4567
tor Mean
UpperBound 92.5408
5% Trimmed Mean 89.5772
Madian 111.5000
Variance 2162.004
Std. Deviation 46.49736
Mnimum 1.00
Maamum 148.00
Range 147.00
Interquartile Range 104.25
Skewness -.783 .121
Kurtosis -1.034 .242
93


Table 15. Normality tests for social support at 15 months for EA and LA
Tests of Normality
Ethnicityl Kolmogorov-Smirnov* Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Sodal_Support_15 .283 704 .000 .761 704 .000
EA .297 721 .000 .741 721 .000
LA .213 405 .000 .803 405 .000
a. Lilliefors Significance Correction
94


Table 16. Case processing data for social support at 15 months for all
groups
Case Processing Summary
English_Sk!lls15 Cases
Valid Missing Total
N Percent N Percent N Percent
Sodal_Support_15 704 80.9% 166 19.1% 870 100.0%
EA 721 88.0% 98 12.0% 819 100.0%
HALA 291 100.0% 0 .0% 291 100.0%
LALA 114 100.0% 0 .0% 114 100.0%
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
The first bar represents the entire sample
95


Table 17. Descriptive statistics for social support at 15 months for all
groups
PBicrlplhM
Engllh_Sklls15 Statistic Std. Error
SodaJ_Support_15 Mean 90.7273 132686
05% Confidence Interval Lower Bound 87.1405

Upper Bound 94.3140
5% Trimmed Mean 92.4905
Median 1153000
Variance 2349349
Ski. Deviation 48.47215
Mnimum 130
Itaimum 147.00
Range 14630
taterquartile Range 10630
Skewness -.860 .092
Kurtoeie -1.014 .184
EA Mean 943150 1.79053
96% Contidenoe Interval Lower Bound 903997

Upper Bound 97.7303
5% Trimmed Mean 963737
Median 1173000
Variance 2311.529
Std. Deviation 4837837
Mnimum 130
htoimum 14930
Range 148.00
Interquartile Range 86.00
Skewness -.970 .091
Kurtoeie -.826 .182
HALA Mean 902818 2.75980
95% Confidence interval Lower Bound 843500

Upper Bound 95.7136
5% Trimmed Mean 923582
Median 1133000
^riance 2216.398
Sto. Deviation 47.07664
Mnimum 130
Mnimum 148.00
Range 147.00
toterquartite Range 107.00
Skewness -.647 .143
Kurtoeie -.966 285
LALA Mean 82.1711 4.18210
95% Contidence Interval Lower Bound 733856

Upper Bound 90.4566
5% Trimmed Mean 832529
Median 962500
Vsionco 1993.858
Std. Deviation 44.65263
Mnimum 130
Merfmum 14130
Range 140.00
hterquartile Range 99.13
Skewness -.670 226
Kurtosis -1.136 .449
96


Table 18. Normality tests for social support at 15 months for all groups
Teats of Normality
English_Skills15 Kolm ogorov-Sm imov* Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Social_Support_15 283 704 .000 .761 704 .000
EA 297 721 .000 .741 721 .000
HALA 247 291 .000 .785 291 .000
LALA 223 114 .000 .826 114 .000
a. Lilliefors Significance Correction
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
Table 19. Case processing data at 26 months for EA and LA
Case Processing Summary
Ethnicityl Cases
Valid Missing Total
N Percent N Percent N Percent
Sodal_Support_26 711 87.9% 98 12.1% 809 100.0%
EA 702 85.7% 117 14.3% 819 100.0%
LA 411 88.2% 55 11.8% 466 100.0%
97


Table 20. Descriptive statistics for social support at 26 months for EA and
LA
Descriptives
Ethnicityl Statistic Std. Error
Sodal_Support_26 Mean 96.5373 1.72281
95% Confidence Interval Lower Bound 93.1549
lor Mean
Upper Bound 99.9197
5% Trimmed Mean 98.8264
Median 117.0000
Variance 2110.306
Std. Delation 45.93807
Mnimum 1.00
Maidmum 148.00
Range 147.00
Interquartile Range 34.00
Skewness -1.128 .092
Kurtosis -.445 .183
EA Mean 97.1360 1.78938
95% Confidence Interval Lower Bound 93.6229
fbrMsan
Upper Bound 100.6492
5% Trimmed Mean 99.4529
Median 1185000
Variance 2247.724
Sid. Delation 47.41016
Mnimum 1.00
Maidmum 149.50
Range 14850
Interquartile Range 40.63
Skewness -1.075 .092
Kurtosis -588 .184
LA MBan 895710 2.34734
95% Confidence Interval Lower Bound 85.2567
for Mean
Upper Bound 94.4854
5% Trimmed Mean 915931
Median 1135000
Variance 2264.609
Sid. DeVation 47.58791
Mnimum 1.00
IMmum 147.00
Range 146.00
Interquartile Range 105.50
Skewness -.844 .120
Kurtosis -563 240
98


Table 21. Normality tests for social support at 26 months for EA and LA
Tests of Normality
Elhnidlyl Kolmogorov-SmirnoV Shapino-Wilk
Statistic df Sig. Statistic df Sig.
Sodal_Support_26 .309 711 .000 .732 711 .000
EA .302 702 .000 .733 702 .000
LA .251 411 .000 .788 411 .000
a. Lilllefors Significance Correction
99


Table 22. Case processing data for social support at 26 months for all
groups
Case Processing Summary
English_Skills26 Cases
Valid Mssing Total
N Percent N Percent N Percent
Sodal_Support_26 711 82.3% 153 17.7% 864 100.0%
EA 702 85.7% 117 14.3% 819 100.0%
HALA 309 100.0% 0 .0% 309 100.0%
LALA 102 100.0% 0 .0% 102 100.0%
European Americans = EA, Highly acculturated LA = HALA and Low
acculturated LA=LALA
100