THE CLINICAL UTILITY OF THE WASI-II AND ITS ASSOCIATION WITH
ACCULTURATION AMONG ARAB AMERICAN ADOLESCENT MALES
By
Ramzi Hasson

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
School Psychology- Doctor of Philosophy
2015

ABSTRACT
THE CLINICAL UTILITY OF THE WASI-II AND ITS ASSOCIATION WITH
ACCULTURATION AMONG ARAB AMERICAN ADOLESCENT MALES
By
Ramzi Hasson
Neuropsychologists attempting to provide an ethical, and clinically useful assessment of
intellectual capability among immigrant youth face great challenges. This is because young
immigrants undergo an acculturation process that has a profound effect on cognitive
development, and the reliability and validity of norm-referenced intelligence tests. This study
explored the clinical utility of the WASI-II among male adolecent Arab Americans (n = 80). It
also explored the association between proxy and systematic acculturation factors with verbal
and language-reduced IQ performances. The possible moderating influence of accutluration on
the predicitive utility of IQ for academic outcomes was also examined. Results showed that
proxy acculturation variables were not associated with WASI-II outcomes. Results showed lower
verbal than language-reduce IQs, but that difference occurred in the context of signficant
variability within and between each IQ index. The difference between each IQ index was mostly
associated with sociodemographic factors. English language competence was associated with
performance on Vocabulary. No acculturation variable was assocaited with language-reduced IQ
after controlling for parent income. Performance on Matrix Reasoning was not influenced by any
sociodemographic or acculturation factors. Estimated Full-Scale IQ was the single best predictor
of basic reading and math skills, which had moderate level association. Acculturation did not
moderate the predicitive association between estimated FSIQ and academic outcomes. Important
clinical and research implications, and limitations were outlined.

ii

ACKNOWLEDGEMENTS

This work is dedicated to those who played an important role in the development and
completion of this dissertation. I’m indebted and deeply grateful to the insight and support of Dr.
Jodene Goldenring Fine. She provided me with furtile ground to develop my passion for
neuropsychology and cultivated scholarly foundations that will serve me a lifetime. I want to
extend my gratitude to my guidance and dissertation committee for their support, instruction,
patients and for simply believing in me. Thank you, Drs. John Carlson, Evelyn Oka, and Cary
Roseth. A special thanks goes to Dr. Andy Pham for his enduring mentroship that began on my
first day as graduate student and continued until the waneing moments of my graduate education.
This project would not have been completed without the support from the two vital community
organizations that are Yemeni American Benevelont Association (YABA), and ACCESS
Community Center for providing support, accesability, and assistance that made recrutiment and
assessment a smooth and efficient process.
I want to dedicate this work to my dear sister, Samira, who represents the best
actualization of what a good, honest, compassionate, hard working, and loving human being can
be. This work will also not be complete without the unwavering support of my dear brothers,
Kassem, Ibrahim, and Gammal. Thank you to all my 13 nieces and nephews (from oldest to
youngest): Remsh, Siham, Ilham, Reem, Khalid, Badr, Tareq, Hashem, Yousef, Ziyad, Majid,
Rana, and Mohamed. This work is also dedicated to my mother and father for constantly pushing
me and not allowing me to forget where I came from. Finally, I want to dedicate this dissertation
to my loving and supportive wife, “sweet” Suffa, my external executive function.

iii

TABLE OF CONTENTS
LIST OF TABLES

vii

LIST OF FIGURES

ix

CHAPTER I: INTRODUCTION

1

CHAPTER II: LITERATURE REVIEW
What is Intelligence?
The Utility of IQ Assessment
Race, Culture, and Intellectual Assessment
Universalist’s Approach
Environmentalist’s Approach
Acculturation and Cognitive Development
The Acculturation Process
Models of Acculturation
Assessing Acculturation
Acculturation of Arab Americans
Key Demographics of Arab Americans
Settlement Dynamics
Arab American Family Dynamics
The Cultural Milieu of Cognitive Assessment
Deconstructing Bias in Assessment
The Influence of Acculturation on Cognitive Performance
Addressing Challenges in the Intellectual Assessment of Immigrant Youth
Ethnic Specific Norms
Language-Reduced and Nonverbal Assessment
Culture-Language Interpretative Matrix (C-LIM)
Cognitive Assessment with Arab Children
Ethical Obligations and Multicultural Competency
Rationale and Aims of the Proposed Study

10
10
11
15
16
20
23
23
25
28
29
32
34
37
38
39
41
44
45
46
47
52
53
55

CHAPTER III: METHODS
Population of Interest
Youth
Parents
Exclusion criteria
Recruitment and Assessment
Variables of Interest and Materials
Demographic variables
Guardian education level
Income level
Intellectual assessment

57
57
57
58
60
60
61
61
61
62
62
iv

Measure of acculturation
Academic achievement
Data Analysis
Power analysis
Three Study Aims, Questions, Rationales, and Analyses
Aim One
Aim Two
Aim Three

64
66
66
67
67
67
68
69

CHAPTER IV: RESULTS
Summary of Purpose
Sample Characteristics
Acculturation levels
Relation between acculturation and sociodemographic variables
Relation between age, country of origin, and main outcome measures
Academic outcomes
WASI-II performances
Aim One Outcomes
Aim Two Outcomes
Aim Three Outcomes
Exploratory Analyses
Subtests and academic outcomes
US-AMAS scales relation to WASI-II subtest scores
First model: vocabulary subtest
Second model: similarities subtest
Third model: matrix reasoning subtest
Forth model: block design subtest
Mother education and WASI-II performances

71
71
71
76
77
78
78
79
80
82
87
89
89
90
90
91
91
91
92

CHAPTER V: DISCUSSION
Purpose of the Study
Discussion of Findings
Aim one
Aim two
Aim three
Integrative Summary
Study Implications and Limitations
Clinical implications
Research implications
Study limitations
Concluding Remarks

94
94
95
95
100
105
106
108
108
110
111
113

APPENDICES
APPENDIX A: PARENT DEMOGRAPHIC QUESTIONNAIRE
APPENDIX B: ADOLESCENT DEMOGRAPHIC SURVEY
APPENDIX C: ACCULTURATION SCALE: AMAS

115
116
118
120

v

APPENDIX D: PARENT INFORMED CONSENT
APPENDIX E: CHILD ASSENT FORM
BIBLIOGRAPHY

123
126
128

vi

LIST OF TABLES

Table 1 Berry’s Bidimensional Model of Acculturation

27

Table 2 Culture-Language Interpretation Matrix

49

Table 3 Country of Birth for Adolescents and Parents

57

Table 4 Demographic Characteristics of Adolescent Participants

58

Table 5 Parents’ Education, Employment, and Income

59

Table 6 Normative Sample Correlations between the WASI-II and
Wechsler’s Academic Outcomes

64

Table 7 Internal Consistency of AMAS with Arab Americans
as Measured by Cronbach’s Alpha

65

Table 8 Intercorrelations of Study Variables, with Means, Standard Deviation,
Kurtosis, and Skewness

73

Table 9 Intercorrelation of Acculturation Subscales, Sociodemographic, and IQ

74

Table 10 Intercorrelations among WASI-II and AMAS Scales

75

Table 11 Internal Consistency, Means, and Range of AMAS
with Arab Americans

77

Table 12 Difference between Obtained and Normative Means

80

Table 13 Classification of Research Sample According to Normative Standards

80

Table 14 Cohen’s d Values from Pair-Wise T-test Comparison of WASI-II Subtests

82

Table 15 Summary of Hierarchical Regression Analysis for Variables Predicting V-IQ

85

Table 16 Summary of Hierarchical Regression Analysis for Variables Predicting LR-IQ 87
Table 17 Summary of Hierarchical Regression Analysis for Variables Predicting
Word Reading

88

Table 18 Summary of Hierarchical Regression Analysis for Variables
Predicting Numerical Operation

89

Table 19 Study and Normative Sample Correlations between the WASI-II
vii

and Wechsler’s Academic Outcomes

89

Table 20 Multiple Regression Results for US-AMAS and WASI-II Subscales

viii

92

LIST OF FIGURES

Figure 1 Cognitive Results as a Function of High and Low Mother Education

93

Figure 2 Sample’s Subtest Performances in Comparison to Normative Outcomes
and C-LIM

96

ix

CHAPTER I: INTRODUCTION
The flood of immigrants to the United States (US) over the past two decades is only
matched by the great immigration waves of the early 20th century (Portes & Rumbaut, 2006).
The new wave of immigration is occurring in the contexts of a substantial demographic shift.
Currently, over 50 percent of school age population in the US is comprised of ethnic minorities
(National Center for Education Statistics, 2009). By 2020, first and second generation immigrant
students are projected to make up approximately 40 percent of public school students (Portes &
Rumbaut, 2006). The factions of ethnic and racial minority groups shape a population mosaic
that is incredibly diverse.
Neuropsychologists are sometimes asked to assess students from a wide range of cultural,
linguistic, and ethnic backgrounds, for a multitude of reasons, using standardized normreferenced tests. The assessment of intelligence is one of the most critical components of a
comprehensive neuropsychological evaluation and it provides a valuable metric for evaluating
the learning potential of students, the integrity of neurocognitive functioning, and in clarifying
differential psychological and neurodevelopmental diagnoses (Sattler, 2008; Lezak, Howieson,
Bigler, & Tranel, 2012). Neuropsychologists administer norm referenced and standardized tests
to immigrant youth. If individuals performed significantly worse than the normative mean then
such performances can be interpreted as reflecting brain pathology (Nell, 2000; Lezak et al.,
2012). However there are a number of factors that can lead to poor performance that do not stem
from brain pathology. For instance effort, fatigue, and pain are common factors
neuropsychologist try to rule out (Tombaugh, 1996; van der Linden, Frese, & Meijman, 2003;
Grigsby, Soenberg, & Busenbark, 1995; Brickman, Cabo, & Manly, 2006). Culture and language
factors also influence performance on intelligence tests (Nell, 2000). Unlike other non-target

1

factors such as effort level, there are no clear ways of understanding how cultural background or
varying bilingual language competence can influence performance on intelligence tests. Cultural
and linguistic experiences influence cognitive development and behavior, but research
examining the role of cultural and language factors on intelligence test remains scarce
(Kennepohl, 1999; Mindt, Burd, Saez, & Manly 2010).
The limited available multicultural assessment research with immigrants has been done
with larger ethnic groups like Hispanic (e.g., Nieves-Brull, 2006) and Asian Americans (Razani
et al., 2007) without careful analysis of how individual and group level factors relate to
assessment outcomes. Further, there is very little research available with smaller ethnic minority
groups that possess unique and salient cultural and language factors. One of those newly arriving
group of immigrants to the US are Arab Americans. There are important cultural and
demographic attributes of Arab Americans that neuropsychologists need to be mindful of when
providing evaluating Arab American youth. Examining how those factors influence intellectual
assessment will provide empirically based guidance to enhance multicultural assessment
competence which is currently non-existent.
As with most ethnic groups in the US, there is substantial heterogeneity within Arab
Americans. Arab Americans are those who may have voluntarily or involuntarily immigrated
from any of the 22 Arab speaking countries (Gregg, 2005). It is not clear what the exact number
of Arab Americans residing in the US is, but estimates show a range of 1.5-3.5 million (AAI,
2009; Brittingham & de la Cruz, 2005). Arab Americans immigrated to the US in three major
waves that span over 100 years. Each wave consisted of immigrants with diverse histories
(Wave-1: 1880-1920; Wave-2: 1938-1958; Wave-3: 1965-current; Gregg, 2005). The first wave
Arab Americans were mostly Christian voluntary immigrants with light skin, and were well

2

educated. Arab American immigrants from the second wave were comprised of volunteer
immigrants that resembled those of the first wave, but also included a substantial number of
Muslim Arab Americans who were fleeing from conflict and instability in their homelands and
sought refugee status. The third wave immigrants were mostly Muslim Arab Americans, who
have less education, and many came because of political conflicts and economic instability in
their homeland (Gregg, 2005). These three waves resulted in experiences that varied depending
when these immigrants came, where they came from, and where they settled. As a result, these
unique acculturative experiences may likely have specific associations with how well a particular
youth may perform on measures of intelligence.
The challenges associated with assessing immigrant youth are wide ranging. Young
immigrants may be developing cultural and language competencies across English and their
native language simultaneously. It has been argued that culture influences the content of tests,
the testing process, frames of thinking, and the development of the neural underpinning of
cognition (Nell, 2000; Ardila, 2005; Park & Huang, 2010). Culture involves the development of
learned traditions and living styles that are shared by a particular group of society resulting in the
internal representations of knowledge, values, beliefs, and modes of behavior (Ardila, 2005). It
follows that the acculturation process is likely to influence how knowledge is acquired, stored,
and expressed.
Acculturation is a process of cultural and psychological change that results from when
individuals interact across multiple cultures (Sam & Berry, 2010). Conceptually, acculturation
was viewed as a unidimensional process where an individual sheds his culture of origin at the
expense of assimilating to the new host culture (Sam & Berry, 2010). It is now accepted that
acculturation is best viewed as a bidimensional process that involves simultaneous maintenance

3

of the culture of origin and assimilation to the host culture (Berry et al., 2005). Others have also
pushed for evaluating specific bidimensional acculturative levels across specific domain within
the host and heritage cultures (Rudmin, 2009). For instance, a person can be bilingual and may
only identify with the heritage culture, but is much more culturally competent within the host
culture. Ultimately, characteristics of the acculturating individual (e.g., acculturation levels,
country of origin, age, gender, education background, and SES), testing process (e.g., individual,
impersonal, specific communication requirements), and those of the test (e.g., translated tests,
normative sample and psychometrics) are factors to consider whenever assessing an immigrant
youth. These variables complicate the interpretation of performances across diverse cultural
groups.
The best available IQ tests are normed in accordance with mostly general US population
census demographic factors, such as age, gender, parental education level, geographic region,
and racial or ethnic backgrounds (Sattler, 2008). Most immigrant youth are not adequately
represented by these general normative classifications. This is significant because it violates the
assumption of comparability of obtained scores with available norms. Namely, acculturating
youth are less likely to meet developmentally appropriate expectations of cognitive, cultural, and
linguistic knowledge acquisition that is directly assessed by the most widely used intelligence
tests (Flanagan & Harrison, 2012; Flanagan & Ortiz, 2007). Even the inclusion of those from
Asian and Hispanic backgrounds in some normative samples does not include the variability of
the immigrant experience associated with their unique acculturation experiences (Ardila, 2005;
Nell, 2000). It has also not been investigated whether there are causal associations between
levels of intellectual capability and adoption of particular acculturation orientations. Questions
such as, “Is it possible that those with higher intelligence are likely to adopt or assimilate at a

4

faster rate than those with a lower IQ?” cannot be examined until our basic understanding of how
acculturation patterns relate to performances on intelligence tests is better developed.
Acculturation is a dynamic process influenced by the level of cultural exposure, level of
linguistic competence, and the development of a cultural identity (Berry et al., 2003).
Additionally, the content of an intelligence test is shaped by a unique cultural framework
and experiences. The development of intellectual tests occurred mostly in the U.S., and to a
lesser extent in Europe (Boake, 2002). As a result, the conception of intelligence and how it is
assessed reflects the values and experiences associated with Western culture (Neisser et al.,
1996, Nell, 2000). Those from non-Western backgrounds may not share or be familiar with the
culturally implicit Western values, and with the modes of perception, reasoning, and knowledge
acquisition involved with individual standardized testing (Nell, 2000; Park & Huang, 2010). A
lack of exposure to the values and content of the test, as well as unfamiliarity with the test taking
process, or what has been referred to as test wiseness, has been documented to attenuate overall
IQ performance (Ardila, 2005; Anastasi, 1982; Flanagan & Harrison, 2012; Razani et al., 2007).
Consequently, the results from IQ assessments with immigrant youth are possibly more
reflective of cultural and linguistic competence than intellectual capability.
It is often assumed that if language demand is controlled, then an equitable and valid
assessment is possible (Rhodes et al., 2005). This is often accomplished through the use of
interpreters and interpreted tests, or the use of language-reduced tests with immigrant youth
(Flanagan & Harrison, 2012). Language-reduced or nonverbal tests do not demand any verbal
language expression by the examinee, and they require limited, if any, verbal instruction during
their administration (Sattler, 2008). For instance, over 80 percent of psychologists who work
with children were surveyed and reported using nonverbal tests as a primary approach for

5

evaluating English Language Learners (Rhodes et al., 2005). However, the sole use of nonverbal
or performance based tests is inherently limited in that it only surveys a limited set of cognitive
skills (Flanagan & Harrison, 2012). Additionally, reducing language based instruction and
expression does not eliminate culturally-mediated communication that occurs during the testing
process as well as the use of culturally-fused test content and stimuli (Nell, 2000). Furthermore,
the predictive validity of such tests is lower than tests that require both verbal and languagereduced performances (Lohman et al., 2008). The performance results from language-reduced
tests, or those obtained with the aid of interpretation, are ultimately interpreted based on
normative samples that are unlikely to be representative of immigrant youth. The development of
ethnic or racial specific norms is rare and when they are available are deemed controversial
because they do not address why there is a difference to begin with (Manly et al., 2005). Another
issue to consider are those immigrant youth who learned their native and English language
simultaneously and may even prefer using English. In these instances, it is likely more
appropriate to use a verbal and nonverbal combined IQ test, but it remains unclear how language
and cultural levels relate to how they perform on combined IQ tests. In summary, interpretation
of test outcomes will be misguided without a clear understanding of how acculturation variables
factor into verbal and language-reduced test performances especially among immigrants who ae
bilingual.
A promising theoretical framework has been introduced by Flanagan and Ortiz (2007). It
attempts to evaluate systematically the influence of the language requirements and cultural
loadings embedded within IQ tests and the interactive role of the individual’s acculturative
characteristics. The framework is called the Cultural-Language Interpretative Matrix (C-LIM),
and it was developed through the consensus of clinical and research experts on the assessment of

6

immigrants. The framework is structured along linguistic demand and cultural loading
dimensions such that each dimension can have low, moderate, or high saturation, thus resulting
in a three by three matrix. The C-LIM model predicts a declining pattern of scores as a function
of limited language competence and level of acculturation. It is assumed that the decline in
performance will be more pronounced on tests that are more culturally-and-linguistically-loaded
than on novel and language-reduced tests (Flanagan & Ortiz, 2007). A specific matrix was
developed for 20 of the most widely-used standardized IQ tests, including the Wechsler scales,
which are considered the gold standard of intellectual assessment and are the most frequently
used tests when evaluating students (Flanagan & Ortiz, 2007; Sattler, 2008).
The clinical utility of the C-LIM framework has been examined in mostly dissertation
studies (Dynda, 2008; Sotelo-Dynega, 2007; Verdorosa, 2007; Nieves-Brull, 2006). The results
from these studies were generally supportive of the clinical utility of the C-LIM when examining
group mean differences. However, there has been two peer-reviewed published studies, finding
mixed support for the C-LIM (Kranzler, Flores, & Coady, 2010; Styck & Watkins, 2013).
Kranzler and colleagues (2010) found no support for the C-LIM predicted patterns when
conducting within-subject instead of between group analyses. Styck and Watkins (2013) also did
not find significant clinical utility of the C-LIM approach to differentiate between ELL and nonELL populations using the Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV).
The C-LIM framework may improve test selection and performance interpretation by
understanding how acculturation factors and test factors interact to allow for better
differentiation between differences and deficits. However, studies that have examined the C-LIM
did not assess how levels of acculturation correspond to the classification patterns of the subtests.
It is unclear if a more systematic evaluation of the acculturation process will improve our

7

understanding of how acculturation relates to performances on verbal and language-reduced IQ.
Also, there have been no verification if the C-LIM classification pattern correspond to direct
measures of acculturation orientations and linguistic competence levels.
According to the American Psychological Association (APA), developing multicultural
competence is essential for practitioners and trainers in meeting general competency and ethical
imperatives (APA, 2003). The APA published two guiding documents in 2002 and 2003
(Guidelines on Multicultural Education, Training, Research, Practice, and Organizational
Change for Psychologists; Ethical Principles and Codes of Conduct) that highlight the
importance of moving away from a monoculture lens towards recognizing and embracing a
multicultural framework that will lead to more precise assessment and ultimately more effective
treatment (APA, 2002, 2003). Mindt, Burd, Saez, & Manly (2010) have published a call to action
for increasing culturally competent neuropsychological services for ethnic minority populations.
They argued that clinical neuropsychology as a field has not kept pace with the needs of ethnic
minority clients. A crucial component in developing cultural competence among clinicians is
increasing one’s knowledge base about their clients, raising awareness of their own potential
biases, and continually enhancing culturally-sensitive services (Eccles, Wong, & Peck, 2006).
With any group of recent immigrants, neuropsychologists will increase their cultural competence
by considering how acculturation levels and the association of other relevant demographic
factors (e.g., age, gender, SES) with intellectual assessment outcomes. Improved understanding
of the acculturation influence on intellectual performance will improve distinguishing between
performances that reflect cultural factors or cognitive deficits.
This study was designed to make multiple contributions to the multicultural
neuropsychological assessment literature. This was accomplished by using the C-LIM

8

predictions as a guide to understand how Arab American youth, an underrepresented population,
performed on standardized verbal and language-reduced IQ tests. Arab Americans are a
heterogeneous ethnic group that present with unique and salient cultural factors that are likely to
affect how they perform on IQ tests. No studies have examined the clinical utility of IQ tests
among Arab American youth. The contrast between Arab American’s heritage host culture
presents a unique opportunity to examine the association between cultural and language
development with IQ tests. The study will be conducted by examining the association between
proxy and systematic measures of acculturation with verbal and language-reduced IQ tests. Also,
the predictive association between IQ performance and academic achievement was examined
and whether there was a moderating effect of the acculturation levels toward US mainstream
culture. This was the first study to directly examine the influence of acculturation and predictive
utility of a commonly used IQ tests (i.e., WASI-II).
In the next chapter, a summary history and the clinical utility of IQ assessment among
ethnic and racial minorities will be outlined. Also, the implications of culture for cognition is
summarized. A review of the available literature on the association between acculturation and
cognitive tests is synthesized. The chapter also provides a detailed overview of the acculturative
experience of Arab Americans.

9

CHAPTER II: LITERATURE REVIEW
What is Intelligence?
There has been considerable debate over the exact nature of human intelligence since
ancient times. There is not a common and widely used conception of intelligence. Historically,
people across many different cultures have conceptualized intelligence using a wide range of
human cognitive, emotional, and behavioral attributes identified at the individual, group, and
social levels of human functioning. Some have argued that attempting to define intelligence is a
futile enterprise (Lynn, 2006). Despite the elusive nature of intelligence, several models of
intelligence have been proposed that have gained some acceptance.
Howard Gardner’s theory of Multiple Intelligence has had strong appeal and is regarded
as one of the most complete conceptions of intelligence (Gardner, 2011). The central tenet of the
Multiple Intelligence theory is that there are multiple and independent intellectual capabilities.
Such abilities include visual-spatial, verbal-linguistic, bodily-kinesthetic, logical-mathematical,
interpersonal, musical, intra-personal, and naturalistic abilities. Although all humans are assumed
to possess these capabilities, they can vary across individuals and across cultural groups. The
values of a particular society may drive the development of one form of intelligence over
another. For instance, in developed societies, the acquisition of literacy is valued and cultivated,
and as a result abstract verbal reasoning and acquired culturally relevant information are
considered foundational to overall intelligence (Gardner, 1985).
Among the most prominent theoretical conceptions of intelligence has been the Triarchic
Theory of Intelligence proposed by Robert Sternberg (1985). The theory defines intelligence as a
set of interactive mental abilities that are directed toward purposive adaptation, selection, and
shaping of real-world environments relevant to one’s life (Sternberg, 1985). The theory’s

10

foundation is based on three discrete yet interactive factors: analytical intelligence, which refers
to abstract problem-solving abilities; creative intelligence, which refers to the processing of new
and novel situations while using past or momentarily acquired knowledge; and practical or
contextual intelligence (i.e., street smarts), which refers to adapting to changing environments.
Despite the positive reception and intuitive face validity of these two theories, there has been
minimal empirical support and practical utility when compared to the psychometric
conceptualizations of intelligence.
The psychometric approach has attracted the most attention and research support of any
conception of intelligence (Sattler, 2008). It is the basis of the most widely used assessment tools
of intelligence (Sattler, 2008). It has been argued that measuring intelligence is one of
psychology’s most important contribution (Rushton & Jensen, 2004). Today, the construct of
intelligence has become almost synonymous with the psychometric derived Intelligence Quotient
(IQ). At the heart of this approach is that intelligence is a phenomenon that can be reliably and
objectively measured. Many psychometric tests measure cognitive abilities that are believed to
be indicative of intelligence, such abstract reasoning, fluid problem solving, level of acquired
knowledge, and processing speed (Sattler, 2008). These capabilities are assumed to be normally
distributed among all humans. The emergence and dominance of the psychometric approach
arose from over a century of research and development that began in Europe, but mostly matured
in United States, and it has become the foundation for nearly all IQ tests.
The Utility of IQ Assessment
Assessment of intellectual ability has become a pervasive practice across vocational,
educational, and clinical settings (Sattler, 2008). One of the most frequent applications of IQ is in
assessing the learning potential of students. The development of standardized intelligence tests

11

arose from the societal need to differentiate scholastic ability and identify cognitively normal and
abnormal children (Sattler, 2008). There is moderate to strong correlation between IQ and
academic achievement (Naglieri & Borenstein, 2003; Rhode & Thompson, 2007; Watkins, Lei,
& Canavaz, 2007; Sattler, 2008; Wechsler, 2011). Performance on global intelligence measures
is the strongest single predictor of academic achievement and total educational attainment
(Sattler, 2008). There is also a strong predictive relationship between IQ and academic
achievement as measured by grade point average (GPA), the Scholastic Aptitude Test (SAT), the
American College Test (ACT), and psychometrically normed academic achievement tests of
reading, writing, and mathematics (Naglieri & Borenstein, 2003; Rhode & Thompson, 2007;
Watkins, Lei, & Canavaz, 2007; Sattler, 2008; Wechsler, 2011). Additionally, specific cognitive
abilities relate differently to various academic outcomes. In general, global cognitive ability is
equally predicative of reading and math outcomes among school age children (Sattler, 2008;
Rhode & Thompson, 2007). However, IQ tests with greater emphasis on spatial ability are better
predictors of mathematical outcomes than verbally loaded ability tests among middle school and
high school students (Rhode & Thompson, 2007; Lubinski et al., 2001).The use of IQ is likely to
continue being an important clinical tool for evaluating the learning potential of students.
It is estimated that nearly two million IQ tests are administered to school-age children
each year (Watkins, Lei, & Ganivez, 2007; Budd, Felix, Ponidexter, Naik-Polan, & Sloss, 2002).
Not only is IQ a strong predictor of academic achievement, it is also viewed to have a causal
association with academic achievement (Watkins, Lei, & Gaivez, 2007). Neuropsychologists
have traditionally used IQ for selection, evaluation, or diagnostic purposes. Frequently, IQ tests
are used to diagnose brain pathology that may underlie occupational, behavioral, adaptive, or
academic difficulties. Definitions of cognitive impairment, by various organizations (e.g.,

12

International Classification of Diseases-Tenth Edition; Diagnostic and Statistical Manual-Fifth
Edition) are partially conceptualized based on IQ performance (Sattler, 2008). Although the new
definitions of learning disabilities under federal special education laws (e.g., IDEA, 2004) do not
require the administration of IQ tests to meet eligibility criteria, it remains the case that the
etiology of a learning disability is best viewed as a result of underlying neurocognitive
processing deficits which have been reliably documented by cognitive tests (MacMillan &
Forness, 1998).
Including an IQ measure as part of a comprehensive neuropsychological evaluation
provides important clinical information (Flanagan, Ortiz, & Alfonson, 2007). Children and
adolescents with cognitive processing differences can look behaviorally similar. Implemented
interventions can be misapplied, and ultimately be ineffective without understanding the nature
of the child’s cognitive profile (Newton & McGrew, 2010; Lezak et al., 2012). The information
gained from including intelligence tests enables clinicians to determine better the status of
cognitive strengths and weaknesses (Newton & McGrew, 2010; Flanagan & Harrison, 2012;
Lezak et al., 2012). Cognitive assessment also provides important clinical information for
documenting and monitoring neurological pathology.
The IQ test is an integral component of neuropsychological evaluations because it
provides an anchor for expected level of performance on other cognitive, behavioral, academic,
and adaptive outcome measures (Fuster, 2008; Lezak et al., 2012). Such evaluations are
conducted to survey brain-behavior integrity. Cognitive tests are sensitive to the effect of
developmental complications and acquired neurological insults (Sattler, 2008; Lezak et al.,
2012). Adverse events occurring in utero or postnatal can alter brain development, and they often
correspond to focal or global cognitive impairments (Breaslau, Johnson, & Lucia, 2001). For

13

instance, individuals who have experienced injuries through exposure to toxins, strokes, or blunt
trauma to the brain are likely to experience significant cognitive dysfunction when compared to
non-injured controls (Fuster, 2008). Pre-natal exposure to certain substances (both illicit and
licit) is also associated with an array of neuropsychological deficits, ranging from focal to more
global cognitive impairments to specific language or visual-spatial processing deficits (DelaneyBlack, Covington, Ondersma, Nordstromklee, et al., 2002). A variety of genetic (e.g.,
neurofibromatosis) and medical disorders (e.g., brain tumors, hydrocephalus, epilepsy) also
affect brain functioning and intellectual development (Breier, Fletcher, Wheless, Clark, Cass, &
Constantionou, 2000; Jokeit & Ebner, 2002). A neuropsychological evaluation can provide a
more nuanced and sensitive approach to assessing brain dysfunction that may not be detected by
traditional neurological studies, but interpretation of neuropsychological assessment is
jeopardized without a reliable and valid measure of global baseline intellectual capability. In
these clinical contexts, it is insufficient to only use nonverbal assessment. Surveying language
based reasoning and processing is often clinically necessary to aid differential diagnoses.
Without understanding how language development and culture influence language mediated tests
could lead to diagnostic errors. Despite the wide use of IQ tests and greater psychometric
robustness, there are significant controversies surrounding their use with individuals who were
not well represented during the normative and content development of these tests.
There have been consistent and persistent findings that show significant cross-racial and
cross-ethnic differences which have affected clinical utility and contributed to socio-political
controversies (Rushton & Jensen, 2005; Neisser et al., 1996). These findings show significantly
higher IQ performance of people who of White and East Asian background and much lower
performances of people from Black and Hispanic background (Rushton & Jensen, 2005). These

14

patterns have been documented across a wide range of age groups but mostly involving adult
samples (e.g., Brooks-Gunn, Klebanov, & Duncan, 1996; Rushton & Jensen, 2005). As a result
of these differences, the appropriate use and clinical utility of IQ tests among ethnic and racial
minorities and those from non-Western and non-English speaking cultures remains controversial
(Neisser et al., 1996). Considerable debate and research has been dedicated to investigating these
issues (Neisser et al., 1996). The focal point of the controversy stems from the explanations put
forth to explain performance differences. For instance, some have interpreted the persistent
cross-racial differences in performance as reflecting differences in inherent ability. Others have
attributed the differences to possible biasing factors associated with the tests, testing
environments, and cultural and environmental context of development (Neisser et al. 1996).
What follows is a brief survey of this controversy to highlight the complex interaction between
cultural, environmental, biological factors in shaping intellectual development and assessment.
Race, Culture, and Intellectual Assessment
Explanations of cross-ethnic and cross-racial IQ differences have essentially fallen along
two schools of thought, namely the Universalists (i.e., nature or genetics) and the
Environmentalists (i.e., nurture, cultural context).
The Universalists assume there are fundamental and universal human cognitive abilities
that are best estimated by global intelligence factor regardless of cultural or environmental
influence (Rushton & Jensen, 2005). Such cognitive abilities are inherently genetic, and cultural
factors have little or no influence on the development and capacity of cognitive functioning. The
Environmentalists emphasize the dramatic influence of the socio-cultural developmental context
on cognitive development and assessment.

15

Universalist’s Approach
Universalists assume that the basis of global cognitive capacity is determined almost
entirely by biological markers such as DNA, developmental maturation, and cortical volume.
These biological factors have been shaped by evolutionary pressures to serve an adaptive
function (Lynn, 2006). Such abilities are posited to vary across different groups of people, as
with other physical morphological variations that also have resulted from unique adaptive
pressures (Neisser et al., 1996; Rushton & Jensen, 2005). For instance, it is argued that Whites
and East Asians may have faced harsher climates that required more elaborate adaptive cognitive
skills than what Blacks endured in Africa. It is assumed that those who lived in the African
continent faced less harsh environmental demands and thus required less adaptive cognitive
pressures (Shorbis, 1996; Lynn, 2006). Thus, when group differences are identified they reflect
differences in aptitude attributed to adaptive genetic predispositions (Rushton & Jensen, 2005;
Lynn, 2006).
One study examined the association between cortical volume and IQ scores among 6,000
U.S. Army personal; it found a strong positive correlation between IQ and cranial capacity after
adjusting for body size (Rushton, 1993). In that same study, East Asians were found to have the
largest capacity, followed by Whites, and the smallest capacity was found among African
Americans. Furthermore, Rushton and Jensen (2005) analyzed data from different groups of
twins and found that the mean Black-White IQ differences were highly related to mean
differences in cortical volume. It was also argued that numerous twin studies from a multitude of
ethnic and racial background groups have suggested that as much as 80% of the variance in mean
IQ scores is explained by genetic heredity and not environmental factors (Jensen, 1980; Rushton
& Jensen, 2005; Shorbis, 1996; Lynn, 2006).

16

Further, a multitude of studies have failed to document systematic bias in IQ assessment
involving various racial and ethnic minorities (Rushton & Jensen, 2005). Among the most
controversial, yet persistent findings is the lower mean scores of African Americans and others
of African descent which fall on average one standard deviation below that of Whites and EastAsians (Lynn, 2006). Findings also show that other ethnic groups, such as Hispanics and Native
Americans, also fall below the averages of Whites but above Blacks (Herrnstein & Murray,
1994; Rushton & Jensen, 2005). Two meta-analyses have suggested that these identified group
differences are systematically linked to g loadings (te Nijenhuis & Dragt, 2010; te Nijenhuis &
Repko, 2011). That is, the more cognitively demanding a task is the more it is loaded onto the g
factor. The g factor is a psychometric property believed to reflect an overall global cognitive
capability (i.e., intelligence). Also, higher heritability associations are found among more
complex cognitive tasks (Lynn, 2006). These findings have provided compelling evidence that
differences in performance between U.S. minority ethnic and racial groups and Whites, and
between non-Western immigrants in Europe with White Europeans, are best attributed to genetic
differences and not societal, cultural, or environmental factors (Lynn, 2006). Controlling for
education and SES levels attenuate the differences, but do not eliminate them, especially on the
more complex cognitive tasks (Herrnstein & Murray, 1994).
Herrnstein and Murray (1994, p. 281-282) argued that “no one has found statistically
reliable evidence of predictive bias against blacks…the cultural content of test items is not the
cause of group differences in scores” and go on to conclude that “the gaping cultural gap
between the habits of the underclass and the habits of the rest of society, far more impassable
than a simple economic gap between poor and not poor, or the racial gap of black and white,

17

will make it increasingly difficult for children who have grown up in the inner city to function in
the larger society even when they want to”(p.524).
Such conclusions imply that the higher performance of Whites and East Asians on IQ
assessment validly reflects inherently higher ability when compared to African Americans,
Hispanics, and other minority groups. However, claims of cross-racial genetic differences in
intelligence have been the subject of serious critiques. Critics have questioned the evidence put
forth to justify such strong conclusions which suggests that differences are biologically based,
that the tests are not biased, and that culture and environment have little or no effect.
Some have argued that biological correlates of cross-racial differences have failed to
discern biological from environmental inheritance (Richards, 2006; Lewontin, 1993).
Additionally, the correlates of biological markers, such as cranial capacity to IQ performance,
are often small and show higher variability within than between cross-racial samples (Lewontin,
1993). Studies citied to support cross-racial differences often failed to provide systematic control
of potential SES factors that may have contributed to possible race or ethnic based differences in
relation to the used biological markers (Goodman, 2000; Nguyen & McDaniel, 2000; McDaniel,
2005). There is also variability in how biological markers are assessed. For example, cranial
capacity has sometimes been measured by using the circumferences of the skeletal cranium, and
only recently have in-vivo measures using MRI been available (McDaniel, 2005). Furthermore,
many scholars do not view race as a biologically discernable construct, rather it is viewed as an
ideological concept ascribed to biology (Goodman, 2000). Even if it is assumed that some
remote discernible genetic basis of race exists, it does not necessitate that certain traits, like
intelligence, are concordant with such genetic predispositions (Goodman, 2000).

18

Critics have questioned the claims that IQ tests are not biased toward certain population
groups (Nell, 2000). For instance, Nell (2000) argued that when the claim of “no bias” is put
forth by the Universalists, like Jensen, Herrnstein, and Murray, they are referring to a lack of
predictive bias that is often measured against the criteria of school or college success. It has been
argued that such culturally specific criteria are as biased as the IQ tests “the educational system
and test system are devised and implemented by a single intellectual elite that shares a wide
range of work-related and cultural values and the two systems are two side of the same coin”
(Nell, 2000, p. 65).
Moreover, a recent study examined whether differences between immigrants in Europe
and white Europeans are best explained by complex cognitive loading (i.e., g) or by cultural
loading (i.e., c). Essentially, the more complex cognitive tasks are, the more they are culturally
loaded. The authors concluded that their results presented a serious challenge to those insisting
that differences among non-Westerners and Westerners are solely attributed to heritable genetic
differences without a serious contribution of cultural or environmental factors (Helms-Lorenz,
van de Vijver, & Portinga, 2003). This finding also indicates that potential cultural bias is
imbedded within cognitively complex tasks. Nell (2000) argued that the Universalists need to
ensure equitable comparisons beyond a reasonable doubt by controlling for significant variables
across comparison groups, such as language proficiency, educational quality, testing familiarity,
cognitive style, and the establishment of cross-cultural construct equivalence of intelligence,
before differences can be primarily attributed to biologically based differences (Nell, 2000). The
limitation of explaining cross-racial and cross-cultural differences as a matter of differences in
biological predispositions gives rise to explanations emphasizing environmental and
sociocultural influences.

19

Environmentalist’s Approach
In contrast to the Universalists’ conceptions, the Environmentalists assume that
intelligence can be viewed as a culturally relative construct in how it is defined and assessed, but
more significantly, that culture can shape the cognitive development of specific cognitive
abilities. This view was strongly advocated by Lev Vygotsky (1978), who argued that the mind
is the product of the material conditions of culture, and the formation of higher level cognitive
abilities arises from a dynamic interaction between biological, developmental, and environmental
factors. Over time, fundamental changes occur in the process of thinking due to changes in the
material conditions in which such processes arise. From these principles, he believed that IQ is
immutable, that learning necessarily trails behind development, and that learning and cognitive
functions are circumscribed by developmental maturation and sociocultural context (Vygotsky,
1978). Thus, learning and cognition are dependent on attained levels of development, and
development is shaped by the context of culture and environment.
Alexander Luria, who was highly influenced by Vygotsky and who is credited with
establishing the foundations of modern cross-cultural neuropsychology, further advanced the
theoretical details of how environment and culture can directly shape cognitive abilities. In 1937,
Luria and colleagues set out to investigate environmental conditions, especially education
attainment, and the new push for acculturation to the cultural modernity after the Bolsheviks
revolution of 1917. The focus was on how sociocultural contexts have shaped cognitive abilities,
by comparing those that have been greatly influenced by the push of modernity and those that
remained traditional. He examined five groups of women: illiterate women from remote villages,
women in remote villages with an individualistic economy, women who had attended short
teaching courses but remained semiliterate, officeholders in the collective system who were

20

barely literate, and women students who were admitted to teaching schools and lived in urban
settings (Luria, 1976). He investigated two modes of cognitive generalization abilities, namely
functional and abstract generalizations or reasoning, by asking participants to choose and select
objects that go together and to explain why they go together. Luria (1976) identified that the
more sophisticated (i.e., more education and more modern) group of women were able to display
abstract categorical associations while the unsophisticated (traditional with very little or no
education) women made functional associations based on the presented items’ everyday use
(e.g., a circle is a plate; a triangle, a kettle stand). Luria observed, “The process of abstraction
and generalization are not invariant at all stages of socioeconomic and cultural development.
Rather, such processes are themselves products of the cultural environment” (Luria, 1976, p. 74).
Although Luria believed there are some basic cognitive processes that may be universal,
he argued that higher level cognitive functions are social in origin, and their development may
depend on the social and cultural milieu in which an individual is embedded. Luria later
formulated his theory of the extracortical organization of brain functions. The theory proposed
three hierarchical structures (Nell, 2000). Basic foundational process can be assumed to be
universal, such as sensation, movement, and elementary processes of attention and memory
storage capacity, but higher level brain functions, such as abstract reasoning, fluid reasoning,
specific memory encoding strategies, and imagination, arise as a function of environmental
demand and sociocultural contexts (Luria, 1976). Luria’s early studies and theoretical
conceptions have been supported by more contemporary studies.
Ogden and MacFarlane-Nathan (1997) examined the performance of indigenous Maori
young men from rural and urban backgrounds in New Zealand and found their overall
performance to be variable in predictable and unpredictable patterns. For instance, the

21

participants scored below the standardized U.S. norms on tests assessing vocabulary and logical
memory, but performed within the above average range on a task involving visual-spatial
abilities (e.g., one standard deviation difference). The authors argued that differences are best
explained by cultural differences in that lower language based measures were evident, but the
superiority in visual-spatial abilities was reflected by Maori cultural values. It was argued that
the Maori may have a particular aptitude for visual-spatial perception, construction, and visual
memory, as reflected by their high value for arts that involves intricate facial painting designs
and carved totems. This study supported the view that culture can give rise or suppress modes of
perceptual and cognitive functions. Similarly, environmental variables can also have dramatic
effects on cognitive development.
Environmental and socio-economic factors can have deleterious effect on neural
development. For instance, environmental toxins, such as lead or fetal-alcohol exposure, can lead
to significant cognitive impairment (Sattler, 2008; Flanagan & Harrison, 2012). Malnutrition and
deprivation during early childhood have been found substantially to lower performance on IQ
tests (Politt et al., 1994). A meta-analysis by Devlin and colleagues (1997) showed a positive
association between breastfeeding duration and IQ performance. Similarly, levels of SES are
associated with IQ performance and heritability. For instance, low SES and low education
attainment were strongly associated with lower IQ performance (Flannigan & Harrison, 2012;
Heaton, Taylor, & Manly, 2003; Pollitt, Gorman, Engle, Martorell, & Riveria, 1993). Both
verbal and language-reduced performances correlated positively with education levels (Heaton,
Taylor, & Manly, 2003; Harris, Tulsky, & Schulthei, 2003). Higher parental education and types
of parental practices positively correlated with higher IQ performances (Molfese, Modglin, &
Molfese, 2004). Among adults, education level was the most important factor for predicting IQ

22

levels, especially verbal IQ (Sattler, 2008). Not only was education level an important predictor
of IQ, the quality of education mattered.
Shuttleworth-Edwards and colleagues (2004) examined the effect of the quality of
schooling on IQ performance; despite the same level of education obtained, those from superior
quality education backgrounds consistently outperformed those with lower quality education
across verbal and language-reduced tasks after age, gender, and ethnicity were controlled. In that
same study, South Africans with higher quality education had commensurate scores with U.S.
standardization norms, but those with lower quality education performed significantly worse,
irrespective of whether the participants were White or Black South Africans. The ShuttlesworthEdwards and colleagues (2004) study suggested that education quality is a more powerful
moderator of IQ performance than ethnicity. Green, Hoffman, Morse, Hayes, & Morgan, (1964)
conducted a study in a Virginia county where school was closed for several years in the 1960s to
avoid integration, leaving most Black children with no formal education at all. When compared
to controls, the IQ scores of these children dropped by a 0.4 standard deviation per missed year.
Thus, environmental conditions, exposure to educational experiences, and cultural values have
important implications for cognitive development and assessment.
Acculturation and Cognitive Development
The Acculturation Process
The process of acculturation can have significant implications for the cognitive, cultural,
and linguistic development of young immigrants. In turn, it has significant implications for the
intellectual assessment of immigrants. The term “acculturation” was first used in 1880 as a way
to describe group-level acculturation phenomenon (Sam & Berry, 2006). A classic definition of
acculturation was provided by anthropologists Redfield, Linton, & Herskovits (1936), who stated

23

that “acculturation comprehends those phenomena which result when groups of individuals
having different cultures come into continuous first hand contact with subsequent changes in the
original culture patterns” (p. 149). Psychological acculturation, which occurs at the individual
level, was first postulated by Graves (1967), who stated that acculturation is a process that
“refers to changes in an individual who is a participant in a culture-contact situation—a person
who is being influenced directly by the external culture and by the changing culture of which the
individual is a member” (Berry, 2003, p. 19). Important changes to the individual’s personal
identity, beliefs, values, attitudes, behavior, cognition, and language competence occur during
acculturative experiences (Berry, 1997).
Acquisition of a new language is arguably one of the most influential changes an
individual experiences. Language competence can itself be a determining factor of how well an
individual acculturates to their host society (Berry, 2003). This in turn has substantial
implications when assessing the cognitive functioning of immigrant youth who may have
variable language skills. The process of second language acquisition is complex and is likely to
vary across individuals, depending on their language of origin, pace of acquisition and
developmental stage of language acquisition.
Cummins (1984) hypothesized that second language acquisition follows two general
stages of development among immigrants. The first stage describes the initial acquisition skills
that involve surface level fluency and capacity referred to as the Basic Interpersonal
Communication Skills (BICS). This level of competence often develops after two to five years of
immersion. Although students who have reached the BICS level of acquisition may appear fluent
in the acquired language as they rely heavily on contextual cues, they may lack significant skills,
including level of vocabulary knowledge, which can indirectly affect more complex

24

comprehension and reasoning. Subsequently, another set of more sophisticated competence skills
are achieved which are referred to as the Cognitive Academic Language Proficiency (CALP),
after five to seven years of immersion (Cummins, 1984). The CALP level of acquisition is
viewed to be essential for success on academic tasks, and potentially for a non-biased
performance on a verbally mediated IQ measure.
It is important to note that the projected pattern of language development is also affected
by the age of exposure. Acquisition of a second language at a younger age is associated with
more efficient processing of both languages, compared to acquisition at an older age (Miyaki et
al., 2000; Carhill, Suarez-Orozco, & Paez, 2013).Additionally, immigrant children’s language
development varies depending on their age of immigration and level of exposure to their native
language. Children born in the U.S. to immigrant parents are likely to develop more competent
language skills than children who learn English as a second language (Carhill, Suarez-Orozco, &
Paez, 2013). Those learning English as a second language are likely to do so by making use of
their existing native language skills (Bialystok, 2011). It remains unclear how patterns of
acculturation and level of language competence are related to performance on IQ and academic
outcomes. There has been important development of theoretical attempts to conceptualize the
dynamic changes that occur during the acculturation process. Such theoretical models can be
used to evaluate how the process of acculturation affects intellectual assessment and academic
achievement.
Models of Acculturation
The acculturation patterns have largely been viewed through unidimensional or
bidimensional models (Rudmin, 2009). The unidimensional approach explains the acculturation
process as a movement between two cultural influences (i.e., host vs. heritage). In general,

25

immigrants undergo a process of assimilation to the host society, and they move away from their
heritage background. It is a zero-sum linear acculturation process that ultimately results in the
individual’s complete assimilation to the host culture and complete shedding of heritage
identification (Gordon, 1964; Berry, 1997; Cabassa, 2004). It has been documented, across many
different ethnic immigrant groups in America, that the cultural orientations of third or even
second generation individuals are far more likely to embrace cultural views, and develop better
cultural and language competencies associated with the host culture, than with their heritage
background (Sam & Berry, 2010; Cabassa, 2004). However, the acculturation experiences of
first generation immigrants, especially those from a more incongruent heritage society, may have
different patterns of acculturation changes (Sam & Berry, 2006).
The need for more nuanced conceptualizations of potentially diverse multiple
acculturation orientations has resulted in the development of the bidimensional approach and its
subsequent modifications (Berry, 1980; Berry et al., 2006; Rudmin, 2009). The bidimensional
approach recognizes the possibility of various acculturation orientations as a result of the
interactions between adaptation and maintenance processes (see Table 1). Central to this model
is the concept that there are two independent (i.e., orthogonal) dimensions underlying the process
of acculturation (i.e., participation in the host culture and maintenance of the heritage culture).
This model is more sensitive to the dynamics of psychological change that occur across the
domains related to the heritage background and the host context, especially for first generation
immigrants.
Berry’s bidiminsional model of acculturation has been the most researched and most
supported model. The assumptions of the model allow for a total of four possible theoretically

26

distinct acculturation orientations that may develop: integration, assimilation, separation, and
marginalization (Berry, 1997).
Table 1
Berry’s Bidimensional Model of Acculturation
Heritage:
Maintenance (YES) Maintenance (NO)
Host: Participation (YES)
Participation (NO)

Integration

Assimilation

Separation

Marginalization

The assimilation orientation describes the momentum toward the adoption of the host
culture’s customs, beliefs, and language while expending lower efforts (less time) on maintaining
one’s heritage background. The forces that shape a person’s motivation may not always be
controlled by the individual however. For instance, children of immigrants or children who
immigrate at a very young age, for developmental and cultural exposure reasons, may find it
easier to assimilate to the host society without having continual and direct contact with the
heritage society (Sam & Berry, 2010; Brown & Zagefka, 2011). The separation orientation is the
continual embrace and maintenance of the heritage culture while not identifying with the host
society’s structure, values, and customs. Individuals who integrate both cultural backgrounds by
combining different aspects of each society to form a unified identity are described as having an
integrated orientation. Lastly, the marginalization orientation describes individuals who have
moved away from both their host and heritage cultures. Overall, the bidimensional framework
has found empirical support for defining acculturation orientations, and it has enabled crosscultural acculturation research among a multitude of ethnic minority immigrants (Ryder et al.,

27

2001; Berry et al., 2006). Berry’s framework has been expanded by applying it to multiple
domains of functioning. That is, an individual can have an integrated acculturation orientation
with language acquisition, but may have a separated or assimilated cultural identity development.
Similarly, an individual can have an integrated cultural identity, but may only speak one
language. Researchers have highlighted the importance of assessing acculturation orientation in a
bidimensional framework along multiple domains (Zea, Asner-Self, Birman, & Buki, 2003;
Rudmin, 2009; Berry et al., 2006).
Assessing Acculturation
The method of conceptualizing the acculturation process has evolved over the years. A
recent 22-year literature review of acculturation research indicated that there has been an
increase in acculturation research using the bidimensional approach to multiple dimensions of
functioning, but the vast majority of studies have been based on unidimensional models looking
at an overall dimension (Yoon, Langrehr, & Ong, 2011; Zea, Asner-Self, Birman, & Buki, 2007).
Both models have been used to develop over 50 acculturation scales which are freely available.
However there are only a few available scales that survey multiple dimensions of acculturation
based on a bidimensional approach and have adequate psychometric properties among younger
and older immigrants (see Zea, et al., 2007 for a review). According to Yoon, Langrehr, & Ong
(2011), adequate acculturation scales should meet four basic criteria: 1) they should assume a
bidimensional framework; 2) they should consider domain specificity adaptation; 3) they should
have sufficient items per domain; and 4) they should have good internal consistency. These
guidelines have been used to develop, adapt, or select relatively comprehensive acculturation
scales. However, available robust scales have mostly been developed for use among Hispanic
and Asian immigrants, and to a lesser extent with other ethnic groups in and out of America

28

(e.g., Berry, et al., 2006; Chung et al., 2004; Costigan & Dokis, 2006). The bidimensional
framework has generalized approach among a wide range of cross-cultural acculturation studies
which have explored the process of acculturation and associated educational, psychological, and
cognitive outcomes. The development of useful models and measurements of acculturation will
contribute to enhancing multicultural competence that can be applied when providing
psychological services to immigrant populations. Even the most reliable and valid acculturation
scales will not capture the vast complexity that is unique to each individual immigrant group and
individual immigrant. Immigrant groups like Arab Americans possess unique cultural, linguistic,
religious, and socio-political histories that shape their acculturative experience in the US.
Acculturation of Arab Americans
The term Arab refers to an ethnic designation of people who reside in 22 countries that
span North Africa and the Middle East. The total Arab population worldwide is estimated to be
367 million (Gregg, 2005). Arabic is the fourth most widely spoken language in the world
(Gregg, 2005), and it is the language in which the Koran is written. The Koran is the Muslims’
holy book. This is a significant cultural fact because it is foundational source of common identity
for Arabs and Muslims living in and outside of the Middle East. Notably, most Arab Americans
are Christians, and they may identify with being Arab more through language than through
religion (Arab American Institute; AAI, 2003). Many Christian Arab Americans came from
predominantly Muslim countries, so they may have been exposed to cultural norms similar to
many Muslim Arab Americans. Being Arabic does not connote a racial description; rather it is an
ethnic classification based on shared cultural, linguistic, and religious values (Gregg, 2005).
Although the Arab identity is imbued with broad characteristics such as shared historical

29

identity, hierarchical interdependence, family and social structures, the Arabic language, and
Abrahamic faiths, there are very important differences.
There are important variations among Arabs in their religious faiths, language dialects,
colonial influences, and physical features. Approximately 92% of Arabs are Muslims, and
approximately 5% are Christians (AAI, nd). A popular misconception is that all Muslims are
Arabs; Arabs in fact make up approximately 17% of the total Muslim population worldwide
(AAI, nd). Countries such as Indonesia, Pakistan, Afghanistan, Turkey, and Iran are
predominantly Muslim, but are not Arab. In countries like Lebanon, approximately half of the
population is Christian (Abudabbeh, 1997). Skin color is another variation among Arabs. People
in countries like Lebanon, Syria, and Palestine have lighter skin than those from the Persian
Gulf, Arabian Peninsula, or North African countries. Also, the socio-historical context based on
colonial influence culturally differentiates between some Arab countries. For instance, Yemen
and Oman were mildly influenced by British colonialism. As a result, their language dialect and
cultural customs were affected to a lesser extent than the North African Arab countries. Morocco
and Algeria were heavily influenced by French colonialism. Consequently, people in these North
African Arab countries began to speak an Arabic dialect that is a hybrid of Arabic and French
(Gregg, 2005). Furthermore, some Arabic countries are much wealthier than others (e.g., Saudi
Arabia vs. Yemen), and as a result their citizens may have acquired greater educational access.
Immigrants from higher SES countries may not face as many cultural hurdles as those with lower
SES during the immigration process (Bradley & Corwyn, 2002). Where an immigrant comes
from, the reason for their immigration, and when they arrive in America, matters to how well
they acculturate to life in America.

30

There have been three major Arab immigration waves that have led to the primary
concentration of settlements in America. The first wave occurred in the 1880s through the 1920s,
during which most of the immigrants were primarily farmers with very little education. Most of
these immigrants were Christians, and they arrived from Syria, Lebanon, and Palestine (AAI,
nd). This wave of immigrants typically settled along the Eastern Coast of America (e.g., New
York and New Jersey). Not very much is known about the adjustment of first wave immigrants
besides that they shared common physical features and religious background with European
Americans. The second wave occurred between 1935 and 1958, during which there was an equal
number of Muslims and Christians, both with higher levels of education (Gregg, 2005). The
second wave was triggered by instability before, during, and after World War II. Most of the
immigrants from the second wave settled on the East Coast and in the Midwest (Gregg, 2005).
The second wave of immigration came to an end with the introduction of new legislation that
required the use of quotas for incoming immigrants (Kayyali, 2006). Many Arab Americans
were not able to immigrate to the U.S. until 1965, which is considered the beginning of the third
wave (Gregg, 2005). The third wave immigrants came from many countries of the Arab world
and were mostly Muslim. Many of the third wave immigrants came to America to escape war
and other social unrest (Gregg, 2005). Most of them settled on the West Coast and in the
Midwest (primarily in the Detroit Metropolitan Area). Arab Americans have immigrated to the
U.S. for well over a century for many different reasons and under many different circumstances.
The impetus for immigration among Arab Americans has not been identical. Although
most came to America on a voluntary basis, a significant number of immigrants left to escape
conflict in their homeland. This is especially true for the second and third wave immigrants, who
fled their countries to seek refuge or asylum (Abudabbeh, 1997). The reason for immigration has

31

been conceptualized under two broad categories, namely voluntary or involuntary (Ogbu &
Simons, 2008). Many Arab American immigrants came to America as asylum seekers and
refugees (Wrobel, Farrag, Hymes, 2009). Many of the refugees and asylum seekers from Iraq
and Lebanon mostly resided in Detroit, New York, and Los Angeles (Gregg, 2005). These
separate waves of immigration have contributed to the diverse assortment that constitutes the
modern Arab American identity. As a result, the umbrella of the Arab American identity
encompasses a highly heterogeneous group that is similar to other broadly described groups,
such as Hispanic or Asian Americans. It is imperative that greater attention is paid to the unique
acculturative patterns and how they directly relate to cognitive outcomes in order to gain a
greater understanding of how the process of acculturation is associated with cognitive
performance.
Key Demographics of Arab Americans
The current number of Arab Americans living in the U.S. is unknown, but is estimated to
be between 1.2 million and 3.5 million (Brittingham & de la Cruz, 2005; AAI, 2009). The first
estimate is based on the 2000 U.S. Census, and the second estimate is based on an
epidemiological study conducted by the Arab American Institute (AAI, 2009) to address some of
the underlying limitations of the Census data. Part of the difficulty in assessing the Arab
American population is that they are not officially recognized as an ethnic minority group by the
U.S. government. People whose origin is in the Middle East or North Africa are considered
White (as identified on the census demographic form); however, Arab Americans are a multicultural, multi-ethnic, and a multi-racial group (Abudabbeh & Nydell, 1993). One Arab
American can have white skin, blue eyes, and red hair while another can have dark skin, brown
eyes, and black hair. This creates a unique set of challenges for Arab Americans; they are not

32

typically recognized as an ethnic minority group, but appear to have experiences of prejudice,
profiling, and discrimination that parallel those of Hispanic, Asian, and African Americans. Arab
American’s identity with White ethnicity is relatively high among the older generation of Arab
American immigrants who are often light skinned and Christian (Ajrouch, 2000). However,
among the more recent Arab American immigrants, only a small minority might identify with
being White (Ajorouch & Jamaal, 2007). These challenges make it very difficult to obtain an
accurate depiction of Arab Americans and their patterns of adjustment in America. It is likely
that such differences in socio-demographic and phenotypic differences affect the pattern of
acculturation in the United States.
According to a nationally representative Zogby poll (2002), approximately 51% of ArabAmericans speak a language other than English at home (predominantly Arabic). It is also
estimated that 40% of Arab Americans are school-age children. Approximately 80% of Arab
Americans are American citizens; 89% have at least a high school diploma; 46% have a
bachelor’s degree, and 19% have a post-graduate degree. Also, about two-thirds of Arab
Americans were born in America (Erickson & Al-Taimimi, 2001). However, there is no
specificity in these demographic rates based on country of origin. It is possible that such
stratification would show that recent immigrants, such as those coming from Iraq or Yemen, and
who are more likely to be Muslim, would have considerably lower SES than Arab Americans
who can trace their roots back to the first and even second immigration waves.
Arab Americans live in every state across America according to a poll conducted by the
Arab American Institute (AAI, nd); although most are concentrated in urban constellations
across the East Coast, the Midwest, and the West Cost. Approximately 94% live in metropolitan
areas of major U.S. cities, such as Detroit, Los Angeles, New York, Chicago, Washington D.C.,

33

and Northeastern New Jersey (AAI, nd). The cultural background of Arabs as well as where the
settle in the US is important for understanding the acculturative experience of Arab Americans.
Settlement Dynamics
Not only are macro-level socio-political factors important for the developments of a
child, but so are the characteristics of the neighborhood. It is repeatedly documented that people
who live in safer, cleaner, and friendly neighborhoods experience higher levels of health and
well-being as they grow older (White, 2008; Bradley & Corwyn, 2002). Those who live in
neighborhoods with more vandalism, graffiti, and crime are associated with poorer health and
overall wellbeing outcomes (e.g., higher depression and anxiety). Other studies have shown that
children raised in lower SES households are less likely to graduate from high school and attend
college (Bradley & Corwyn, 2002). The assessment of SES in the reviewed studies was mostly
based on quantifying family income and parental education. Immigrants have historically sought
settlement in areas with higher low-skill labor markets such as manufacturing (Gregg, 2005).
The settlement patterns of Arab Americans are also highly variable, and as a result could
influence how well they adjust to life in America. For instance, the Metro-Detroit area is home to
the largest Arab concentration outside the Arab world and Paris. Nearly 500,000 Arab
Americans were estimated to reside in the Metro-Detroit area over a decade ago (Zogby, 2002).
Although the vast majority of Arab Americans are Christian, the majority in the Detroit-Metro
area are Muslims who arrived during the third major wave of immigration (Wrobel, Farrag, &
Hymes, 2009). Many immigrants from the second and third immigration called the Detroit area
home because of the historically great manufacturing job opportunities. During the past few
decades Detroit has become symbolic with deterioration, poverty, violence, low economic
opportunities, and dysfunctional school systems (Mrozowski, 2008). Such changes in settlement

34

contexts change affected Arab American youth and their families because they add additional
stressors that will affect their developmental context (Coll, Lamberty, Jenkins, McAdoo, Crnic,
Wasi, & Garcia, 1996).
Arab American Family Dynamics
There are unique and salient factors about Arab American families that are essential for
researchers, educational and mental health practitioners to be knowledgeable of. The family
system for Arab Americans is an integral component of their self-identity in dynamic ways. Arab
Americans come from ethnic backgrounds that are largely traditional, tribal, and collectivistic, in
which the family is the core of their ethnic and self-identity (Abudabbeh, 1997). Individuals who
come from collectivistic societies experience greater adaptation challenges. In general, a
collectivistic society values interpersonal connectivity, strong family connectivity, and an
identity that is based on such interdependence (Triandis, 1995). There is less focus on
individuality and more value on communal attitudes and behaviors (Triandis, 1995). In addition,
collectivism can be divided into horizontal collectivism and vertical collectivism. Horizontal
collectivism is relatively independent of hierarchical relationships, and there is relative equality
among individuals. Decentralized collective communist societies are considered to have
horizontal collectivism (Triandis, 1995). Vertical collectivism refers to a more hierarchical
structure of power and moral and centralized cultural conformity (e.g., Arab countries). In
general, individualistic societies are characterized by their values of personal freedom, selfreliance, and individuality. Psychological and cognitive style changes are implicated as a result
of exposure to two such relatively diverse cultural frameworks. Also, the individual assessment
context could prove to be an unfamiliar alienating process, especially when there is no gender
concordance between the examinee and the examiner.

35

Gender is an important dimension of Arab American family life. In general, gender roles
are largely fixed by tradition and religious values. For instance, marriage is often arranged, and
women are allowed to divorce only in very rare circumstances. This cultural fact has important
implications for gender interaction during mental health service delivery. Intercourse before
marriage is taboo, especially for females. Engaging in sexual behavior prior to marriage runs the
risk of ruining family honor and family reputation (Arabic terms sharaf al-aila; somaat al-aila);
this is more so for females than for males (Amer, 2005). In more traditional Arab families, a
female of child bearing age cannot be alone with another male without the accompaniment of an
adult male family member. Thus, dating before marriage is controversial and could elicit strong
feelings from the family that may negatively impact the individual’s wellbeing. The traditional
gender role expectation is likely to affect how Arab American youth acculturate to life in
America.
Arab American youth raised in America may strive to participate in mainstream
American norms at the expense of strong family rebuke. Traditional roles contribute to some
challenges children may face growing up in America. Many youth may not resolve such
differences with their family without experiencing substantial mental health difficulties or
separatist acculturation orientation (e.g., Berry et al., 2006). Children are expected to be obedient
to their parents and fully to adopt family values. Children are expected to take care of their
parents as they get older; the concept of a retirement home is not culturally practiced by Arab
Americans. Corporal punishment may be considered proper parenting by many Arab Americans,
as has been documented with other immigrant and ethnic groups (e.g., Mexican, African
American; Berlin, Ispa, Fine et al., 2009); overall, parenting style is less structured than
European Americans and is gender specific (Abudabbeh & Hays, 2005). Parents are likely to be

36

more permissive with males and more restrictive with females (Abudabbeh & Hays, 2005). The
father is often involved with discipline and needs outside the home, whereas the mother attends
to the child’s home and sometimes to school needs. Not only is gender relevant to possible
strategies of acculturation for the youth it also plays a role in how their parents are able to
interact in and with American mainstream society.
Many immigrant parents from developing countries do not view themselves as having a
large role in their child’s education, and so they may simply defer to the school staff and other
educational authority figures (Lopez et al., 2001). This may be true for Arab Americans as well,
since they may face the same challenges. Such challenges may include cultural and language
barriers to participation. It is plausible that school expectations for behavior may not be
consistent with home expectations, and as result behavioral challenges could manifest in school
(Berlin et al., 2009). As a result, Arab American youth must adjust to two different disciplinary
environments at home and at school. Children who come from households with a more relaxed
structure may struggle with understanding the more subtle, less corporal methods of behavior
modification often employed within school environments (Downey & Pribesh, 2004). Cultural
differences between students and teachers are associated with greater risk of behavioral
disruption and lower academic achievement (Downey & Pribesh, 2004).
In summary, Arab American adolescents live under unique cultural circumstances. They
may grow up in America witnessing the perceived cultural clash between America and the
Middle East; they may even be discriminated against as a result of socio-political tensions. The
push and pull between two cultural identities is likely to affect how Arab American youth
develop and acculturate to American mainstream society as has been identified among
adolescent boys of Hispanic background (Romero & Roberts, 2003; Gil, Vega, Dimas, 1994).

37

Arab Americans have called America home for over 100 years, and they have become a
substantial immigrant group, although they are still not recognized as an official ethnic group.
Very little is known about the way Arab Americans adjust to life in America. What is clear is
that the acculturation process has substantial influence in how immigrant youth and their families
function across multiple domains. What is not clear is the acculturative contexts influences
cognitive development and assessment.
The Cultural Milieu of Cognitive Assessment
Psychological testing arose from a Western cultural tradition, and its foundational
development was driven by elite White men from Europe and the United States (Boake, 2002).
Also, contemporary normative standards that guide interpretation continue to be based on
predominately monolingual White individuals of middle class status living in the United States
(Sattler, 2008; Flannigan & Harrison, 2012). As a result, the testing process is essentially a
culturally relative practice that was created to measure culturally specific functions.
The value-laden testing situation and its effect on performance is not a matter of
theoretical suppositions, but has been empirically substantiated (Shuttleworth-Edwards et al.,
2004). The familiarity with testing, referred to as “test-wiseness,” has been found to moderate
test performance significantly (Anastasi & Urbina, 1997; Nell, 2000). Individuals who have been
brought up in a test-taking culture are likely to have more experience with similar test-taking
situations, and to have more advanced strategies that will improve their performance. In one
study, elementary school children from diverse cultural backgrounds who were not familiar with
individual testing performed significantly lower than native children who were familiar with
individual testing (Tzuriel & Kaufman, 1999). However, after the migrant children were
provided with some initial training of the testing scenario, their performance significantly

38

improved much more than the performance among the native children, who were also provided
with the same level of training (Tzuriel & Kaufman, 1999). In another study, van de Vijver
(2008) examined the cross-cultural differences of basic reaction-time tasks among secondary
school students in Zimbabwe and the Netherlands, using a total of four trials. The results
indicated that the Dutch students performed faster on initial trials, but the Zimbabwe students
made the greatest performance gain in subsequent trials, and overall there were no subsequent
between-group differences. The author concluded that extra caution is warranted when
interpreting initial cross-cultural differences because they may reflect non-target variance
attributed to understanding and exposure of the task, and not to inherent ability (van de Vijver,
2008). Similarly, non-target factors such as differences in cultural experiences and lower
language competence levels may depress or bias cognitive performances that are not directly
related to overall cognitive ability.
Deconstructing Bias in Assessment
Arguably, among the biggest challenge practitioners face in their assessment of
immigrant youth is ensuring that performance and result interpretation are not biased by
acculturation factors. In general, bias occurs when the intended target of measurement is not
reflected in the obtained performance. Bias in the context of intellectual assessment among nonmainstream samples has been deconstructed into three forms: construct bias, method bias, and
item bias (van de Vijver & Tanzer, 2004). Construct bias occurs when the intended construct is
not identically measured across cultural groups. For example, individuals from cultures that
value social aspects of intelligence and have less emphasis on speeded performances may
underperform on intelligence measures that are heavily determined by demonstrating processing
efficiency and individualized reasoning (Nell, 2000).

39

Method bias refers to problems that occur because of the test components and testing
process (van de Vijver & Tanzer, 2004). Individuals from cultural backgrounds who are not
familiar with the process of standardized testing are likely to have their performance reflect their
testing experience and not their targeted measured ability. The use of stimuli such as geometric
blocks, paper, and pencil problem-solving tasks, or analyzing culturally specific pictures, may be
novel to some immigrant children, but are relatively over learned skills for many American born
children (Nell, 2000). Method bias can also occur if there are communication problems during
the test’s administration. This is likely to occur with individuals who are not fluent in English, or
if there are salient characteristics difference between examinee and examiner.
Method bias may also refer to sample bias. Namely, the standardized scores are based on
sampling a population that may not reflect the characteristics of the individual being assessed.
For instance, assumption of comparability across salient factors such as educational attainment,
parental education backgrounds, level of cultural exposure, and other salient socio-demographic
factors (e.g., income, access to health care) may not be met, thus making the comparison
potentially invalid. No available IQ tests have stratified norms for individuals who are English
language learners and have variable levels of acculturation to U.S. society (Sattler, 2008;
Kranzler et al., 2010). Thus, method bias becomes a strong possibility when assessing immigrant
children who have not had adequate cultural and linguistic exposure.
Lastly, item bias, which refers to distortions at the item level, may occur in a number of
ways. Biased items may have different meanings and importance across cultures, and this is
likely to be encountered on tests that emphasize acquired knowledge. Many IQ tests require the
demonstration of acquired experiential knowledge and the ability to use such knowledge to solve
problems. For instance, the Verbal Comprehension Index from the Wechsler scales, or other

40

measures of crystallized intelligence, are essentially cognitive acculturation measures because
they require the demonstration and use of age-expected word, informational, and experiential
knowledge that is culturally specific. For example, an immigrant student may struggle with
naming relatively familiar items to very young American children, such as a “fire hydrant.” They
may also struggle to reason about common cultural experiences, such as providing a rationale
about the importance of exercising and eating a healthy diet. Thus, the process of acculturation
has important implications not only for cognitive development but also for administering and
interpreting performances on measures of cognitive ability.
The Influence of Acculturation on Cognitive Performance
There has been evidence of performance differences on cognitive tests between
immigrants and natives since the early 20th century. Results from a study conducted by Yerkes
(1921) using the Stanford-Binet intelligence test showed that performance increased as a
function of years of residency. Individuals who resided for five years or less in the U.S. averaged
11.9 mental age score points, where the expected average mental age was 13.08. Those who
resided for 10 years or more met or exceeded the expected mental age (Yerkes, 1921). The
results showed similar and more systematic trends using the Wechsler scales with Hispanic
groups. Vukovich and Figueroa (1972, as citied in Flanagan & Harrison, 2012) assessed a group
of Hispanic adult participants and showed significantly below average performance on language
and culturally saturated tasks (e.g., Vocabulary, Information, Comprehension subtests)
contrasted by within the average range performance on language-reduced or performance tasks
(e.g., Block Design, Picture Concepts). The same sample was tested a decade later and the
participants performed within the average range across verbal and language-reduced tasks
(Vukovich & Figueroa, 1982). The sample improved on verbal tasks as well as on language-

41

reduced tasks, suggesting some acculturative effect on language-reduced tasks. Another study
also showed better performance on language-reduced than verbal tasks, but the overall
performance was below average the across verbal and language-reduced tasks (Cummins, 1982).
Only recently has there been an effort to identify the systematic influence of acculturation on
measures of cognitive functioning.
Razani and colleagues (2007) investigated the relation between some acculturation
variables and a variety of cognitive outcomes among Hispanic, Asian, Middle Eastern, and
European American non-clinical samples. Acculturation was conceptualized by a unidimensional
measure of language acquisition, level of education, and duration of stay in the host culture.
Results indicated that European Americans performed significantly better than ethnic immigrant
minorities on the verbal but not on the language-reduced, subtests from the Wechsler
Abbreviated Scale of Intelligence (WASI; Wechsler, 1999). Acculturation variables indicative of
greater assimilation are generally associated with higher performance on the Vocabulary (i.e.,
word knowledge) and Similarities (i.e., verbal categorical and abstract reasoning task) subtests
than individuals with high heritage identity. The differences between more assimilated ethnic
individuals and European Americans were not statistically significant. Language loaded tests are
highly likely to be affected by acculturation levels, and by some possible influence with
language-reduced tests as well.
The relation between acculturation and language-reduced intelligence was assessed
among pluralistic group of immigrant and native children in Europe (van de Vijver, HelmsLorenz, & Feltzer, 1999). The study included children who were Moroccan, Turkish, Bosnian,
Serbian, Somali, and Pakistani who resided in the Netherlands. Positive correlations among
acculturation orientation and verbal and language-reduced cognitive outcomes were higher

42

among first generation compared to second generation immigrants (van de Vijver, HelmsLorenz, & Feltzer, 1999). Those with higher assimilation performed better than those with lower
assimilation. Also, those with higher heritage were likely to be less assimilated, and thus
generally performed lower than those with lower heritage acculturation. However, the
independent effect of heritage acculturation, or enculturation was not separately analyzed.
Survey of the literature also failed to identify any study that has examined the possible
moderating effect of heritage acculturation even though acculturation is best viewed as a
bidimensional construct. The results from the reviewed study were somewhat variable among the
different immigrant groups, but the study’s sample was not adequate to detect any statistically
meaningful differences. Thus, it was not clear if there was an interactive association between
acculturation and cognitive outcomes between different ethnic groups.
Some language-reduced neuropsychological measures (e.g., Wisconsin Card Sorting
Test) are assumed to be relatively culture-free, but normative differences were documented in a
sample of Mexican American adults (Coffey, Marmol, Schock, & Adams, 2005). Variations
within a Mexican American sample were significantly associated across different levels of
cultural assimilation, but it was not clear if level of ethnic association was related with the
obtained results. Another study, among Mexican and African American children, found that a
greater number of “Anglo” sociocultural characteristics were associated with higher IQ
performance (Mercer, 1973). However, both of these studies did not systematically analyze the
experience of acculturation by differentiating between proximal measures and levels of
assimilation across different domains of functioning (e.g., language competence or cultural
knowledge). These results are consistent with previous studies in documenting that better

43

cognitive performance is associated with greater assimilation on verbal and language-reduced
cognitive tasks (Dana, 2005).
The available literature on the influence of acculturation on cognitive performance is
limited. Most of the available studies have been conducted with adult samples and have not
systematically and concordantly controlled for acculturative variables such as duration of
residency, language competence, acculturation orientation, cultural identification, and generation
status. The few available studies with children have similar limitations. Studies have also
neglected to assess the clinical utility of IQ tests among immigrant youth, despite possible bias.
Although it is clear that acculturation factors influence cognitive performance it remains
unknown what predictive value these tests can serve among such populations. It remains unclear
what acculturation variables are prominent in affecting intellectual assessment. Clinicians
assessing children from culturally and linguistically diverse backgrounds continue to face great
hurdles with their attempts to provide reliable, valid, and non-discriminatory assessments.
Addressing Challenges in the Intellectual Assessment of Immigrant Youth
There are a limited number of strategies available to help clinicians reduce bias and nontarget confounds during the assessment of immigrant youth. The vast majority of practitioners
lack a clear understanding about how acculturation variables affect cognitive assessment
(Rhodes et al., 2005). Approximately one of five young people under the age of 20 speaks a
language other than English at home and by 2040 it is projected to be one of three (Hernandez,
Denton, & Macartney, 2007). A national survey, completed by over 400 school psychologists
across the U.S., indicated that over 80 percent of psychologists did not feel that they were
properly trained by their graduate programs to conduct assessments of bilingual children (Ochoa,
Rivera, & Ford, 2004). Over 75 percent responded that they did not feel competent assessing

44

English Language Learners (Ochoa, Rivera, & Ford, 2004). Rhodes and colleagues (2005)
highlighted serious shortcomings in the assessment of culturally and linguistically diverse
students through the use of untrained interpreters, and through conducting cognitive assessment
without assessing acculturation levels or even language competence. These deficient practices
were partly attributed to the lack of empirically based guidelines of what the specific effects of
language and acculturation may be during cognitive evaluations. There are relatively more
resources available to practitioners when assessing larger minority groups such as Hispanics and
African Americans than there are for smaller minority groups. Although development of
translated and adapted IQ tests and sometimes ethnic specific norms, have become available (see
Heaton et al., 2002) the use of translated tests and population-specific norms remains
controversial (Manly et al., 2004).
Ethnic Specific Norms
Developing racial or ethnic based norms has been advocated by some as a way to
enhance the sensitivity and accuracy of cognitive measures (Heaton, Taylor, & Manly, 2003).
However, significant drawbacks are associated with such an approach. First, specific norms are
minimally available to large ethnic groups such as African and Hispanic Americans. Also, the
generalizability of such norms may not be completely appropriate, given the significant
heterogeneity among ethnic groups like Hispanics, Asians, or African Americans. Assuming that
such norms can be readily available for a wide range of ethnic groups, there are still foundational
problems that arise from this approach. For instance, Manly (2008) noted the following
drawbacks: 1) race or ethnic specific forms do not explain how or why test scores vary as a
function of group membership; 2) issues of test bias or cultural equivalence are not sufficiently
addressed; 3) such an approach will lead to irresponsible social, biological, and genetic

45

conclusions regarding differences; and4) development of such norms is logistically difficult to
develop for existing tests. Practitioners have attempted to attenuate the effect of ethnicity by
using standardized language-reduced tests.
Language-Reduced and Nonverbal Assessment
Other methods of reducing cultural and linguistic bias have been through using languagereduced tests. Language-reduced tests are often referred to as “culturally free” or “culturally
fairer” because they involve novel stimuli, are performance based, and do not require expressive
language from the examinee, and some tasks are administered using pantomime instructions.
Such methods may be an improvement over culturally or linguistically complex tests, but they
are not considered a comprehensive or unbiased alternative (Flanagan & Harrison, 2012). Just
because no expressive language is required to administer and complete the tasks does not
eliminate the occurrence of language-reduced communication that ensues during the evaluation.
Furthermore, others have also found that language-reduced tests are equally if not more affected
by culture and language factors than verbal based measures (van de Vijver, 2007; Rushton &
Jensen, 2005). Performances on major language-reduced tests are similarly influenced by
sociocultural factors, such as years of education (Ardila, Rosselli, & Rosas, 1989; Heaton,
Taylor, & Manly, 2003), level of acculturation (Arnold et al., 1994; Boone et al., 2007; Coffey et
al., 2005; Harris) and ethnicity (Coffey et al., 2005). It has been argued that the idea of a “culture
free” test is illusory, because the testing experience is a culturally specific exercise that requires a
significant level of overt and covert communication between the examiner and examinee (Nell,
2000). Also, language-reduced tests have been found to have limited predictive validity
(Giguero, 1989; Lohman et al., 2008). Despite limitations, the use of language-reduced tests is
very popular; 88 percent of surveyed practitioners choose to administer language-reduced tests

46

when evaluating immigrant populations (Sotelo-Dynega et al., 2011). More recently, crosscultural and multicultural assessment researchers have advocated more systematic approaches to
control simultaneously for assessment bias associated with cultural and language demands of the
test and the cultural and linguistic competencies characteristics of the examinee (Flanagan &
Ortiz, 2007).
Culture-Language Interpretative Matrix (C-LIM)
Flanagan and Ortiz (2001) proposed a systematic process when assessing individuals
from linguistic and cultural minority backgrounds. It was recognized that language and cultural
factors attenuate performance on standardized cognitive assessment among immigrant
individuals. As has been reviewed, language and cultural competence is likely to influence
performance on intelligence tests at multiple levels. At one level, the assessment process itself
involves culturally bound and culturally specific forms of communication and functional
dispositions. At another level, the language and cultural knowledge embedded within the tests
are likely to affect performance. What Flanagan and Ortiz (2001) proposed was a method to
increase the reliability and validity of assessment administration and performance interpretation.
The process was meant to enable selection of more appropriate tests and to provide an a priori
expected pattern of performance based on the cultural and linguistic loadings of the test and also
the acculturation levels of the examinee.
Based on expert consensus and review of the available literature, Flanagan and Ortiz
(2001) developed the Culture-Language Interpretative Matrix (C-LIM). The C-LIM is a
bidimensional (i.e., Language vs. Culture) three by three matrix (i.e., high, moderate, and low
saturation levels; See Table 2). It allows practitioners to examine possible systematic bias in the
assessment procedures and content of the tests. For example, immigrant individuals are likely to

47

perform more poorly on tests that are linguistically demanding and culturally loaded (e.g., verbal
tests), compared to tests with lower language and cultural demand (e.g., language-reduced tests).
Also, the variability in performance is assumed to vary as a function of the test and the
examinee’s characteristics. Flanagan and Ortiz categorized an individual’s level of acculturation
into three broad categories based on how culturally different they are from the host culture.
Individuals deemed Slightly Different are those that are third generation or later, have very high
levels of English language competence (e.g., advanced BICS and Emerging CALP), and have a
high level of acculturation. Those deemed Moderately Different are those who are second
generation, have moderate levels of English language competences (e.g., intermediate to
advanced BICS), and have a moderate level of acculturation. Those deemed Markedly Different
are likely to be first generation, have very low levels of English language competences (e.g.,
early BICS) and have a low levels of acculturation (see Table 2). Thus, the combination of test
characteristics (e.g., levels of linguistic demand and levels of cultural loading) and the
individual’s acculturation characteristics (e.g., slightly, moderately, or markedly different) are
used to estimate, in standardized points, the level of performance (see Table 2; Flanagan & Oritz,
2007). This is a promising approach but there has be limited empirical support.

48

Table 2
Culture-Language Interpretation Matrix
(Developed from Flanagan & Harrison, 2012)

Language Loading

Culture Loading

Low

Moderate

High

Low

Moderate

High

Slightly Different 3-5 points

Slightly Different 5-7 points

I Slightly Different 7-10 points

Moderately Different 5-7 points

Moderately Different 7-10 points

Moderately Different 10-15 points

Markedly Different 7-10 points

Markedly Different10-15 points

Markedly Different 15-20 points

Slightly Different 5-7 points

Slightly Different 7-10 points

Slightly Different 10-15 points

Moderately Different 7-10 points

Moderately Different 10-15 points

Moderately Different 15-20 points

Markedly Different 10-15 points

Markedly Different 15-20 pints

Markedly Different 20-25 points

Slightly Different 7-10

Slightly Different 10-15 points

Slightly Different 15-20 points

Moderately Different 15-20

Moderately Different 15-20 points

Moderately Different 20-30 points

Markedly Different 20-25

Markedly Different 20-25 points

Markedly Different 25-35 points

49

A number of dissertations have examined the clinical utility of the C-LIM with
immigrant youth that were largely supportive of the framework. Nieves-Brull (2006) examined
the utility of C-LIM with monolingual and bilingual samples using a Wechsler scale (i.e., WISCIV). The mean squared differences between the two groups were consistent with the predictions
of the C-LIM model. A greater group of mean differences was more associated with culturally
and linguistically demanding tasks than with less demanding tasks. The average difference
between monolinguals and bilinguals ranged from 8.76-16.08 points on verbal tasks (i.e.,
Vocabulary, Similarities, and Comprehension subtests) and between .21-.1.99 on languagereduced tests (i.e., Block Design, Matrix Reasoning, and Concept Formation subtests). Dynda
(2008) examined the association between the WASI and language competency with immigrant
and non-immigrant elementary school students. No differences emerged on language-reduced
tasks (e.g., Matrix Reasoning and Block Design), but significant differences were observed on
verbal tasks (e.g. Vocabulary and Similarities). Similarly, Sotelo-Dynega (2007) examined the
association of acculturation measures and verbal and language-reduced IQs, and found a relation
between the level of language competence and the overall cognitive performance (including
verbal and language-reduced). Those with low language competence performed within the
borderline-impaired range; those with intermediate to advanced language competence performed
within the below average range; and those with proficient language competence performed
within the average range. Verdorasa (2007) tested the utility of C-LIM with a preschool bilingual
sample using the Differential Abilities Test-Second Edition (DAS-II). The results showed
systematic outcomes between English language competence and performance on verbal and
language-reduced clusters. The mean difference between those with high language competence
and the low language competence was 13.65 standardized points on the verbal index and 5.9

50

standardized points on the language-reduced index. The above reviewed studies were relatively
consistent in their findings across three different cognitive batteries and samples. However, all of
the above reviewed studies examined mean group differences and implemented proxy measures
of acculturation (i.e., language competence, length of stay). Kranzler and colleagues (2010)
argued that the model is not substantiated when examining within-individual performance.
Kranzler and colleagues (2010) examined the utility of the C-LIM with a diverse sample
of 46, 5 to18 year-old immigrant students by using a within-subject analytical strategy (Kranzler,
Flores, & Coady, 2010). The results showed generally consistent patterns of performance as
predicted by the C-LIM within the overall sample. However, such patterns were not statistically
significant differences based on within-subject analyses. Overall, a large proportion of the
sample did not reflect pattern of scores that were predicted by the C-LIM model. They found
13% of all participants had scores consistent with the patterns predicted for the effects of
linguistic demand, cultural loading, and their combination. Almost half of the sample had scores
that did not follow any of the predicted patterns in the C-LIM. The authors concluded that their
results, when using the Woodcock-Johnson Tests of Cognitive Abilities among a diverse sample
of immigrant children, did not substantiate the use of C-LIM model. What this study highlights is
the importance of examining within-subject patterns of associations between acculturation and
cognitive performance for contributing important information for individual assessment.
Although the reviewed studies show relatively consistent support for the C-LIM model at
the group level, there are important limitations when examining the model at the individual level,
as highlighted by the Kranzler and colleagues (2010) study. The C-LIM utility was also
challenged when attempting to discriminate between ELL and non-ELL students (Styck &
Watkins, 2014). It also remains unclear if the process of acculturation has unique contributions to

51

understanding the effect of culture and language on cognitive performance because there was no
systematic evaluation of acculturation levels. The one study that used a within-subject design did
not analyze the level of acculturation or even the generation status of the ethnically,
linguistically, and culturally heterogeneous sample assessed. It is not clear if there are unique
cultural and linguistic factors within a specific ethnic or linguistic group that may be more
pronounced in how they substantiate the model, or in how acculturation variables relate to verbal
and language-reduced performances. There are no available studies that have examined the
clinical utility of IQ testing among Arab Americans who possesses unique cultural and linguistic
attributes. There have been some studies that have evaluated Arab children using Wechsler
scales outside of America.
Cognitive Assessment with Arab Children
There have been no published studies that have examined the cognitive performance of
Arab American youth or Arab youth in Europe. However, there have been some studies of
examining the cognitive performance of Arab youth residing in Arab countries and in Israel. The
available studies are mostly normative studies using the Raven’s Colored Progressive Matrices, a
language-reduced reasoning task (i.e., Raven’s Matrices). Results show that the mean
performance varies from the lowest mean of 78 to the highest mean of 87 (Lynn, 2006; Dickins,
Sear, & Well, 2007). However, the available studies were often conducted in the late 1980s and
early 1990s, and the normative comparisons used were based on a sample of British children.
The studies also lacked adequate controls for important educational and geographical setting
variables. The most detailed studies of Arab youth were conducted in Israel.
There have been several studies using full intelligence assessment batteries with Arab and
Israeli children that have revealed some important findings. Liblich and Kugelmass (1981)

52

looked at the patterns of intellectual ability of Arab school children (age 6-16) in Israel using the
WISC-R. This particular study followed up prior studies that analyzed the intellectual profiles of
much younger Arab children using the WPPSI (Kugelmass et al., 1974). Results overall
indicated higher performances on verbal than language-reduced tasks. This difference was much
more pronounced among younger age groups (e.g., 6-12 years old). Arab children were also
compared to Israeli children. In general, Israeli children scored higher, but the difference nearly
disappeared when comparing samples that lived in the same city and had comparable SES
factors. However, some significant differences continued to be observed on visual perceptual and
speed-tests. The authors posited some culturally based explanations of the observed differences.
For instance, they highlighted that Arab cultural attitudes toward time and speeded tasks are
negative, citing popular Arab sayings that “speed is from the devil” and implying preference to
work in a relaxed, collaborative, and unhurried fashion. Such speculations were supported by the
higher performance Picture Completion subtest, which is an untimed visual-perceptual
categorical reasoning task. Another cultural-specific factor that may explain weaker visual
perceptual abilities may be associated with Islam and its prohibition of pictorial arts and
sculptures. Because of these popular cultural values, there may be less developmental interaction
with drawing pictures and construction activities in Arab homes. The authors argued that their
analysis of Arab youth’s performance on the WISC-R challenges the notion that languagereduced performance based tests are more “culturally fair.”
Ethical Obligations and Multicultural Competency
According to the American Psychological Association (APA), developing multicultural
competence is essential for practitioners and trainers in meeting general competency and ethical
imperatives (APA, 2003). The APA published two guiding documents in 2002 and 2003

53

(Guidelines on Multicultural Education, Training, Research, Practice, and Organizational
Change for Psychologists; Ethical Principles and Codes of Conduct) that highlight the
importance of moving away from a monoculture lens towards recognizing and embracing a
multicultural framework that will lead to more precise assessment and ultimately more effective
treatment (APA, 2002, 2003). Developing cultural competence allow for better delivery of
psychological assessment services that are more consistent with established ethical and
professional guidelines. Sue and colleagues (1998) outlined four dimensions in the process of
developing cultural competence. First, it requires involves being aware of one’s own
assumptions, values, biases, and stereotypes about an ethnic minorities. Second, it involves
developing knowledge and understanding regarding one’s own worldview and that of one’s
client; specific knowledge regarding the culture of one’s clients and underlying sociopolitical
influences. Third, cultural competence requires acquisition of specific, culturally appropriate
assessment and communication skills necessary to effectively work with ethnic minority groups.
Lastly, it also requires the development of core cultural competencies based on new theories,
practices, policies, and organizational structures. In light of these guidelines, intellectual
assessment among minorities has received very little attention that would allow practitioners to
enhance their multicultural competence and provide ethically and professionally responsible
services.
A crucial component in developing cultural competence among clinicians is increasing
one’s knowledge base about their clients, raising awareness of their own potential biases, and
continually enhancing culturally sensitive services (Rogers, Ingraham, Bursztyn, et al., 1999).
With any group of recent immigrants, clinicians, educators, and researchers will increase their
cultural competence by considering how well they adjust and the effect of other relevant

54

demographic factors (e.g., age, gender). Meeting these goals will require a better understanding
of how the process of acculturation affects intellectual assessment. Improved understanding of
the acculturation influence on intellectual performance will help better interpret results that
reflect a cultural difference from results that are indicative of cognitive processing deficits.
Rationale and Aims of the Proposed Study
Assessment of global cognitive functioning has played an important societal role across
clinical, vocational, and educational settings. The available research has not provided an
adequate analysis of how the acculturation process influences performance on cognitive tests by
immigrant populations. Cultural exposure to the host culture and the development of language
competence has been documented to have dramatic influence on cognitive outcomes (Flannigan
& Ortiz, 2007). Although a substantial amount of research has highlighted significant moderating
effects of some acculturation variables on verbal and language-reduced intelligences tests, most
of that research has been conducted with adult samples. Additionally, the available studies with
children have often utilized proxy measures of acculturation (Flanagan & Harrison, 2012).
Acculturation is a dynamic process that leads to changes in the cultural, affective, and cognitive
domains (Sam & Berry, 2006; Nell, 2000). Proxy level conceptualizations of acculturation may
not provide an adequate account of the association between the acculturation process and
cognitive assessment. Not accounting for acculturation variables will likely threaten the
reliability and validity of the outcomes, lessen the quality of care provided, and fall short of
meeting professional multicultural competence and ethical standards. Thus our understanding of
how the acculturation process affects cognitive development and assessment is fragmented and
has not resulted in foundational clinical implications for assessing immigrant youth.

55

Models such as the C-LIM have provided a systematic framework to understand how the
level of linguistic and cultural loadings within IQ tests could affect performances and help
differentiate between cultural differences and disability. This framework has found some
empirical support, but important limitations have been highlighted (see Kranzler et al., 2010).
The current study is designed to contribute to the gaps in the literature, as well as to document
the clinical utility of intelligence test with Arab American youth. No study has examined the
influence of a systematic conceptualization of acculturation on performance of intelligence tests
and how well such tests predict academic achievement. This will also be the first study to
examine the bidimensional nature of acculturation and its potential association with verbal and
language-reduced cognitive performance. It remains unclear if enculturation factors have
independent association with performance on intelligence tests, and whether that association
differs across verbal and language-reduced tests.
The proposed study has three main aims. The first aim is to examine whether there is a
clinically meaningful difference between the Arab American youths performances on verbal and
language-reduced IQ tests. The second aim is to examine the association between proxy and
systematic acculturation factors with verbal and language-reduced IQ performances. The third
aim is to identify how well IQ predicts academic achievement and whether that association is
moderated by acculturation factors. Despite potential bias in the intelligence assessment among
immigrant groups, it has been found that they may not be biased when it comes to their
predictive validity. The predictive validity of IQ among Arab American youth is unknown.

56

CHAPTER III: METHODS
Population of Interest
Youth. A total of 83 non-clinical, community based Arab American males between the
age of 12 and 17 were recruited for the study, but only 80 participated. Three participants were
dropped because they were currently undergoing aggressive epilepsy treatment. Approximately
39 percent (n = 31) were between the age of 12 and 14, and approximately 61 percent (n = 49)
were between the age of 15 and 17. The average age was m = 15.34, sd = 1.7. All participants
were general education students at charter and public schools. Approximately 55 percent (n = 44)
of the youth were born in the US, while 45 percent (n = 36) were born outside the US. All
participants were Arab American, lived in the US for a minimum of three years, and spoke fluent
English. A total of nine youths preferred using Arabic over English in general, while the rest
preferred English. The sample was predominantly composed of participants from Yemeni origins
(77.5 %), but included others from Iraq, Jordan, Lebanon, Syria, and Palestine. At least one
parent provided demographic information for each youth (see Tables 1 & 2).
Table 3
Country of Birth for Adolescents and Parents
Country of Birth
Adolescents (n = 80)
n
%
United States
44
55.0
Yemen
24
30.1
Iraq
4
5.0
Lebanon
3
3.0
Jordan
3
3.8
Palestine
2
3.0
Syria
1
1.3

57

Parents (n = 80)
n
3
61
8
1
2
3
1

%
3.8
7.36
10.0
1.3
3.0
3.8
1.3

Table 4
Demographic Characteristics of Adolescent Participants
Total Participants Total Percentage
Age
12-14
49
62.2
15-17
31
38.8
Language Preference
English
71
88.8
Arabic
9
11.2
Language Used at Home
Mostly English
15
18.8
Mostly Arabic
27
33.7
Both Equally
38
47.5
Language Use at School
Only English
22
27.5
Mostly English
57
71.3
Mostly Arabic
0
0
Both Equally
1
1.3
Friends
Mostly Arab
79
98.7
Mostly Non-Arab
1
1.3
Place of Birth
US
44
55
Outside US
36
45

Parents. Out of 83 parents who provided consent to participate in the study, 80 were
included in the analysis. Seventy-six of 80 reporters were fathers. Parents’ socio-economic
statuses were generally in the low range (Hollingshead Index score m = 30.06, sd = 8.93, Median
= 28, Range = 8 to 53; Hollingshead, 1975). Approximately 90 percent of the reporters were
married to the child’s other parent. Eighty eight percent (n = 70) of the parents were born outside
the US. Approximately 70 percent (n = 56) of the mothers and 39 percent (n = 31) of the fathers
earned less than a high school diploma. Slightly over 11 percent (n = 9) of the mothers and 26
percent (n = 21) of the fathers had some college or higher. Only four of the 80 mothers had at
least part-time employment. The majority of fathers were employed; 66.25 percent (n = 53) were
full-time employed, 12.5 percent (n = 10) were part-time, and 21.25 percent (n = 17) were

58

unemployed. The majority of those employed were blue collar workers in primarily the
manufacturing and service industry (e.g., restaurant staff, cashiers). Total family SES was in the
low range (Hollingshead Index m = 30.06, sd = 8.93). The average household consisted of six to
seven individuals (m = 6.68, sd = 1.66). A total of 34 percent (n = 27) reported communicating
with their child primarily in Arabic, 18 percent (n = 15) primarily in English, and 48 percent (n =
38) reported using both languages equally (See Tables 2 & 3).
Table 5
Parents’ Education, Employment, and Income
Variable
Mother (n =80) Father (n = 80)
Education
Grade 0-3
23
5
Grades 4-6
22
3
Grades 7-9
11
23
Grades 10-12
14
26
Some college
6
13
Bachelors or higher 3
8
Employment
Full Time
2
53
Part Time
2
10
Not working
76
17
Family Income (frequency/%)
< 10, 000
19 (23.8%)
10,001-20,000
26 (32.5%)
20,001-30,000
12 (15%)
30,001-40,000
12 (15%
> 40,001
11 (13.8%)
Family SESa
Mean (SD)
30.06 (8.93)
Range
8-53
a
Family SES based on Hollingshead Index

59

Exclusion criteria. Participants with any psychiatric, developmental, neurological, or
behavioral disorders were excluded from the study. The exclusions included children receiving
special education services at their schools for neurodevelopmental disabilities (e.g., ASD,
ADHD, and Specific Learning Disability). Additional exclusion criteria included individuals
with psychological (e.g., depression), neurological (e.g., seizures, traumatic brain injury), or
cognitive impairment (e.g., intellectual disability). Also, those who resided in the US for less
than three years or did not identify as Arab American were excluded. A total of three individuals
were dropped from the analyses because they had a history of or were currently undergoing
aggressive treatment for epilepsy. Female participants were also excluded because controlling for
possible examiner and examinee interaction as a function of gender, ethnic, and religious
background would have introduced great pragmatic challenges that could not be realistically
controlled for in the current dissertation study. As a result only a single male examiner evaluated
only males.
Recruitment and Assessment
Recruitment procedures were multifaceted. Overall, recruitment was conducted through
schools, a community center, and by posting flyers in community centers, mosques, schools, and
local markets. Most of the participants were recruited from two schools (one public and one
charter) and a community center. Those affiliated with the community center were all enrolled in
after school programs located in a public high school. The primary investigator cooperated with
school and community center staff to introduce the study and pass out study descriptions,
consent, and demographic forms. School staff and community center coordinators collected
completed consent forms and some demographic forms. The parent demographic forms were
mostly completed through an in-person or over the phone interview by the primary investigator.

60

Students were tested after consent and demographic information were collected. Those recruited
via flyers were provided with an overview of the study and in some cases e-mailed study
descriptions and consent forms. Those interested were scheduled for testing after parent consent
and demographic information were obtained.
All testing was conducted in English only and in accordance with standardization
procedures of the instruments used. The participants were all individually assessed in quiet and
private rooms located in two different schools and one community center. The individual student
evaluation included measures of verbal and language-reduced intellectual abilities, measures of
basic academic reading and mathematics skills, brief demographic information report, and a selfreport acculturation scale. Administration time for each student took approximately 60 minutes.
Upon completion of the study requirements, each youth received a 20-dollar gift card. Three 50dollar gift cards were also provided to three different parents who were selected through a raffle.
See Appendices D, E, and F for consent/assent and recruitment materials.
Variables of Interest and Materials
Demographic variables. Guardians and youth completed brief demographic forms that
were developed by the principle investigator (see Appendixes A & B). The parent form included
the following information about the family: country of origin, length of stay in the U.S. (i.e.,
number of years resided in the U.S.), relation to child, child’s age of arrival, languages spoken in
the home, total family members, occupation status, household income level, parent education
levels, and screening questions of learning, psychological, developmental, or neurological
impairments.
Guardian education level. Defined as follows: Scale from 1 to 7: 1= third grade or below;
2 = fourth grade up to sixth grade; 3 = seventh grade up to ninth grade; 4 = 10th grade up to high

61

school graduate; 5) beyond a high graduate and up to two years of college; 6) bachelor’s degree;
7) master’s degree or higher.
Income level. Defined as follows: scale from 1-10: 1 = 5,000 or below; 2 = 5,001-10,000;
3 = 10,001-20,000, 4 = 20,001-30,000; 5 = 30,001-40,000; 6 = 40,001-50,000; 7 = 50,00160,000; 8 = 60,001-70,000; 9 =70,001-100,000; 10) More than 100,000.
These measures, used for assessing income and education level, have been viewed as
adequate measures to control for SES effects (Bialystok, 2001). The Hollingshead Index (1975)
method was used to construct the SES variable that took into account type of job and education
levels. The youth demographic forms included questions about language preference, friendships,
religiosity, ethnicity, place of birth, school attending, school grades, and city of residence.
Intellectual assessment. The Wechsler Abbreviated Scale of Intelligence -Second Edition
(WASI-II; Wechsler, 2011) was used to estimate verbal and language-reduced intelligence. The
WASI-II provides four major indices. It has an overall composite score that is comprised of all
four subtests (i.e., FSIQ-4) and another overall abbreviated measure that is comprised of one
verbal and one language reduced subtests (i.e., FSIQ-2). The WASI-II is also structured into an
abbreviated measure of Verbal Comprehension Index (VCI) and Perceptual Reasoning Index
(PRI).
The VCI may be referred to as an estimated verbal IQ [V-IQ]) and is composed of two
subtests, Vocabulary and Similarities. The Vocabulary subtest has 31 items, including three
picture items and 28 verbal items. Vocabulary requires the participant to orally express
definitions for words that increase in difficulty. For picture items, the examinee names the
object, and for verbal items, the examinee defines words that are presented visually and orally.
The Vocabulary subtest is designed to measure word knowledge and verbal concept formation. It

62

is also believed to be a reliable estimate of the examinee’s crystallized intelligence, fund of
knowledge, learning ability, degree of language development, and long-term memory (Wechsler,
2011; Lezak et al., 2012). The Vocabulary subtest has strong internal consistency among youth
populations (Average Cronbach’s Alpha = .91; Wechsler, 2011).
The Similarities subtest includes three basal picture items and 24 verbal items. The
examinee is orally presented two words that represent common objects or concepts and are asked
to describe how they are similar. The Similarities subtest is designed to measure abstract verbal
concept formation and categorical reasoning. It is also assumed to be a reliable indicator of
crystallized intelligence, auditory comprehension, categorical thinking, and verbal expression
(Wechsler, 2011, Lezak et al., 2012). It also has strong internal consistency among youth
populations (Average Cronhbach’s Alpha = .89; Wechsler, 2011).
The PRI may be referred to as an estimated language reduced IQ (LR-IQ). It is composed
of two subtests (Block Design and Matrix Reasoning) that involve reduced levels of language
instruction, comprehension, and processing to perform the tasks. The Block Design subtest is
composed of 13 modeled or printed two-dimensional geometric patterns that the examinee
replicates within specified time limits using two-color blocks. It is designed to assess the ability
to analyze and synthesize abstract visual-spatial stimuli. It also involves nonverbal concept
formation and reasoning and is regarded as an indicator of broad visual intelligence, fluid
intelligence, and visual perception and organization (Lezak et al., 2012). It also requires
simultaneous processing, visual-motor coordination, and learning and problem solving ability
(Wechsler, 2011, Sattler, 2008). Block Design has strong internal consistency among youth
populations (Cronbach’s Alpha = .89, Wechsler, 2011). The second language reduced subtest is
Matrix Reasoning, which has 30 items. The examinee views a series of incomplete matrices and

63

completes them by selecting the correct response from a set of choices. Matrix Reasoning is
viewed as a measure of fluid and nonverbal reasoning. It taps into broad visual intelligence,
perceptual organization, classification, spatial ability, and knowledge of part-whole relationships
(Sattler, 2008; Wechsler, 2011). Matrix Reasoning has strong internal consistency among youth
populations (Cronhbach’s Alpha = .87).
The four subtests and two subtests estimated FSIQ scores have strong concurrent validity
with more comprehensive IQ measures ranging from .78 to .91 (Wechsler, 2011). The
correlation between the WASI-II four subtests FSIQ had moderate to strong correlation with
Wechsler’s Word Reading and Numerical Operations. Please see table (Table 6) below for
individual composite and individual subtests correlations. The total estimated administration
time for the WASI-II ranges from 30 to 45 minutes (Table 6).
Table 6
Normative Sample Correlations between the WASI-II and Wechsler’s Academic Outcomes
Wechsler Academic Achievement Tests
WASI-II Subtests/ Composites
Word Reading
Numerical Operations
Block Design
.46
.50
Matrix Reasoning
.53
.60
Vocabulary
.64
.53
Similarities
.62
.53
VCI
.68
.56
PRI
.59
.65
FSIQ-4
.70
.68
FSIQ-2
.67
.65

Measure of acculturation. The Abbreviated Multidimensional Acculturation Scale
(AMAS; Zea, Asner-Self, Birman, & Buki, 2003) is a bilinear and multidimensional scale used to
assess acculturation toward American mainstream society (AMAS-Total American) and
maintenance of heritage culture or enculturation (AMAS-Total Arabic). The AMAS-Total
American includes three subscales: English Language Competence, American Cultural

64

Competence, and American Cultural Identity. The AMAS-Total Arabic also includes three
comparable subscales: Arabic Language Competence, Arabic Cultural Competence, and Arabic
Identity Levels. Thus, the scale allows for unidimensional and bidimensional conceptualizations
of acculturation across multiple domains.
The AMAS has been used and analyzed across diverse samples of Hispanic and Asian
young immigrants and demonstrated robust reliability indicators across groups and across scale
domains (Cronbach’s alpha ranging from .90 to .97; Zea et al., 2003). The scale also
demonstrated strong internal consistency with the current study sample when examining the
entire scale and individual scales (see Table 7).
Table 7
Internal Consistency of AMAS with Arab Americans as Measured by Cronbach’s Alpha
AMAS-Scales
English
Arabic
Total Scale
.82
.85
Language Competence
.84
.89
Cultural Competence
.79
.77
Cultural Identity
.77
.93

The structure of the questions allowed for relatively facile adaptations by simply
changing the reference group when referring to the heritage acculturation (e.g., “I think of myself
as_________” or “I have a strong sense of being ________).” There are 12 questions that survey
cultural identity, 12 questions that survey cultural competence, and 18 questions that survey
language competence on the acculturation and enculturation domains (AMAS-Total American
and AMAS-Total Arabic). Each question was answered using a Likert-Scale ranging from 1
(strongly disagree) to 4 (strongly agree). The scale is designed so that higher scores reflect higher
levels of acculturation and enculturation. Approximately a fourth grade reading level is required

65

for participants to individually complete the scale. The completion of the AMAS acculturation
scale took approximately 5 to 10 minutes.
Academic achievement. The Word Reading and Numerical Operations subtests from the
Wechsler Individual Achievement Test-Third Edition (WIAT-III; Wechsler, 2009) will be used as
measures of academic outcomes. The Word Reading subtest measures the rate and accuracy of
word reading. The examinee is asked to read aloud from a list of words that increase in
difficulty. This subtest allows for a timed and untimed interpretation. Only untimed
performances will be used for the current study. The Numerical Operation subtest is an untimed
task that presents the examinee with a wide range of math calculation problems to be solved in a
paper and pencil format without the aid of a calculator. The subtest assesses a range of skills that
include basic arithmetic, basic operation with integers, geometry, algebra, trigonometry, and
calculus. Both achievement subtests are standardized on kindergarten through 12th grade and
have demonstrated strong predictive validity with overall school achievement and excellent
internal reliability (Word Reading, r = .98; Numerical Operations, r = .93). These two subtests
have also been shown to be very reliable across age, gender, and ethnic backgrounds (Wechsler,
2009). Administration of both academic tasks takes between 15 to 20 minutes.
Data Analysis
The SPSS (version 20) software was used to conduct descriptive and inferential statistical
analyses. For the first question, a paired-sample t-test was conducted to examine withinindividual differences between verbal and language reduced IQ. Cohen’s d effect size was used
to determine the magnitude of the difference between verbal and language-reduced IQ. The
difference between V-IQ and LR-IQ was analyzed by creating an IQ Difference variable that was
generated by subtracting V-IQ from LR-IQ. A hierarchical regression equation was conducted to

66

evaluate what important sociodemographic variables explained IQ Difference. Two hierarchical
regression analyses were conducted to examine the association between proxy and systematic
measures of acculturation with V-IQ and LR-IQ. Follow-up analyses were conducted using
multiple regression procedures to examine which aspects of acculturation best explained
variance in V-IQ and LR-IQ performances. A multiple regression equation was used to evaluate
how well the FSIQ-4 predicted basic reading and math skills. An interaction term was added to
explore the possible moderating influence of acculturation levels on the predicative association
between IQ and achievement. All terms were centered when interaction terms were entered to all
equations.
Power analysis. A statistical power approaching .80 is considered adequate for rejecting
the null hypothesis if it were false. For multiple regression procedures with up to four
independent variables, given an alpha level of .05, assuming a small to moderate effect size (e.g.,
.15), a sample size of up to 80 participants will achieve a total power of .82.
Three Study Aims, Questions, Rationales, and Analyses
Aim One
The first aim was to examine whether there existed a significant difference between Arab
American youth’s performances on V-IQ and LR-IQ as predicted by the C-LIM framework.
Studies with immigrants have identified that verbal IQ is often underestimated because of
acculturation factors (Flanagan & Ortiz, 2007). Results have found that language-reduced IQ
assessment can be less sensitive to cultural and language factors and may be a more reliable
indicator of overall cognitive ability and possibly a better predictor of important academic skills
(Flanagan & Ortiz, 2007; Sattler, 2008).

67

Question 1: Is there a significant difference between the performance on V-IQ (i.e.,
estimated Verbal IQ) and LR-IQ (i.e., estimated language-reduced IQ) among Arab American
youth? As predicted by the C-LIM framework, it was anticipated that Arab American youth’s
scores would be significantly greater on LR-IQ than on V-IQ. A paired-sample t-test was used to
examine how big the difference was between V-IQ and LR-IQ. Further, the difference between
V-IQ and LR-IQ was regressed against important sociodemographic variables. A moderate to
large effect size (Cohen’s d > .5) represented a clinically meaningful difference.
Aim Two
The second aim was to evaluate the association between acculturation variables with VIQ and LR-IQ performances. As outlined in the literature, there is a considerable influence of
acculturation factors on intellectual assessment. Moreover, there has been inconsistent
conceptualization of the acculturation process among available studies that have examined
intellectual capability among immigrant populations. Some studies have used proxy and
sociodemographic factors to account for possible acculturation influence. It has been argued that
acculturation is a dynamic process that is not sufficiently captured by proxy variables and that a
systematic bidimensional assessment instrument better organizes and conceptualizes the
complexity involved. If more systematic measures are important, it remains unclear whether
there is an added explanatory benefit of conceptualization acculturation as a unidimensional or
bidimensional construct when assessing its potential effect on IQ performance. The following
question was aimed at evaluating the relation between the acculturation process and IQ
performances and determining whether there was a difference in explanatory power when using
proxy and more systematic evaluations of acculturation.

68

Question 2a: Is acculturation associated with performance on measures of verbal IQ
among Arab American youth? The influence of acculturation on V-IQ was examined using
hierarchical regression equation. Sociodemographic variables were entered first followed by
acculturation factors that were significantly associated with V-IQ. This allows for evaluating the
association of acculturation independent of sociodemographic factors.
Question 2b: Is acculturation associated with performance on measures of LanguageReduced IQ among Arab American youth? The influence of acculturation on LR-IQ was
examined using a hierarchical regression equation. Sociodemographic variables were entered
first followed by acculturation factors that were significantly associated with LR-IQ. This allows
for evaluating the association of acculturation independent of sociodemographic factors.
Aim Three
The third aim was to evaluate how well FSIQ-4 predicted basic academic skills and whether
there was a moderating effect of the acculturation levels toward American mainstream society.
On average, there is a moderate to strong predictive association between IQ and standardized
academic measures when using the Wechsler scales (Watkins, Lei, & Ganivez, 2007; Gagne &
St. Pere, 2001; Wechsler, 2011, 2009). Specifically, verbal IQ was found to be the strongest
single predictor of word reading, reading comprehension, and math calculation (Mayes &
Calhoun, 2007). Overall, Full-Scale IQ (i.e., combined verbal and language-reduced IQs) has
been viewed as the most robust predictor of academic achievement (Mayes & Calhoun, 2007;
Sattler, 2008). Studies with immigrant groups have shown similar but more variable associations
between intelligence performance and academic achievement (Rushton & Jensen, 2005; Flynn,
2008). It is not clear if such variability is partly explained by levels of acculturation or language
competence levels. In general, despite potential bias in the process of intellectual assessment

69

among immigrant groups, it has been found that intelligence tests may be unbiased in how they
predict academic achievement (Rushton & Jensen, 2005). This can hold important clinical utility
when examining immigrant youth who may be immersed in two different cultural and language
backgrounds with no viable assessment alternatives. The capability to estimate learning potential
more robustly is a clinically and ethically important undertaking. It is unknown whether this
predictive association can be generalized to understudied groups such as Arab American youth
and whether acculturation is an important moderating variable in that association. To meet this
aim, the following questions were put forth:
Question 3. Is FSIQ-4 predictive of academic outcomes (e.g., Word Reading and Numerical
Operations)?
3a. FSIQ-4 is expected to be moderately to strongly predictive of Word Reading.
3b. FSIQ-4 is expected to be moderately to strongly predictive of Numerical Operations.
Question 4. Is the predictive association moderated by acculturation?
It was anticipated that the predictive association between FSIQ-4 and academic achievement
will be influenced by level of acculturation toward American mainstream society. The predictive
association was anticipated to increase as the level of AMAS-American increased.
A hierarchical regression model for each outcome variable was completed (i.e., Word Reading
and Numerical Operations). The order of entry was centered FSIQ-4, followed by a centered USAMAS, and lastly a centered interaction term was included (i.e., FSIQ-4 X US-AMAS).
4a. The association between FSIQ-4 and Word Reading will be positively moderated by
US-AMAS.
4b. The association between FSIQ-4 and Numerical Operations will be positively moderated
by US-AMAS.

70

CHAPTER IV: RESULTS
Summary of Purpose
The overall goal of the study was to evaluate the clinical utility of the WASI-II among
Arab American youth by examining how well it predicts academic achievement and what effect
acculturation factors have on verbal (e.g., V-IQ) and language-reduced (e.g., LR-IQ) intelligence
tests. Three specific aims were outlined in the previous chapter. The first aim was to evaluate if
there was a meaningful difference between the youth’s performances on V-IQ and LR-IQ. The
second aim was to evaluate the association between proxy and systematic measures of
acculturation (i.e., rating scale acculturation assessment) with intelligence tests performances.
The third aim was to evaluate the predicative association between FSIQ-4 and basic academic
skills (e.g., reading and math) and determine whether acculturation moderates the relation. The
results are presented by providing descriptive, correlational, and inferential statistics. For
descriptive statistics, multiple tables are presented that provide an overview of the sample
characteristics. Matrices of correlations with means and standard deviations are presented for the
main study variables. Inferential analyses are organized around the research questions and stated
hypotheses. Exploratory analyses were added at the end of the chapter as a result of the outcomes
observed in the core analyses.
Sample Characteristics
The data were inspected for form, distribution, and completeness prior to their use in any
analyses. For non-demographic items, the mean score replaced missing responses to preserve
power. This was the most appropriate imputation strategy since there was a very small amount of
missing data. Approximately 0.1% of items required replacement. A Shapiro-Wilk’s test (p>.05;
Shapiro & Wilk, 1965; Razali & Wah, 2011) and visual inspection of their histograms, normal
Q-Q plots, and box plots showed that the IQ and achievement scores were normally distributed
71

for the entire sample (Cramer, 1998; Cramer & Howitt, 2004; Doane & Seward, 2011).
Inspection of Mahlanobis and Cook’s values indicated no presence of outliers for the main study
variables including acculturation, IQ, and achievement. The instruments and scales used in the
study had appropriate characteristics to allow for the use of standard, inferential, parametric
statistics. Please see Tables 6, 7, and 8 for means, standard deviation, correlations, skewness, and
kurtosis for the main variables.

72

Table 8
Intercorrelations of Study Variables, with Means, Standard Deviation, Kurtosis, and Skewness
1
2
3
4
5
6
7
8
9
1. 4-FSIQb
2. 2-FSIQc
.93**
3. V-IQ
.84**
.78**
4. LR-IQ
.88**
.81**
.50*
5. Reading
.51**
.50**
.56**
.32**
6. Math
.60**
.57**
.43**
.57**
.39*
7. Arab-AMAS -.10**
-.12
-.08
-.09
.19
-.05
8. US-AMAS
.28**
.24*
.29**
.22
-.26*
.28
.26
9. Age
.00
.06
-.01
-.01
-.01
.06
-.14
-.08
10. LOSx
.16
.11
.13
.15
.17
.09
-.27*
-.05
.23*
.15
.04
-.01
.25*
.04
.21
.09
.36**
-.12
11. SESy
Mean
89.29a
93.30a
88.51a
92.25a
89.16a
94.10a
61.70
67.15
15.27
(SD)
(11.20) (11.66) (11.03) (13.33) (13.20) (13.22)
(10.58) (9.17)
(1.70)
Kurtosis
.18
.03
.09
.14
-.27
.23
-1.24
1.69
-1.81
Skewness
1.44
.98
1.55
.28
-1.47
1.02
-1.04
.69
-1.45
a
Scores are based on a standard score, mean = 100, SD ± 15; b4-FSIQ comprised all four WASI-II subtests; c2FSIQ comprised the Vocabulary and Matrix Reasoning subtests; x Length of Stay (LOS) equals total years or US
residence; y Socio-economic status (SES) based on Hollingshead method (1975); Note: * P < .05, **P < .001

73

10

.24*
12.73
(3.36)
.05
4.01

11

30.06
(8.93)
.18
1.44

Table 9
Intercorrelation of Acculturation Subscales, Sociodemographic, and IQ
1
2
3
4
5
6
1. US-AMAS
2. English
.55**
Competence
3. US Culture
.57** .27*
Competence
4. US Identity
.57** .10
.10
5. Arab-AMAS
.26*
.16
.35** .15
6. Arabic
.15
.02
.28*
.03
.80**
Competence
7. Arab Culture
.24*
.01
.42** .03
.77** .62**
Competence
8. Arab Identity
.19
.05
-.08
.19
.30*
-.07
9. Length of Stay
-.07
-.01
-.08
-.12
-.29* .31**
10. SES
.36** .10
.20
.31** .10
-.01
11. Income
.28*
.10
.15
.25*
.11
-.01
12. Mother’s
.21
.08
.23*
.12
-.17
-.22*
Education
13. Father’s
.24*
.08
.11
.19
.07
.06
Education
14. Age
.08
.04
-.10
.03
-.14
.02
15. V-IQ

.29*

.35**

.06

-.17

-.08

-.16

16. LR-IQ

.22*

.17

.01

.26*

-.09

-.10

Note: *<.05, **<.001

74

7

.11
.22
.11
.09
.09
.13
.14
.15
.02

8

9

10

11

12

13

14

.28*
.13
.06

.24*
.25*
.31*

.90**
.34*

.16

.02

-.02

.44**

.03

.33*

-.16

.20

-.12

-.05

-.14
.39**

.21
.10

.12

.13

-.01

-.06

-.01

.02

.15

.25*

.27*

.15

.01

-.01

15

.00

50**

Table 10
Intercorrelations among WASI-II and AMAS Scales
1
2
3
4
5
1. US C.C.a
2. US ID.b
.10
3. Eng. C.c
.27* .10
4. A.C.C.d
.42** .03
.02
e
5. A. ID.
-.08
.19
.05
.03
6. Arabic
.28* .03
.02
.62** -.07
7. Vocabulary -.01
.20
.40** -.24
.10
8. Similarities -.13
.08
.15
-.05
.11
9. MRf
-.04
-.00
.10
-.10
-.00
10. BDg
.02
.27* .20
.02
.27**
h
.09
.29** .24* -.08
.11
11. NO
12. WRi
-.03
.06
.32
-.25* -.02
.06
.17
.35** -.15
.12
13. V-IQj
14. LR-IQk
.01
.26* .17
-.03
.03
15. 4-FSIQ
.01
.27* .26* -.12
.11
16. 2-FSIQ
-.01
.26
.26* -.15
.01

6

7

8

9

10

11

12

-.17
-.12
-.12
-.10
-.14
-.32**
-.16
-.10
-.16
-.13

.54**
.38**
.38**
.41**
.56**
.89**
.39**
.72**
.76**

.47**
.47**
.40**
.35**
.85**
.46**
.75**
.46**

.66**
.54**
.41**
.48**
.90**
.82**
.85**

.55**
.26*
.45**
.91**
.80**
.6**

.39**
.43**
.57**
.60**
.57**

.56**
.32** .50**
.51** .88** .87**
.50** .81** .81** .93**

a

13

14

15

16

= AMAS American Cultural Competence; b = AMAS American Identity; c = AMAS English Competence; d = AMAS Cultural Competence;
= AMAS Arab Identity; f = Matrix Reasoning; g = Block Design; h = Numerical Operation; I = Word Reading; j = Verbal IQ;
k
= Language-Reduced IQ
* p < .05, **p<.001
e

75

Acculturation levels. The sample acculturation characteristics are described below,
followed by examining the internal reliability of the AMAS acculturation scale, as measured by
Cronbach’s Alpha. Further, the means and intercorrelations between systematic and proxy
acculturation, and sociodemographic factors are summarized (see Tables 7 and 8). The entire
sample was composed of first and second generation Arab American male immigrants. A total of
44 (55%) males were born in the US, while 36 (45 %) were born outside the US. Most of those
born abroad came at a very young age. The Length of Stay (LOS) reflected the total number of
residence years in the US (m = 12.73 sd = 3.36). Approximately 89 percent (n=71) of the sample
preferred speaking English over Arabic. However, 100 percent of the sample spoke English and
did not require any modifications or interpretations of the assessments. Each male completed the
AMAS Acculturation Scale (Zea et. al., 2003) to assess their acculturation, bidimensionally,
across language competence, cultural competence, and cultural identity domains (See Table 11
for means, range, and internal reliability and Table 10 for subscales intercorrelations).
Overall, the results show higher US-AMAS than Arab-AMAS means. There was also a
positive association between Arab-AMAS and US-AMAS (r (78) =.26, p = 02). There were
differences in level of acculturations within American and Arab cultural domains. A total of 72
(90%) participants rated higher Arab than American identity, but 64 (80%) participants endorsed
higher English and American cultural competences than Arabic language and Arab culture
competences.

76

Table 11
Internal Consistency, Means, and Range of AMAS with Arab Americans
AMAS-Scales
Englisha Mean (sd)b
Range Arabic
Mean (sd)
Total Scale
.82
67.15 (9.17) 47-84 .85
61.7 (10.58)
Language Competence .84
31.94 (4.33) 22-36 .89
24.98 (6.57)
Cultural Competence
.79
17.78 (3.84) 7-24 .77
14.75 (4.26)
Cultural Identity
.77
18.05 (4.17) 6-24 .93
21.84 (3.42)
a
Cronbach’s Alpha used to measure internal consistency
b
Mean scores reflect likert-totals (1-low, 4-high) for each scale and subscale

Range
42-83
9-36
7-23
7-24

Relation between acculturation and sociodemographic variables. Some AMAS
acculturation scales were significantly associated with proxy acculturation and
sociodemographic variables. The acculturation levels toward the US or Arab culture were not
related to what language individuals preferred using or where they were born. The longer
individuals stayed in the US (i.e., LOS) the lower they rated their total acculturation levels
toward Arab culture [Total Arab AMAS (78) = -.27, p = .02; Arab Cultural Competence r (78) =
-.24, p = .03]. However, the longer they stayed in the US did not positively relate to increased
ratings of US cultural competence, identity, or English levels. Participants who came from higher
SES background rated higher acculturation levels toward the US [US-AMAS r (78) = .36, p <
.001; US Identity[r (78) = .31, p < .001] and rated having higher Arab Identity [r (78) = .28, p <
.001]. Individuals from higher SES were also more likely to have lived in the US longer [LOS r
(78) =.24, p = .02], and had more educated mothers [r (78) = .34, p < .001]. Those from families
with higher incomes were more likely to have lived longer in the US and rated having higher US
acculturation levels toward the US [US-AMAS [r (78) = .28, p =.01]; US Identity r (78) = .25, p
= .02; LOS [r (78) = .25, p = .02]. The higher the mother education level the higher the ratings of
US cultural competence [US Cultural Competence r (78) = .23, p = .02] but had lower ratings of
Arab cultural competence [Arabic Competence r (78) = -.22, p = .05]. Higher mother education
was also positively associated with longer stays in the US [LOS r (78) =.31, p < .001].

77

Participants who had more educated fathers were more likely to rate having higher acculturation
levels toward the US [US-AMAS r (78) = .24, p = .02].
Relations between age, country of origin, and main outcome measures. The possible
effect of age and country of origin group were examined. The sample was broken into two age
groups, those 14 years or younger and those 15 years and older. Results from an independent
sample t-test showed no significant association between age group with IQ, academic, or
acculturation factors. The sample was also divided into two groups by country of origin. Group
one was composed of all Yemeni American males and sample two was all other Arab American
male. Results from an independent sample t-test showed no significant association between the
two ethnic groups with respect to IQ, academic, or acculturation outcomes. As a result, age and
ethnic background were not controlled for in subsequent analyses.
Academic outcomes. Academic subtests were first examined to identify the sample mean
academic strength and weakness. Next, each academic sample mean was compared to national
normative data. Academic outcomes were examined in how they relate to intelligence and
acculturation outcomes.
The participants performed significantly higher on Numerical Operations (m = 94.10, sd
= 13.22) than on Word Reading (m = 89.16, sd = 13.20), [t (2, 78) = 2.34, p = .01, Cohen’s d
=.37]. The sample performed significantly lower than the normative mean for Word Reading and
Numerical Operations (Word Reading m = 89.16, sd = 13.20; Cohen’s d = -1.07; Numerical
Operations m = 94.10, sd = 13.22; Cohen’s d = -.57). Both academic subtests were significantly
correlated with FSIQ-4 [Word Reading: r (78) = .51, p < .001; Numerical Operations: r (78) =
.60, p < .001]. The V-IQ was strongly associated with Word Reading [r (78) = .56, p < .001],
whereas LR-IQ was strongly associated with Numerical Operations [r (78) = .57, p < .001].

78

Word Reading performance results were negatively associated with total Arabic Competence [r
(78) = -.32, p < .001]. Total Arab Cultural Competence [r (78) = .32, p < .001] was positively
related with English Competence [r (78) = .32, p < .001]. Performance on Numerical Operations
was positively correlated with US-AMAS [r (78) = .28, p < .001] (see Tables 6, 7, and 8).
WASI-II performances. The sample IQ scores were described in comparison to
normative means and designated qualitative descriptors. The sample mean FSIQ-4 was
approximately 11 standardized points below the national normative mean (FSIQ-4 m = 89.29, sd
= 11.20, Cohen’s d = -.92). The sample mean V-IQ was approximately 12 standardized points
lower than the normative mean (V-IQ m = 88.51, sd =11.03, Cohen’s d= -.87), and the sample
mean LR-IQ was approximately 8 standardized points lower than the normative sample (LR-IQ
m = 92.25, sd = 13.33, Cohen’s d= -.55). The highest performance was on Matrix Reasoning,
followed by Vocabulary, Block Design, and Similarities (see Table 12). The lowest performance
on Similarities was unexpected. Performances on the WASI-II are classified based on where a
score falls in a normal distribution. Approximately 6 percent (n = 5) of participants performed
within the Above Average range; 42.4 percent (n = 33) performed within the Average range;
30.1 percent (n = 24) performed within the Below Average range; 18.8 percent (n = 5) performed
within the Borderline-Impaired range; and 2.5 percent (n = 2) performed within the Impaired
range (see Table 13).

79

Table 12
Difference between Obtained and Normative Means
Subtest
Sample T-Score
Normative
Mean (sd)
T-score
Mean (sd)
Vocabulary
44.84 (8.07)
50.8 (8.5)
Similarities
41.11 (6.96)
49.5 (7.8)
Block Design
43.59 (8.43)
50.2 (9.7)
Matrix Reasoning 47.68 (8.79)
49.8 (8.9)
4-FSIQ
89.29 (11.20)
100.2 (12.5)
2-FSIQ
93.30 (11.66)
100.1 (12.1)

Difference
Cohen’s d
-.72
-1.13
-.73
-.24
-.92
-.57

Table 13
Classification of Research Sample According to Normative Standards
FSIQ-4
Classification Current Sample
WASI-II Normative Sample
(%)
130 and above
Very Superior
0
2.1
120-129
Superior
0
7.4
110-119
High Average
6.2
15.8
90-109
Average
42.4
50.9
80-89
Low Average
30.1
16.1
70-79
Borderline
18.8
5.6
69 or below
Extremely
2.5
1.1
Low

Aim One Outcomes
The goal of the first aim is to determine if a significant difference in performance
between V-IQ and LR-IQ exists. It was hypothesized that the participants would perform
significantly higher on LR-IQ than V-IQ. A total of 50 of the participants had higher LR-IQ than
VIQ, 26 had higher V-IQ than LR-IQ, and four performed equally well on V-IQ and LR-IQ. A
paired-sample t-test was conducted to examine within-subject differences between V-IQ and LRIQ. Results show a statistically significant difference [t (79) = -2.70, p = .01, Cohen’s d = .61]. A
new variable, IQ Difference, was created by subtracting V-IQ from LR-IQ to better understand
whether the identified difference was associated with acculturation or sociodemographic factors.

80

Statistically significant results showed that Parent Income [r (78) = .35, p < .001] and
SES [r (78) = .28, = .01] were positively associated with IQ Difference. A small but not
significant negative association between Mother Education and IQ Difference was also observed.
A multiple regression equation was conducted to examine how much variance in IQ Difference
was explained by Parent Income and Mother Education. Socio Economic Status was not
calculated because of its very high association with Parent Income [r (78) = .90, p < .001] and its
overall lower association with IQ Difference when compared with Parent Income.
The overall model was statistically significant [F (2, 78) = 8.32, p < .001, R2 = .18.].
Parent Income explained 14 percent of the variance (b = -3.15, ΔR2Parent Income =.14), and Mother
Education explained an additional 3 percent of the variance (b= 2.04, ΔR2Mother Ed= .03) after
controlling for Parent Income. Results showed that for every $10,000 increase in Parent Income,
the difference between V-IQ and LR-IQ lessened by 3.15 standard scale points. In contrast, for
each three years of education obtained by mothers, the gap between V-IQ and LR-IQ widened by
2.04 standard scale points.
Contrary to expectations, there was notable variability within each index. To investigate
the differences between each subtest, six pair-wise t-tests were conducted with appropriate
Bonferroni corrections (p-value = .008). The t-test values were converted to Cohen’s d effect
sizes to demonstrate the difference magnitude. Results showed that performance on the
Vocabulary subtest was significantly higher than Similarities (Cohen’s d =1.03), significantly
lower than Matrix Reasoning (Cohen’s d = -.61), and it did not significantly differ with the
performance on Block Design. Further, results showed the performance on Similarities was
significantly lower than Matrix Reasoning (Cohen’s d = - 1.15), but it did not significantly differ
from Block Design. Results showed that Block Design was significantly lower than Matrix

81

Reasoning (Cohen’s d =-.58, see Table 14). The performance difference within the V-IQ index
show that the sample performed higher on concrete word knowledge retrieval task than on verbal
categorical reasoning and concept formation task. Within the LR-IQ index, the sample performed
higher on untimed fluid reasoning task than on a timed visual-spatial construction task.
In summary, results from Aim 1 show that the sample performed higher on LR-IQ than
on V-IQ overall, but there was equally discrepant performances within each index as there was
between the indices. In addition, mother education and parent income were related to the
difference between V-IQ and LR-IQ. Specifically, higher mother education levels were
associated with higher V-IQ (r (78) =.39, p < .001), and higher parent income was associated
with higher LR-IQ (r (78) =.27, p =. 03; Table 9). Below is an exploration of how acculturation
variables relate to V-IQ and LR-IQ when controlling for these relevant sociodemographic
variables.
Table 14
Cohen’s d Values from Pair-Wise T-test Comparison of WASI-II Subtests
1
2
3
1. Vocabulary
1.03
-.61
2. Similarities
-1.60
3. Matrix Reasoning
4.Block Design

4
.27
-.58
1.15

Aim Two Outcomes
The possible association between acculturation and WASI-II performance was evaluated
by considering V-IQ and LR-IQ separately. Acculturation was conceptualized using proxy and
systematic factors. Proxy variables were first considered, followed by systematic bidimensional
variables. It was hypothesized that V-IQ and LR-IQ performances would be related to proxy and
systematic acculturation variables. Further, it was hypothesized that systematic measures of

82

acculturation would explain additional variance of the dependent measures beyond that
explained by proxy variables. A greater association was anticipated between the acculturation
levels toward the US with V-IQ than LR-IQ than acculturation toward heritage culture.
Explorations, using bivariate correlations and t-tests, of how demographic, proxy, and systematic
acculturation variables relate to intelligence tests performances were conducted before testing the
outlined hypotheses.
Results from correlation and t-test analyses between proxy acculturation variables, such
as length of stay (LOS), place of birth (POB), and language preference (LP), and intelligence
outcomes (e.g., V-IQ and LR-IQ) were not significant. A significant association between Mother
Education and V-IQ was observed[r (78) = .39, p < .001]. Also, a significant association between
Parent Income and LR-IQ was identified [r (78) = .27, p = .02]. As a result, proxy variables were
eliminated from further regression analyses, but Mother Education was controlled for when
evaluating V-IQ and Parent Income was controlled for when evaluating LR-IQ. The systematic
measures of acculturation variables were centered, and an interaction term was developed from
the product of centered US-AMAS and centered Arab-AMAS. The centered interaction term was
negatively associated with V-IQ [r (78) = -.23, p = .04] but not with LR-IQ.
A hierarchical regression (HR) equation was conducted to test the association between
bidimensional acculturation and V-IQ. A centered Mother Education variable was entered first as
a control, followed by the centered US-AMAS, centered Arab-AMAS, and then the centered
interaction acculturation product term. Variance Inflation Factors (VIF) was less than 10,
suggesting no multicollinearity (Bowerman & O’Connell, 1990). The overall model was
significant and explained approximately 21% of the variance [F (4, 76) = 4.84, p =.002, R2 =
.21]. However, the contributions of Arab-AMAS and the interaction were not significant after

83

controlling for Mother Education and US-AMAS. These results showed that the majority of
variance explained was attributed to Mother Education [F (1, 79) = 14.10, p < .001, ΔR2=.15,
βMotherEd = .39] followed by US-AMAS [F (2, 78) = 9.46, p < .001, ΔR2= .04, βUS-AMAS = .21].
These results show that for every three years of education obtained by the mothers, the V-IQ
increased by 2.36 standard scale points with all other variables controlled (Table 15).

84

Table 15
Summary of Hierarchical Regression Analysis for Variables Predicting V-IQ
Model 1
Model 2
Model 3

Model 4

Variables

b

b

SE b β

Mother Education 2.90 .77
US-AMAS
Arab-AMAS
Interaction Term

b

SE b β

.39** 2.56 .78
.26 .13

b

SE B β

.35** 2.42 .80
.22* .29 .13
-.08 .11

ΔR2
.15.
.04
.01
Total R2
.15
.20
.20
Note: Interaction term is product of centered US-AMAS and Arab-AMAS
*p < .05 **p <.001.

85

SE b β

.33** 2.36 .82
.24* .27 .14
-.08
-.06 .13
-.01 .01
.00
.21

.32**
.22*
-.06
-.06

A second regression was conducted to examine the association between acculturation and
LR-IQ. Correlation and t-test results showed that proxy acculturation variables and Arab-AMAS
were not significantly associated with LR-IQ; only Parent Income [r (78) = .27, p = .01], SES [r
(78) = .25, p = .02], and US-AMAS [r (78) = .22, p = .06] were significantly associated with LRIQ. However, SES was no longer significant after controlling for Parent Income. Thus, the
association of US-AMAS with LR-IQ was explored when controlling for Parent Income. Results
indicated a significant overall model [F (2, 78) = 4.06, p =.02, R2 = .10, β = .27], which
explained 10% of the LR-IQ variance. However, the contribution of US-AMAS was not
significant when Parent Income was controlled for (US-AMAS ΔR2 = .02). Parent Income
explained approximately 8 of the total 10 percent [F (2, 78) = 6.03, p = .01, R2 = .08, β =
.23].These results show that an increase of 10,000 dollars in Parent Income leads to 2.04
standard scale points increase in LR-IQ (Table 16).
In summary, the associations of acculturation and sociodemographic factors were
separately related to V-IQ and LR-IQ. Acculturation toward the US mainstream culture was
significantly associated with performance on V-IQ but a higher association was found between
V-IQ and mother education levels. Only Parent Income was significantly related to performance
on LR-IQ.

86

Table 16
Summary of Hierarchical Regression Analysis for Variables Predicting LR-IQ
Step 1
Step 2
Variable
b
SE b β
Parent Income 2.41 .96
.27*
US-AMAS
TotalR2
.08
2
ΔR
.08
*p < .05. **p < .001.

b
SE b β
2.04 1.00 .23*
.22 .16
.15
.10
.02*

Aim Three Outcomes
The goal of the third aim is identify the predictive association between FSIQ-4 and
academic skills (i.e., Word Reading and Numerical Operations) and to explore if acculturation
levels toward US mainstream culture (i.e., US-AMAS) moderated that predictive association.
Two hierarchical regression equations were conducted to examine the predictive association
between FSIQ-4 with Word Reading and Numerical Operations, and whether US-AMAS
moderated such associations. It was hypothesized that FSIQ-4 would have a moderate to strong
association with Word Reading and Numerical Operations. Higher US-AMAS was hypothesized
to increase the association between FSIQ-4 and academic outcomes and lower US-AMAS to
lower that association. All independent predictors were centered, including the interaction term
(FSIQ-4 X US-AMAS). The interaction term was positively related to Word Reading [r (78) =
.23, p = .04]. Total US-AMAS was not significantly related to Word Reading [r (78) = .19, p =
.10]. The order of independent predictors was the same for both regression models: centered
FSIQ-4 was entered first, centered US-AMAS was entered second, and the centered interaction
term was entered third.
Results from the first model revealed a significant overall model [F (3, 76) = 8.99, p <
.001, β = .51, Model R2 = .26], with FSIQ-4 emerging as a significant independent predictor [F

87

(1, 79) = 26.91, p < .001, β = .51, ΔR2 = .26], accounting for 25.7 percent of the variance in
Word Reading. Additionally, US-AMAS did not moderate the relationship between FSIQ-4 and
Word Reading (See Table 17).These results show that each standard scale point increase in
FSIQ-4 leads to .60 standard scale points increase in Word Reading.
A second regression was conducted to examine FSIQ-4 association with Numerical
Operations and whether US-AMAS moderated that association. A significant association was
observed between the interaction term and Numerical Operations [r (78) = .33, p < .001]. Results
from the second regression revealed a significant overall model [F (3, 76) = 15.70, p < .001, β =
.51, Model R2 = .38]. The FSIQ-4 emerging as a significant independent predictor [F (1, 79) =
44.39, p < .001, β = .62, ΔR2= .36] accounting for 36 percent of the variance in Numerical
Operation. Additionally, US-AMAS did not moderate the association between FSIQ-4 and
Numerical Operation (See Table 18).These results show that each standard scale point increase
in FSIQ-4 accounts for .71 standard scale point increase in Numerical Operations. In summary,
the FSIQ-4 was moderately to strongly predictive of Word Reading and Numerical Operations
scores, and acculturation did not moderate the association.

Table 17
Summary of Hierarchical Regression Analysis for Variables Predicting Word Reading
Step 1
Step 2
Step 3
Variable

b

SE b

β

FSIQ-4
.60
.12
.51**
US-AMAS
Interaction
Total R2
.26
2
ΔR
.26
Note: *p <.05. **p <.001, (N=80)

b

SE b

β

b

SE b

Β

.58
.07

12
.15

.49**
.05
.

.53
-.54
.01
.26
.00

.15
1.07
.01

.45**
-.38
.44

.26
.00

88

Table 18
Summary of Hierarchical Regression Analysis for Variables Predicting Numerical Operation
Step 1
Step 2
Step3
Variable

b

SE b

β

FSIQ-4
.71
.11
.60**
US-AMAS
Interaction
.36**
Total R2
ΔR2
.36**
Note: *p<.05. **p <.001, (N = 80)

b

SE b

β

b

SE b

Β

.67
.17

11
.14

.57**
.12
.

.60
-.71
.01
.38**
.01

.14
.98
.01

.51**
-.49
.64

.38**
.01

Exploratory Analyses
Subtests and academic outcomes. Further correlational analyses of how other WASI-II
scales and subtests relate to academic outcomes in comparison to normative associations were
conducted. In the current sample, Word Reading was most related to the Vocabulary subtests [r
(79) = .58, p < .001] and least related to Block Design [r (79) = .26, p =.02]. The normative data
for the WASI-II showed that Word Reading was most related to FSIQ-4 and least related to
Block Design. In the current sample, Numerical Operations was most related to FSIQ-4 [r (79) =
.60, p < .001] and least related to Similarities (See Table 16). The normative data for the WASIII showed that Numerical Operation was most related to FSIQ-4 and least related to Block
Design (See Table 19).
Table 19
Study and Normative Sample Correlations between the WASI-II and Wechsler’s Academic Outcomes
Wechsler Academic Achievement Tests
WASI-II Subtests/ Composites
Word Reading
Numerical Operations
Norms Current Sample
Norms
Current Sample
Block Design
.46
.26
.50
.55
Matrix Reasoning
.53
.36
.60
.54
Vocabulary
.64
.56
.53
.41
Similarities
.62
.41
.53
.35
V-IQ
.68
.56
.56
.43
LR-IQ
.59
.32
.65
.57
FSIQ-4
.70
.51
.68
.60
FSIQ-2
.67
.50
.65
.57

89

US-AMAS scales relation to WASI-II subtest scores. The results showed that the
associations between sociodemographic and acculturation variables with WASI-II performances
were primarily related to the US-AMAS scale, mother education, and parent income.
Exploratory regressions were conducted to evaluate which US-AMAS subscale explained the
most variance in each of the four WASI-II subtests after controlling for Mother Education and
Parent Income levels.
Four hierarchical regressions were conducted to examine the contribution of each AMAS
scale for predicting each subtest of the WASI-II after controlling for Parent Income and Mother
Education. Each model included Mother Education and Parent Income as controls in the first
block, followed by the three US-AMAS scales: English Competence, US Cultural Competence,
and US Identity scales, as independent predictors of performance on Vocabulary, Similarities,
Matrix Reasoning, and Block Design. If the US-AMAS scales explained significant amount of
variance beyond what was explained by sociodemographic variables, then multiple regression
equations were conducted to investigate which specific US-AMAS scale made significant
contributions.
First model: vocabulary subtest. The overall model of the first regression was significant
[F (5, 74) = 7.11, p < .001, Model R2 = .32]. The sociodemographic variables (Mother Education
and Parent Income) accounted for 10.4 percent of the variance in Vocabulary (ΔR2= 10.4, p =
0.14). With sociodemographic variables controlled, the US-AMAS scales accounted for an
additional 22 percent of the variance (ΔR2 = 22.0, p < .001). The English Language Competence
scale was significant with other variables controlled indicating that it, alone, could uniquely
explain Vocabulary scores. The English Language Competence accounted for 17.3 percent of
Vocabulary variance in the sample (ΔR2 = 17.3, p < .001).

90

Second model: similarities subtest. The overall model of the second regression model
was also significant [F (5, 74) = 2.70, p = .03, Model R2 = .15]. However, contributions of the
US-AMAS scales were not significant after controlling for sociodemographic variables (USAMAS scales ΔR2= .02, p = .22).
Third model: matrix reasoning subtest. The overall regression model was not significant.
There was no significant association between sociodemographic or US-AMAS subscales with
Matrix Reasoning.
Fourth model: block design subtest. The overall model of the fourth regression was
significant [F (5, 74) = 3.48, p = .002, Model R2 = .19]. Sociodemographic variables accounted
for 13 percent of the explained Block Design variance. After controlling for sociodemographic
variables, the US-AMAS scales accounted for an additional 6.4 percent of the variance (See
Table 17). The US Culture Identity and English Competence scales each explained
approximately three percent of Block Design variance.
Overall, the four models related differently to sociodemographic and acculturation
variables. Vocabulary was significantly associated with Mother Education [t (5, 74) = 3.10, p <.
001, β = .31], Parent Income [t (5, 74) = -2.10, p <. 001, β =.-22], and English Competence [t (5,
74) = 4.14, p <. 001, β = .44]. Similarities was not significantly associated with any of the USAMAS scales. A significant association between Similarities and Mother Education was
observed [t (5, 74) = 3.27, p < .001, β = .36]. No significant association was found between
sociodemographic and US-AMAS with Matrix Reasoning. Block Design was significantly
associated with Mother Education [t (5, 74) = 2.54, p =. 02, β = .28].

91

Table 20
Multiple Regression Results for US-AMAS and WASI-II Subscales
Similarities Vocabulary
Matrix Reasoning
2
2
ΔR
β
ΔR
β
ΔR2
Β
Sociodemographic
.13*
.10*
.03
Mother Education
.36**
.31**
.10
Parent Income
-.09
-.22*
.12
English Competence .02
.14
.17** .44** .01
.10
Cultural Competence .00
-.02
.03
-.19
.01
-.13
Cultural Identity
.00
.05
.03
.19
.03
.17
2
Total R
.15*
.32**
.06
Note: * p < .05, ** p <.001

Block Design
ΔR2 Β
13*
.28*
.10
.03 .18
.01 -.11
.03 .18
.19*

Mother education and WASI-II performances. Mother Education was significantly
associated with most of the WASI-II subtests. An extreme group analysis was constructed as a
function of Mother Education levels. One group included participants whose mothers earned at
least a high school diploma (n = 24), and a second group included participants whose mothers
earned less than a high school diploma (n = 56). The higher education group had higher scores
on all WASI-II and academic subtests. Results from an independent sample t-test showed
significant group differences across FSIQ-4, V-IQ, and Similarities (see Figure 1).

92

Figure 1.
Cogntive Results as a Function of High and Low Mother Education
105
100
95
90
85
80
75
High Education (N = 24)
Low Education (N = 56)

FSIQ-4

V-IQ

Vocabulary

Similarities

LR-IQ

93.17
87.63

93.54
86.36

95.3
92.6

91.4
85.2

95.04
91.05

Matrix
Reasoning
99.3
95.3

Block Design
94.1
90.1

Word
Reading
92.67
87.67

Math
94.13
94.1

Figure 1. Individuals whose mothers had at least a high school diploma outperformed those individuals whose mothers had
education levels below high school diploma on all cognitive outcomes.

93

CHAPTER V: DISCUSSION
Purpose of the Study
Psychologists are sometimes asked to assess individuals from a wide range of cultural,
linguistic, and ethnic backgrounds using standardized norm-referenced intelligence tests. This
can be challenging when evaluating the intellectual ability of young first or second generation
immigrants who are in the process of acculturative and developmental transformations. The
assessment of intelligence is the most critical component of a comprehensive neuropsychological
or psychological evaluation, as it provides a valuable marker for understanding the learning
potential of students, the integrity of neurocognitive functioning, and it aids in differential
psychological and neurodevelopmental diagnoses (Sattler, 2008; Lezak et al., 2012). The
difficulty emerges when administering standardized tests to individuals who are not well
represented by the normative samples of the most commonly used IQ tests. There is a growing
but very limited body of literature that highlights the profound influence of cultural and language
factors on intelligence test performances (Flanagan & Ortiz, 2007; van de Vijver & Phalet,
2004). Practitioners often resort to empirically unsupported practices when evaluating immigrant
male because they have received minimal training in multicultural assessment and because of the
scarcity of research evaluating the clinical utility of intelligence tests among immigrant
populations (Rhodes et al., 2005). Consequently, immigrant students are twice as likely to be
identified as having learning, intellectual, or language impairments (Sullivan, 2011).
Available multicultural assessment research with immigrants has been done with larger
ethnic groups such as Hispanic and Asian Americans, but very little research is available with
smaller ethnic minority groups such as Arab Americans. The available literature is also limited
because there has been no direct assessment of how acculturation and language levels relate to

94

verbal and language-reduced tests. No study has evaluated the association between proxy and
systematic measures of acculturation with verbal and language-reduced assessment among
typically developing immigrant adolescents. There are also no studies to date that have examined
the clinical utility of intelligence tests among the Arab American population.
The current study was designed to expand the research on the influence of acculturation
and language levels on verbal and language-reduced IQs. Specifically, the study examined the
association between proxy and systematic measures of acculturation with IQ assessment among
an adolescent population while controlling for important sociodemographic variables. A total of
80 typically developing Arab American male adolescents participated in the study. The majority
of participants were born in the US (n = 45) and were of Yemeni backgrounds (n = 63). Each
participant was evaluated using the WASI-II, an abbreviated test of intelligence, and the
Numerical Operations and Word Reading subtests of the WIAT-IV, a comprehensive test of
achievement. Participants also completed the AMAS acculturation rating scale and a
demographic form. At least one guardian provided demographic information for each male.
Discussion of Findings
The study had three primary aims. The first aim was to investigate if performance
differences existed between verbal and language reduced intelligence tests that are clinically
meaningful. The second aim was to evaluate how specific acculturation factors relate to verbal
and language reduced tests. The third aim was to examine if acculturation levels toward the US
moderated the predictive association between the WASI-II and academic achievement.
Aim one. With regards to aim one, it was hypothesized that the study sample would
perform higher on language-reduced tests than on verbal tests. Flanagan and Ortiz (2007) argue
that immigrant students, who are considered moderately different from the normative sample, are

95

expected on average to perform 20-30 standard scale points below the mean on Vocabulary and
Similarities, and approximately 7-10 standard scale points below the mean on Block Design and
5-7 standardized points Matrix Reasoning (see Figure 2).
Figure 2.
Sample's Subtest Performances in Comparison to Normative Outcomes and C-LIM
110
105
100
95
90
85
80
75
70
65
60

100

100

100

100
96.52

92.26

94

90.84 91.5

86.66
75

Vocabulary

75

Similarities
Matrix Reasoning
Normative
Sample
C-LIM

Block Design

Figure 2. The sample’s performances on verbal and language reduced subtests varies
from the normative and the C-LIM projections.
A moderate to large mean difference was observed between LR-IQ and V-IQ. This
finding supported the stated hypothesis, and it was also largely consistent with available studies
that have used similar Wechsler’s IQ tests (Dynda, 2008; Reed, 2000; Vardarosa, 2007; Dana,
2005; Vukovich & Figueroa, 1972, as cited in Flanagan & Harrison, 2012). The magnitude of the
difference was also clinically meaningful However, the study participants’ performance varied
notably within each index; the sample’s highest performance was on Matrix Reasoning, followed
by Vocabulary, Block Design, and Similarities. This pattern of performances and the magnitude
of variation from the normative mean within each index were unexpected. These results were

96

also inconsistent with projected performances as outlined by the C-LIM framework and with
results from prior studies with other immigrant groups (Flanagan & Ortiz, 2007; Cummins,
1981; Reed, 2000; Sotelo-Dynega, 2007; Donoso, 2010; Coffey et al., 2005).
The C-LIM framework was used as a guide to understand WASI-II performances in the
current sample. Overall, there was considerable variability among subtests that were not greatly
consistent with expected C-LIM patterns. The divergence of performances between Vocabulary
and Similarities likely reflected the culturally and linguistically mediated cognitive demands
required by each subtest, which is not captured by the three-by-three C-LIM classification
framework. Results from the current study challenge the C-LIM classification assumption that
Vocabulary and Similarities reflect equal language and cultural demands. Vocabulary is a rote
verbal test that depends on verbal retrieval from an acquired fund of vocabulary knowledge,
whereas Similarities not only requires vocabulary knowledge, it also requires verbally mediated
abstract categorical reasoning (Sattler, 2008; Lezak et al., 2012). Fluent English-speaking
immigrant students have more exposure to English vocabulary at school, in the community, with
peers, and through the media than to the demands and style of reasoning required by Similarities.
It is possible that the cultural loading within Similarities is higher than for Vocabulary. For
example, culture-based differences in level of abstraction and modes of categorization were
documented by Luria’s ethnographic studies in Uzbekistan (Luria, 1979) and by James Flynn,
who attributed part of the increase in IQ over the last seven decades (i.e., Flynn Effect) to
increased abstraction reasoning skills, such as taking on the hypothetical as if it were real and an
expansion of educational and cultural complexity (Flynn, 2012). Other studies have also
identified that the greater cognitive complexity a test has, the greater the cultural loading it
carries (Helms-Lorenz, van de Vijver, & Portinga, 2003). The significant difference between the

97

two Vocabulary and Similarities deemphasizes the value of examining the differences between
V-IQ and LR-IQ and highlight the importance of considering within index variation.
The variation of the two subtests that compose the LR-IQ was more consistent with CLIM expectations than V-IQ outcomes discussed above. According to C-LIM framework’s
projections for the current sample, Block Design standard scores were expected to be 90 to 93
and Matrix Reasoning standard scores were expected to be 93 to 95. Performance on Block
Design was anticipated to be lower because it is classified as having moderate level language
requirements and low cultural demand; whereas Matrix Reasoning was classified as having low
cultural and language demands. The current sample means were approximately 91 on Block
Design and 96 on Matrix Reasoning (see Figure 2).
The small variation observed within the LR-IQ index may be partly attributed to the
characteristics of the test and how it was administered. The Block Design subtest in the WASI-II
has a shorter discontinuation rule when compared to other Wechsler full-battery Block Design.
As a result of a shorter discontinuation rule, it is feasible that performance efficiency becomes
weighted more (Lezak et al., 2012). Although the target cognitive process assessed is visualspatial processing, the underperformance by the current sample could be attributed to slower
performances because the test is timed. Ardilla (2005) argued that speed is a value that varies
cross-culturally. Some cultures, like the US, value speed and fast production whereas in other
cultures, slower and careful processing may be primarily valued. Another possible explanation is
that the task itself is affected by cultural peculiarities such as exposure to visual perceptual
stimuli, block manipulation experiences, and negative cultural attitude toward speeded tasks
(Liblich & Kugelmass, 1981; Kugelmass, 1974). A study with elementary school Arab students
in Israel also documented significantly lower performance on Block Design than Matrix

98

Reasoning (Liblich & Kugelmass, 1981). The authors suspected that the dominance of oral
traditions and less emphasis on imagery in Arabic and Islamic culture was a possible explanation
(Liblich & Kugelmass, 1981). Culture emphasis on imagery among the indigenous Maori young
men from rural and urban settings in New Zealand was believed to be the underlying factor that
explained the nearly 15 standard scale points above the US normative mean performance on
Block Design (Ogden & MacFarlane-Nathan, 1997). Liblich and Kugelmass (1981) also
identified overall higher verbal than visual test performance among Arab children in Israel and
noted significant deficits with processing speed. Other possible explanations could be that Arab
American male may not be as familiar with block manipulation tasks as those children included
in the normative sample. This assumption is based on the overall low SES; consistent
associations between low SES, low education attainment, and lower overall IQ performances
have been observed across a number of studies (Flannigan & Harrison, 2012; Heaton, Taylor, &
Manly, 2003; Pollitt, Gorman, Engle, Martorell, & Riveria, 1993). This explanation is consistent
with the small to moderate association between the mother’s education level and performance on
Block Design. Mother’s education was also positively associated with longer stay in the US and
higher income (see Figure 1).
In summary, results from aim one showed a measurable and statistically significant
difference between verbal and language reduced IQ. However, the difference was less
meaningful when considering the significant variation within each index and across the four
WASI-II subtests. Overall, results did not substantiate the pattern of performance predicted by CLIM framework within the V-IQ. The C-LIM patterns were more consistent with performances
within the LR-IQ index. The empirical support for the clinical utility of the C-LIM has been
variable. Support for its use has come mainly from unpublished dissertation studies. However,

99

two available published studies have underscored the lack of utility for the C-LIM (Styck &
Watkins, 2013; Kranzler et al., 2010). For instance, Kranzler and colleagues (2010) found
limited utility of the C-LIM when using Woodcock Johnson Tests of Cognitive Abilities, Third
Edition (WJ-III) among a multi-ethnic sample of referred elementary school students. Kranzler
and colleagues (2010) explained that when examining within-individual variation, the C-LIM
patterns of performances were not substantiated, but the overall sample means were consistent
with expected C-LIM patterns. More recently, Styck, and Watkins (2013) showed that the CLIM approach failed to discriminate between ELL from non-ELL children beyond chance levels
when using the WISC-IV. All prior studies have used clinically referred-samples, which further
complicates examining the association with acculturation and language development
independent of possible neurocognitive impairments (Nieves-Brull, 2006; Sotelo-Dynega, 2007;
Verdorasa, 2007; Dynda, 2008; Kranzler et al., 2010; Styck & Watkins, 2013). The current study
used an underrepresented non-referred community-based sample and examined the association of
language and acculturation for each subtest. The applicability of C-LIM classifications across
different ethnic groups and different IQ tests likely varies which undermines the clinical utility of
the C-LIM approach for systematically understanding the interaction between culture, language,
and test factors across all ethnic groups. A closer examination of how different acculturation and
language factors relate to performances on V-IQ and LR-IQ was conducted in aim two of the
study.
Aim two. To meet the second aim, sociodemographic, proxy acculturation, and
systematic acculturation factors were examined in relation to performances on V-IQ and LR-IQ
indices. Sociodemographic variables included Parent Income, Mother Education, Father
Education, and SES. Proxy acculturation variables included the Length of Stay (LOS), Place of

100

Birth (POB), and Language Preference (LP). The POB and LP variables were both dichotomous
(i.e., Born in or outside the US; Prefer English or other language). Systematic acculturation was
measured by the AMAS, a bidimensional multi-domain acculturation scale. Acculturation
variables have been noted to be important for understanding performances on IQ tests (van de
Vijver & Phalet, 2004).
Contrary to expectations, there was no measurable association between proxy
acculturation and systematic acculturation measures. Total US-AMAS was not significantly
associated with total length of stay, as was identified in the validation study of the AMAS (Zea et
al., 2003). There was a small negative relation between total length of stay with Arab
acculturation levels. The lack of strong associations between proxy and systematic acculturation
measures are likely reflecting the lack of variation within the proxy acculturation variables.
Nearly the entire sample preferred using English over Arabic. The majority of the sample was
born in the US, and those born outside the US, immigrated at a very young age. The restricted
variation among proxy acculturation also resulted in no significant association with V-IQ and
LR-IQ. As a result, these proxy acculturation measures were not meaningful for understanding
variation in IQ performances with this particular sample. Acculturation was also measured using
systematic rating scales.
The evaluation of acculturation using the AMAS scale showed significant variations
across US and Arab cultural domains. Overall, there was a positive association between
American and Arab Acculturation scales. The results showed that the majority of the sample
endorsed high ratings on both Arab and American acculturation levels, reflecting bicultural
acculturation orientation for the majority of the sample. However, acculturation orientation
within the sample varied as a function of the specific acculturation domains assessed. For

101

instance, the majority of the sample rated having higher English and American Cultural
Competencies than Arabic and Arab Cultural Competencies, but 78 of 80 rated having higher
Arab than American Identity, although there was a small, positive association between Arab and
American Identity ratings. This is not surprising given that the sample was composed of firstand second-generation immigrants with an average LOS of approximately 12 years. Studies with
other immigrant groups have shown that cultural identity is often viewed as a deeper form of
assimilation and may be the last cultural domain to be influenced by the host culture (Phinney,
2003; Berry et al., 2006; Zea et al., 2003). Overall, these domain specific variations are
consistent with models that view acculturation as an interactional, hierarchical, and dynamic
process that can vary as a function of specific domains (Rudmin, 2009; Arends-Tóth & van de
Vijver, 2003, 2006). This implies that bidimensional acculturation scales that do not separately
evaluate multiple domains of acculturation run the risk of oversimplification and a loss of insight
into how acculturation orientations vary not only among and within groups, but among and
within individuals. The AMAS scale was valuable for capturing varying levels of variations
across multiple domains. Such variation allowed for a more nuanced evaluation of how cultural
and language factors related to WASI-II performances.
Cultural and language levels were assumed to systematically relate to WASI-II
performance as expected by C-LIM predictions (Flanagan & Ortiz, 2007). It was hypothesized
that systematic measures of acculturation would contribute additional explanatory power beyond
proxy acculturation or sociodemographic measures. Assimilation toward the US was anticipated
to be positively associated with intelligence test outcomes, and the maintenance of Arab culture
was anticipated to be negatively associated with intelligence test outcomes. Results showed no
association between proxy acculturation variables with performances across verbal and

102

language-reduced IQ tests. However, important associations emerged between sociodemographic
variables and WASI-II outcomes.
It is well established in the literature that SES is positively associated with IQ
performance (Lezak et al., 2012; Sattler, 2008; Heaton, Taylor, & Manly, 2003). Total family
income and mother education have also been positively linked with overall IQ performances
(Lezak et al., 2012; Heaton, Taylor, & Manly, 2003). In the current study, there was a small to
moderate level association between mother education levels with V-IQ and a small association
between total parent incomes with LR-IQ. The mother’s education level was also mildly
associated with total length of stay and parent income. In this sample, it appears as if those
families with a longer length of stay were also more likely to have higher income and greater
mother education attainment. It can be argued that such higher sociodemographic factors may
result in a more enriched environment that better approximates that of the normative sample.
Approximately 95 percent of the WASI-II normative sample had a parent with at least a high
school diploma (Wechsler, 2011). As a result, IQ performances were closer to the normative
means (see Figure 2). The evaluation of systematic measures of acculturation was conducted
after controlling for these important associations.
Examination of systematic acculturation variables showed some correlation across US
and Arab acculturation orientations. Total US acculturation was positively associated with V-IQ
and LR-IQ; the strength of that association was small. The interaction between American and
Arab Acculturation orientations was significantly associated with V-IQ, but not with LR-IQ.
Overall, maintenance of Arab culture was not significantly related to V-IQ or LR-IQ outcomes.
There is one previous study that found higher scores with higher assimilation and lower scores
with higher heritage endorsement (van de Vijver, Helms-Lorenz, & Feltzer, 1999). Further

103

exploration was conducted to measure how specific host and heritage scales related to WASI-II
subtests.
The association between domain specific host and heritage acculturation levels with
specific WASI-II subtest revealed important associations. For instance, a small negative
association was observed with Vocabulary and Arab cultural competence and a positive
moderate association with Vocabulary and English competence. Also, results showed that
English competence and mother education were equally important for predicting Vocabulary. No
acculturation variable was associated with Similarities. Thus, the overall positive association
between US acculturation and V-IQ was better explained by specific associations between
mother education and English levels with Vocabulary. The mother’s education level was equally
important for predicting Similarities performance. There were also specific association identified
between sociodemographic and acculturation factors with LR-IQ.
The association between acculturation levels and LR-IQ performance was mostly
explained by total parent income levels. A small-to-moderate positive association between total
parent income and Block Design was observed. A small positive association was observed
between levels of US cultural identity and Block Design. However, the association was no longer
significant after controlling for mother education and parent income. Performance on Matrix
Reasoning was not substantially associated with any acculturation or sociodemographic factors.
This suggests that Matrix Reasoning may be the most robust WASI-II subtest against the
influence of sociodemographic and acculturation factors among Arab American male. This is
consistent with other studies that found Matrix Reasoning or other cognitive tests with similar
formats (e.g., Raven’s Matrices) to be the least affected by culture and language factors

104

(Flanagan & Ortiz, 2007; Nell, 2000, Brouwers, Van de Vijver, & Van Hemert, 2009, Rushton &
Jensen, 2010).
In summary, exploration of aim two showed that different components of
sociodemographic and acculturation variables were uniquely associated with specific WASI-II
subtests. Proxy measures of acculturation were not useful for understanding WASI-II
performances. The relation between language and cultural factors was not similarly distributed
across V-IQ and LR-IQ indices. Aim two results showed that Vocabulary subtest is much more
sensitive to English competence levels than Similarities. Whereas performances on Similarities
was mostly predicted by the education level of the mother. These variations were unexpected in
light of the C-LIM predictions. No acculturation factor was significantly associated with LR-IQ
performances after controlling for mother education and parent income levels. Performances on
the LR-IQ were also more consistent with C-LIM predictions and were not associated with any
measured acculturation factors. A better understanding of how acculturation and
sociodemographic factors explained variance in V-IQ and LR-IQ but it remained unclear if these
tests were clinically useful for predicting basic reading and math skills.
Aim three. The third aim was to examine whether acculturation levels toward the US
moderated the predictive association between FSIQ-4 and basic academic skills. It is well
established that IQ is the single best predictor of academic achievement and total academic
attainment (Naglieri & Borenstein, 2003; Lezak et al., 2012). A moderate to strong association is
documented between FSIQ and standardized academic measures (Wechsler, 2011; Lezak et al.,
2012). It was hypothesized that a moderate to strong association would be observed between
FSIQ-4 with basic math and reading outcomes as was identified in the normative sample
(Wechsler, 2011). It was further hypothesized that acculturation would positively moderate both

105

associations. Both hypotheses (i.e., FSIQ-4 with reading and math) were supported; results
showed a moderate level association between FSIQ-4 with basic math and reading outcomes.
The predicative association between FSIQ-4 and math was slightly higher than that between
FSIQ-4 and reading. The levels of US acculturation were examined for possible moderating
effect on the association between FSIQ-4 with basic reading and math skills. Results showed that
acculturation was not a moderating factor for both associations. This is not surprising because
there were limited association identified between US acculturation level and WASI-II
performances. Additional exploratory analyses were conducted to examine whether individual
scales or subtests had higher association with academic outcomes than FSIQ-4. In general,
language-reduced tests have been found to have limited predictive power when compared to
verbal or combined IQs (Giguero, 1989; Lohman et al., 2008). The conducted analyses did not
identify any single scale or subtest that predicted academic outcomes significantly better than
FSIQ-4, which is consistent with findings with other populations (Mayes & Calhoun, 2007;
Sattler, 2008). Results from aim three imply that the use of FSIQ-4 among Arab American male
holds good clinical utility for predicting basic academic skills independent of moderating effects
of acculturation levels.
Integrative Summary
With regards to the measurement of acculturation, proxy acculturation variables were not
associated with WASI-II performances. The multidimensional AMAS outcomes illustrated
important acculturation variation between and within individuals. Single domain acculturation
scale may be limited in capturing between and within individual differences among Arab
American male, as was observed in the current study. The AMAS subscales were mildly
informative for understanding variation between and within V-IQ and LR-IQ. The most

106

important subscale was English Competence scale, which had a small association with only
Vocabulary. There was a significant discrepancy between Vocabulary and Similarities scores
that was likely associated with cognitive processing demands and possible higher cultural
loading of Similarities. No association was identified between Acculturation toward American or
Arab culture with Similarities, Block Design, or Matrix Reasoning after controlling for mother
education and parent income. Although a notable difference emerged between V-IQ and LR-IQ,
such a difference was not systematically associated with language and cultural factors as
anticipated by C-LIM. Instead, the difference was better explained by variations in mother
education and parent income.
Sociodemographic variables proved to be important factors for understanding WASI-II
performances. Specifically, total parent income was associated with Similarities and Block
Design. Mother education was moderately associated with Vocabulary, and it was relevant for
understanding performances on all WAS-II subtests. Individuals whose mother had at least a
high school diploma scored within the average range across all subtests. These results are
consistent with the Environmentalists’ view of IQ development in that the cultural and
sociodemographic milieu are important factors to understanding variation in levels of IQ among
cultural and language minority populations (Nell, 2000; Helms-Lorenz, van de Vijver, &
Portinga, 2003). Despite the influence of some acculturation and sociodemographic factors, these
results showed that the FSIQ-4 had a moderate level predicative association with academic
outcomes, which was not moderated by acculturation levels. The overall predicative utility was
largely comparable with associations between FSIQ-4 and academic outcomes among the
normative sample, especially for math skills. Overall, the FSIQ-4 was the best single predicative
value for basic reading and math outcomes. These results also support the assessment of both

107

verbal and language-reduced domains with Arab American male. This is important information
in light of the challenges neuropsychologist face when attempting to provide a useful and
ethically responsible assessment. Results from this study highlight continued evaluation of how
specific individual and test factors influence overall clinical utility. Although, a much higher
number of study participants performed within the below average and impaired ranges than the
normative distribution. This skew in performance likely reflects the cumulative divergence of
cultural, linguistic, and socidemographic outcomes of the study participants when compared to
the normative sample. Without a randomized expanded sample it is difficult to determine if such
skewed distribution will persists. The US will continue to experience shifts in its demographics
which will place great difficulties on all psychologist working with children. Continued
evaluation of how norm-referenced tests such as the WASI-II will enhance multicultural
assessment competence by developing the empirical foundation for selecting and interpreting
performances. The results from this study provided useful research and clinical implications.
With any empirical study, there are limitations. A summary of notable limitations are provided
below.
Study Implications and Limitations
Clinical implications. Results from this study support recommendations that the
interpretation of specific WASI-II subtests should be completed cautiously. For example, Arab
American male performed the lowest on Similarities, which also had the lowest association with
academic outcomes. In contrast, they performed significantly higher on Vocabulary than
Similarities, which had a moderate level association with basic reading skills. Performance on
Vocabulary reflects a better measure for crystalized knowledge than Similarities. There were no
significant influences of sociodemographic and acculturation factors on Matrix Reasoning.

108

Clinicians assessing Arab Americans can have some confidence that Matrix Reasoning likely
measures the intended target cognitive skills, namely nonverbal and fluid reasoning, without
great cultural or linguistic influence. Clinicians should be cautious when administering and
interpreting the Block Design subtest to this population. Block Design on the WASI-II may be
sensitive to cultural influences regarding speed due to a lower discontinuation limit on this
measure compared to, for example Block Design on other available tests such as the Wechsler
Individual Scale for Children-Fourth Edition (WISC-IV) and Developmental Abilities ScaleSecond Edition (DAS-II). Ardila (2005) argued that speed is a value that varies cross-culturally.
Some cultures, like the US, value speed and fast production whereas in other cultures, slower and
careful processing may be primarily valued. Clinicians will be best served if they use Block
Design scoring and administration that allows for untimed interpretation and with a higher
discontinuation rule to obtain a better assessment of visual-spatial processing skills. Despite the
lower performance on Block Design, it proved to be a moderately good predictor of math skills.
Even though there was variability within and between each WASI-II index, the FSIQ-4
was the single best predictor of reading and math skills. This is consistent with the identified
factor structure of the WASI-II among the normative group as it reflects a latent measure of
wider set of cognitive abilities. Overall, the clinical utility of the WASI-II among English
speaking Arab American population is adequate for anticipating academic outcomes. Important
attention should be given to mother education, parent income, and English language competence
levels when administering and interpreting test scores from this measure. Results from this study
also showed that a much higher number of individuals performed within the below average to
impaired range across some subtests and overall IQ. Interpretation of such outcomes need to be
interpreted in light of the possible cultural, linguistic, and sociodemographic effects. It is unclear

109

if this distribution pattern will persist if a larger randomized sample of Arab American male was
evaluated.
Research implications. Important research implications can be gleaned from the current
study. The purpose of the current study was to contribute to the multicultural assessment
research specific to intellectual assessment. Results from the current study highlighted important
factors to consider in future studies with Arab American and possibly other immigrant groups.
However, there are several factors in the current study that limit the generalizability of the
results. These factors should be examined in detail in future research. One area in which the
current study could be extended is by examining Arab Americans who are not residing in an
ethnic dense urban setting. Important acculturation variations occur as a result of the host
context. For instance, there may be less pressure to learn English and develop US cultural
competencies than those who live without connection to immigrants from similar cultural,
linguistic, or religious background (Sam & Berry, 2006). It is unclear if the acculturation context
differences could also lead to IQ performance differences.
The study can also be extended by including female participants. Due to logistical
complications, female youth were not included in the study. It can be argued that Arab and
Muslim girls and young women are likely to have different acculturation and educational
experiences than males, and as a result, could vary in their IQ and academic scores. It is also
possible that no difference would emerge since there was weak association between IQ and
acculturation factors in the current study.
Future research can also examine possible differences on how Arab Americans perform
when compared to national and local norms. It is unclear if performance differences would have
emerged if they were compared to a control sample comprised of non-immigrant individuals

110

residing in the same context and who have comparable sociodemographic factors. However, it
will be challenging to find an adequate sample control comparison in light of the unique cultural,
linguistic, and sociodemographic attributes of the study sample.
It is also unclear if assessment of vocabulary knowledge is best conducted using an
expressive or receptive format. The current study showed that expressive Vocabulary was a test
that moderately predicted basic reading skills and was mildly associated with English
competence skills and mother education level. It is possible that the use of a receptive vocabulary
task can mitigate expressive language competence factors and provide a better measure for
vocabulary knowledge. Lastly, the clinical utility of the WASI-II and similar tests could be
examined when predicting other forms of academic outcomes, such as standardized national and
state tests or school-based grades.
Study limitations. The present study was attempted to understand the influence of
acculturation on intellectual assessment and the clinical utility of WASI-II with Arab American
males. However, there were significant limitations to this study. First, the current study had a
non-experimental design and used mostly correlations, t-tests, and linear regression analytic
techniques. Participants were composed of non-random convenience sample located in urban and
ethnically dense communities and schools.
There were other notable limitations specific to the study sample and instrument
characteristics. As discussed by Ardila (2005), psychometric cognitive testing is laden with
cultural values and the assessment context is culturally shaped. One particular value is what he
referred to as isolated environment, in that tests are often done in an isolated room with closed
doors for an extended period of time. Such value may be culturally insensitive, particularly given
religious values with the targeted population of the study. It is culturally inappropriate for an

111

Islamic girl or woman to be alone with a man without the presence of a male relative. The
inclusion of female participants would have required controlling for examinee and examiner
gender, religious, and ethnic background. Such control would have introduced pragmatic and
statistical demands that could not be practically met for this dissertation study. As a result of
these complications, the current study included only males. In addition, the examiner was a male
Arab American. It is possible that characteristics of the examiner could influence the assessment
context and performance (Ardila, 2005). Future studies can examine the possibility of concordant
and discordant assessor and assessee dynamics.
Another important limitation is the restricted ranges of SES, immigration, and country of
origin in the study’s sample. The sample was primarily comprised of first and second generation
Yemeni immigrants residing in an urban ethnic enclave dense setting. This limitation is relevant
in two ways: First, such composition inhibits generalization of the current results to other Arab
Americans who are not Yemeni or do not live in a primarily urban ethnic dense context.
However, when evaluating results across Yemeni individuals versus others groups, no significant
difference emerged. Second, restricted sociodemographic range could have obfuscated the effect
of acculturation levels on WASI-II performances due to the lack of variability in the composition
of the sample.
There were also possible limitations of the instruments used. The uses of a full IQ battery
include paper and pencil processing speed tasks that would have allowed exploration of how
acculturation factors may be associated with speeded performance. Cross-cultural differences
with processing speed performances have been documented (Nell, 2000). In Western culture,
speed is valued, but in other cultures, slow and careful processing is valued (Ardila, 2005; Nell,
2000). In Arabic and Islamic cultures, speed may be not be valued as much as in Western

112

culture, and speeded performances carry a negative cultural connotation (Liblich & Kugelmass,
1981). The inclusion of a full Wechsler battery would have allowed better understanding why the
current sample performed much lower than anticipated on Block Design.
Furthermore, assessing English Language competence by an acculturation rating scale
may have been a limiting factor for illuminating the association between language development
and cognitive functioning. A standardized measure to determine English expression and
comprehension competencies could have allowed greater precision in determining the
association between language levels with verbal and language reduced tests. Important
information can be deduced from this study despite these important limitations.
Concluding Remarks
The current study contributed important information for clinicians and researchers despite
significant limitations. Administering verbal and language reduced tests to first and second
generation immigrants proved to be useful for predicting basic math and reading skills. This
study directly examined language and cultural levels for all subtests while using the C-LIM
framework as a guide to understanding language demand and cultural loading patterns for each
WASI-II subtests. The C-LIM classification framework provided very little utility for explaining
the performance variability observed by the study’s sample. The current study presented an
underrepresented ethnic population that possessed unique cultural and linguistic attributes that
are disparate when compared to the cultural and language of mainstream America. Despite the
great cultural and language contrasts, sociodemographic factors were the most salient factors for
understanding WASI-II performances, and acculturation factors, although to a much lesser
extent. Finally, results from the study highlight that psychologist need to be better aware of the
tests they use when they are administered to individuals not reflected by the normative

113

standardization. It is the psychologist’s responsibility to know their test, to know what they can
and cannot do when established reliability and validity of a test is threatened.

114

APPENDICES

115

APPENDIX A: PARENT DEMOGRAPHIC QUESTIONNAIRE
1) Name:______________________________

2) Relation to child?__________

3) What country did you immigrate from?

___________________________

4) When did you arrive in the U.S.?

___________________________

5) What is your marital status? (Check one below)
Single____

Married____

Divorced_____

Separated_____

Other_____

6) What is the COMBINED household yearly income? (Check one below)
$ 5, 000 or less

$ 40,001-50,000

$ 5,001-10,000

$ 50,001-60,000

$ 10,001-20,000

$ 60,001-70,000

$ 20,001-30,000

$ 70,001-100,000

$ 30,001-400,000

More than $ 100,000

7) What is the highest education level achieved by the mother and father?
(Check for each)
Education Level

Mother

Kindergarten up to 3rd grade
4th grade up to 6th grade
7th grade up to 9th grade
10th grade up to high school graduate
One or two years of college
Bachelor’s degree
Master, professional or doctorate degree
8) What is the primary language spoken at home?

116

Father

Mostly Arabic

Mostly English

Both Equal

Other

9) What language do you use to speak with your child?
Mostly Arabic

Mostly English

Both Equal

Other

10) Total number of family members in the household? _______________________
11) What is the occupation of the mother and father
a) Mother’s Occupation___________________ b) Father’s
Occupation__________________
12) Please check all that apply to your child:
Learning Disability

Seizures

Developmental Disability

Brain Injury

Psychiatric Diagnoses

Vision Impairments

Pre-mature Birth

Hearing Impairment

117

APPENDIX B: ADOLESCENT DEMOGRAPHIC SURVEY
1.
2.
3.
4.
5.
6.
7.

What is your age? _____
What country were you born in? ____________________
If born outside the U.S., what age did you arrive in the U.S.?___________________
Have you ever visited your country of origin, if so, how many times? ___________________
What country was your mother and father born in? ____________________
How long have your family lived in the U.S.? ____________________
How many years have you attended English as a Second Language (ESL) classes?
________________
8. What is your overall estimated GPA?____________
9. How do you identify yourself? (please choose one)
1- Arabic

2- American

3- Arab American

10. How do your parents identify themselves? (please choose one)
1- Arabic

2- American

3- Arab American

11. How religious are you? (please choose one)
1- Not Religious

2- Somewhat Religious

3- Very Religious

12. How often do you practice your religion? (please choose one)
1- Never

2- Sometimes/ Special Occasion like Eid or Ramadan

3- Daily

13. How religious are your parents? (please choose one)
1- Not Religious

2- Somewhat Religious

3- Very Religious

14. In school, most of your friends are? (please choose one)
1- Arab

2- non-Arab

15. Outside of school, most of your friends are? (please choose one)
1- Arab

2- non-Arab

16. How religious are you? (please choose one)
1- Not Religious

2- Somewhat Religious

3- Very Religious

17. How often do you communicate with family in your country of origin? (please choose one)
1- Daily

2- Weekly

3- Monthly

4-Yearly

5- Never

18. In school, you speak English or Arabic? (please choose one)
1- English Only

2- Mostly English

3- About the Same

118

4- Mostly Arabic

5- Only Arabic

19. At home, you speak English or Arabic? (please choose one
1- English Only

2- Mostly English

3- About the Same

119

4- Mostly Arabic

5- Only Arabic

APPENDIX C: ACCULTURATION SCALE: AMAS
Abbreviated Multidimensional Acculturation Scale ( Zea, Asner-Self, Birman, & Buki, 2003)
The following section contains questions about your culture of origin and your native language.
By culture of origin we are referring to the culture of the country either you or your parents came
from (e.g., Yemen). By native language we refer to the language of that country, spoken by you
or your parents in that country (e.g. Arabic).
Instructions: Please mark the number from the scale that best correspond to your answer.
1
Strongly Disagree

2
Somewhat Disagree

3

4

Somewhat Agree

Strongly Agree

ABBREVIATED MULTIDIMENSIONAL ACCULTURATION SCALE
1. I think of myself as being American.
2. I feel good about being American.
3. Being American plays an important part in my life.
4. I feel that I am part of American culture.
5. I have a strong sense of being American.
6. I am proud of being American.
7. I think of myself as being Arabic
8. I feel good about being Arabic
9. Begin Arabic plays an important part in my life.
10. I feel that I am part of the Arabic culture.
11. I have a strong sense of being Arabic.
12. I am proud of being Arabic.
Please answer the questions below using the following response
1
Not at all

2

3

A little

Pretty well

How well do you speak English:
13. At school?
14. With Americans?
15. On the phone?
16. With strangers?
120

4
Extremely well

17. In general?
How well do you understand English:
18. On television or in movies?
19. In newspapers and magazines?
20. Words in songs?
21. In general?
How well do you speak Arabic:
22. with family?
23. With Arabic or other Arabs?
24. On the phone?
25. With strangers?
26. In general?
How well do you understand Arabic:
27. on television or in movies?
28. in newspapers and magazines?
29. words in songs?
30. in general?
How well do you know:
31. American national heroes?
32. Popular American television shows?
33. Popular American newspapers and magazines?
34. Popular American actors and actresses?
35. American history?
36. American political leaders?
37. Arabic national heroes
38. popular Arabic/Arabic television shows
39. popular Arabic newspapers and magazines?
40. popular Arabic actors and actresses
41. Arabic, Islamic, or Arabic history?
42. Arabic/ Arabic political leaders

121

Subscales available:
IDENTITY:
American Cultural Identity (Possible scores range 6-24; 6-12 = low, 13-24 = high)
Arabic Cultural Identity (Possible scores range 6-24; 6-12 = low, 13-24 = high)
LANGUAGE:
English Language Competence (Possible scores range 9-36; 9-18 = low, 19-36 = high)
Arabic Language Competence (Possible scores range 9-36; 9-18 = low, 19-36 = high)
CULTURE:
American Culture Competence (Possible scores range 6-24; 6-12 = low, 13-24 = high)
Arabic/Arabic Culture Competence (Possible scores range 6-24; 6-12 = low, 13-24 = high)
UNIDIMENSIONAL MODEL OF ACCULTURATION
Total possible Arabic/Arabic 21-84 (21-41 = low, 42 = high)
Total possible American 21-48 (21-41 = low, 42 = high)
BERRY’S BIDIMENSIONAL CLASSIFICATION
Integrated = High American and High Arabic
Separate = High Arabic and Low American
Assimilated = High American and Low Arabic
Marginalized = Low Arabic and Low American (will not be included to anticipated low
numbers)

122

APPENDIX D: PARENT INFORMED CONSENT
Your child is being asked to participate in a research project. Researchers are required to provide
a consent form to inform you about the study, to inform you that participation is voluntary, to
explain risks and benefits of participation, and to allow you to make an informed decision. You
should feel free to ask the researchers any questions you may have.
Study: The Association between Acculturation and Intellectual Assessment among Arab American
Adolescents
Researchers: Ramzi M. Hasson, Doctoral Candidate in School Psychology
Jodene G. Fine, Ph.D., Assistant Professor, School Psychology & Educational
Psychology
Department of Counseling, Educational Psychology, and Special Education
Michigan State University
PURPOSE OF RESEARCH:
Research shows that your cultural background and language levels influence your performance
on cognitive tests. The assessment of cognitive ability plays an important role in clinical and
educational services provided to individuals from many different cultures. There is no research,
however, on the connection between culture, language and cognitive ability among Arab
American children. In this study, the researchers hope to learn about how your child’s
development in the U.S. is related to how well they perform on cognitive tests. All children and
adolescents ages 12-17 that identify themselves as being of Arab descent are invited to
participate in this research study. Your child’s participation in the study will take about 60 to 90
minutes. Your participation will take approximately 5 minutes.
WHAT YOU AND YOUR CHILD WILL DO:
Your child will complete a brief background information form, an acculturation scale and
undergo a brief psycho-educational assessment. The acculturation scale will ask questions about
cultural identity, language competence, and cultural knowledge. An adult will either read the
questions to your child or your child can complete the survey by him or herself. The direct
assessment will involve an evaluation of cognitive ability and academic functioning. You will be
asked to complete a demographic survey that will include questions about your family
background, home life, language use, and your child’s developmental history. You will be able
to complete this survey individually at home or through a phone call by the examiners. The
surveys will be completed for research purposes only, and the results will not be shared with you
or your child. Please note, this consent form must be signed and returned for participation to be
possible.
POTENTIAL BENEFITS:
You or your child will not directly benefit from participation in this study, but participation may
help contribute to a better understanding of how culture and language influences cognitive
performance. This study, along with future research, may increase our knowledge to reduce bias
during standardized intellectual assessment among immigrant youth.

123

POTENTIAL RISKS:
This study poses minimal risk for you and your child. You will complete a survey that includes
some questions that may cause you to experience some discomfort. Your child will be complete
a survey that includes some questions about adjustment, and cultural identity, which may cause
him/her to experience some discomfort or distress. You and your child may skip any question.
Your child will also undergo a cognitive evaluation that will last approximately 60-90 minutes.
Some discomfort maybe experienced during the assessment.
PRIVACY AND CONFIDENTIALITY:
The data for this project will be kept confidential to the greatest extent allowable by law. Only
the researchers and Michigan State University’s Human Research Protection Program will have
access to the data. After you and your child complete the survey, an identification number will
be assigned to the survey and the cognitive assessment results. The consent form with you and
your child’s names will be separated from the survey and filed in a locked cabinet. Completed
surveys will be kept in a in a locked office of the researcher. All documents will be destroyed ten
years after completion. The results of this study may be published or presented at professional
meetings, but the identities of all research participants will remain anonymous. It will not be
possible for readers to know who participated in the study.
As in all research, there is a possibility that some information is shared by a child about doing
serious harm to themselves or others. If keeping information obtained in this study private would
immediately put them or someone else in danger, the investigators would release that
information to protect them or another person.
YOUR RIGHTS TO PARTICIPATE, SAY NO, OR WITHDRAW:
Participation in this research project is completely voluntary. You and your child have the right
to say no. You or your child may change your minds at any time and withdraw from the study.
You and your child may also choose not to answer specific question or to stop participating at
any time.
COSTS AND COMPENSATION FOR BEING IN THE STUDY:
It does not cost anything to participate in this study. As an appreciation of your time, you will be
entered into a raffle for a chance to win a $50 gift card. Your child will be given a $10 gift card
as an appreciation of their time.
CONTACT INFORMATION FOR QUESTIONS AND CONCERNS:
If you have concerns or questions about this study, such as scientific issues, how to do any part
of it, or to report an injury, please contact the researchers, Ramzi Hasson, by phone: 313-2836056; email: hassonra@msu.edu or Dr. Jodene G. Fine, by phone: 517-884-0443; email:
finej@msu.edu; 439 Erickson Hall, East Lansing, MI, 48824. If you have questions or concerns
about your role and rights as a research participant, would like to obtain information or offer
input, or would like to register a complaint about this study, you may contact, anonymously if
you wish, the Michigan State University’s Human Research Protection Program at 517-3552180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail at 202 Olds Hall, MSU, East
Lansing, MI 48824.
124

DOCUMENTATION OF INFORMED CONSENT
The Association between Acculturation and Intellectual Assessment among Arab American
Adolescents
Please select a box, fill in your child’s name, and sign below.
Yes, my child _____________________________________ may participate in this
research study.
Please Print Child’s Name
No, my child ______________________________________ may not participate in this
research study.
Please Print Child’s Name

________________________________________
Signature

_____________________________
Date

PLEASE PRINT:

_______________________________________

Circle One:

Mother

Father

Other________ Parent/Guardian Name

125

APPENDIX E: CHILD ASSENT FORM
Study Title: The Association between Acculturation and Intellectual Assessment among Arab
American Adolescents
This is a research study and you do not have to take part. You are being asked to take part in this
study because you identified yourself as being of Arab descent. In this study, the researchers
hope to learn about how the experience of Arab American youth influences performance on
cognitive tests. About 80 adolescents will participate in this study.
What will happen if I take part in this study?
If you agree to be in this study, you will complete a survey and undergo a cognitive assessment
that takes between 60 to 90 minutes. The surveys will ask about your culture, family, and
language background. An adult will either read the questions to you or you can complete the
survey by yourself. The surveys are for research purposes only, and the results will not be shared
with you, your teachers, or your parents.
Are there any risks to me or my privacy?
Some of the survey questions and the direct assessment may make you feel uncomfortable. You
may skip any question. We will do our best to protect the information we collect from you. We
will give you an ID code and not use your name. The completed surveys and tests will be kept
secure and separate from information which identifies you. Only a small number of researchers
will have access to the surveys. Names and information that might identify you will not be used
if the study is published or presented at scientific meetings.
Are there benefits?
There is no direct benefit to you. Your participation in this study may help us how to best assess
children from immigrant backgrounds.
Can I say “No”?
Yes, you do not have to complete a survey or the testing. If you choose not to be in this study
you will not lose any of your regular benefits.
Are there any payments or costs?
It does not cost anything to participate in this study. To thank you for your time, you will be
given a $10 gift card.
Who can answer my questions about the study?
You can talk with the study researcher about any questions, concerns, or complaints you have
about this study. Contact the study researchers: Ramzi Hasson, by phone: 313-283-6056; email:
hassonra@msu.edu or Dr. Jodene Fine, by phone: 517-884-0443; email: finej@msu.edu; 435
Erickson Hall, East Lansing, MI, 48824.
If you wish to ask questions about the study or your rights as a research participant to someone
other than the researchers please call the Michigan State University’s Human Research

126

Protection Program at 517-355-2180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail
at 202 Olds Hall, MSU, East Lansing, MI 48824.
Assent
PARTICIPATION IN RESEARCH IS VOLUNTARY.
You have been given copies of this consent form to keep. If you wish to be in this study, please
sign below.

Date

Participant's Signature for Assent

Date

Person Obtaining Assent

127

BIBLIOGRAPHY

128

BIBLIOGRAPHY
Arab American Institute. (2002). Zogby International Survey. Retrieved from
http://www.aaiusa.org/arab-americans/22/demographics
Arab American Institute Foundation (n.d.). Quick facts about Arab Americans. Washington, DC:
Author. Retrieved October 27, 2010 from http//www.aaiusa.org/PDF/quickfast.pdf
Arab American Institute. (2007). U.S. Bureau of the Census 2005-2007 American Community
Survey 3-year estimates Arab American Institute Foundation.
Abudabbeh, N. (1996). Arab families. In M. McGooldrick, J., Giordano, & Pearce, K. (Eds.),
Ethnicity and family therapy (333-346). New York: Guilford.
Abudabbeh, N., & Hays, P. A. (2006). Cognitive-behavioral therapy with people of Arab
heritage.In Hays, P. A. & Gayle, Y. (Eds.), Culturally responsive cognitive-behavioral
therapy: assessment, practice, and supervision (141-159). Washington, DC: American
Psychological Association.
Ahmed, W., & Bruinsma, M. (2006). A structural model of self-concept, autonomous motivation
and academic performance in cross-cultural perspective. Electronic Journal of Research
in Educational Psychology, 4, 551-576.
Ajrouch, K. J., & Jamal, A. (2007). Assimilating to a white identity: the case of Arab
Americans. International Migration Review, 41, 860-879.
Amer, M. M., & Hovey, J. D. (2005). Examination of the impact of acculturation, stress, and
religiosity on mental health variables for second-generation Arab Americans. Ethnic
Dissertations, 15(1 Suppl 1), S1-111.
Anastasi, A., & Urbina, S. (1997). Psychological testing (7th ed). Upper Saddle River, NJ:
Eribaum.
American Psychological Association (2003). Guidelines on multicultural education, training,
research, practice, and organizational change for psychologists. American Psychologist,
58, 377-402.
Ardila, A. (2005). Cultural values underlying psychometric cognitive testing. Neuropsychology
Review, 15, 185-195.
Ardila, A., Rosselli, M., & Rosas, P. (1989). Neuropsychological assessment in illiterates:
visuospatial and memory abilities. Brain and Cognition, 11, 147-166.
Berry, J. W. (2003). Conceptual approaches to acculturation. In K. M. Chun, P. B. Organista &

129

G. Marin (Eds.), Acculturation: Advances in Theory, Measurement, and Applied
Research(17-37). Washington, D.C.: American Psychological Association.
Berlin, L. J., Ispa, J. M., Fine, M. A., Malone, P. S., Brooks‐Gunn, J., Brady‐Smith, C…&
Bai,Y. (2009). Correlates and consequences of spanking and verbal punishment for low‐
income White, African American, and Mexican American toddlers. Child Development,
80, 1403-1420.
Berry, J. W. (1997). Immigration, acculturation, and adaptation. Applied Psychology,46, 5-34.
Bialystok, E. (2011). Reshaping the mind: the benefits of bilingualism. Canadian Journal of
Experimental Psychology/Revue canadienne de psychologie expérimentale, 65, 229-257.
Boake, C. (2002). From the Binet-Simon to the Wechsler-Bellevue: Tracing the history of
intelligence testing. Journal of Clinical and Experimental Psychology, 24, 383-405.
Boone, K. B., Victor, T. L., Wen, J., Razani, J., & Pontón, M. (2007). The association between
neuropsychological scores and ethnicity, language, and acculturation variables in a large
patient population. Archives of Clinical Neuropsychology, 22, 355-365.
Bradley, R. H., & Corwyn, R. F. (2002). Socioeconomic status and child development. Annual
Review of Psychology, 53, 371-399.
Breslau, N., Johnson, E. O., & Lucia, V. C. (2001). Academic achievement of low birthweight
children at age 11: the role of cognitive abilities at school entry. Journal of Abnormal
Child Psychology, 29, 273-279.
Breier, J. I., Fletcher, J. M., Wheless, J. W., Clark, A., Cass, J., & Constantinou, J. E. (2000).
Profiles of cognitive performance associated with reading disability in temporal lobe
epilepsy. Journal of Clinical and Experimental Neuropsychology, 22, 804-816.
Brittingham, A., & De la Cruz, G. P. (2005).We the people of Arab ancestry in the United States
(pp. 1-24). Washington, DC: US Census Bureau.
Brown, R., & Zagefka, H. (2011). The dynamics of acculturation: An intergroup perspective.
Advances in Experimental Social Psychology, 44, 129.
Budd, K. S., Felix, E. D., Poindexter, L. M., Naik-Polan, A. T., & Sloss, C. F. (2002). Clinical
assessment of children in child protection cases: An empirical analysis. Professional
Psychology: Research and Practice, 33, 3.
Cabassa, L. J. (2003). Measuring acculturation: Where we are and where we need to go.
Hispanic Journal of Behavioral Sciences, 25, 127-146.
Cahan, S., & Cohen, N. (1989). Age versus schooling effects on intelligence development. Child
Development, 60, 1239-1249.

130

Carhill, A., Suárez-Orozco, C., & Páez, M. (2008). Explaining English language proficiency
among adolescent immigrant students. American Educational Research Journal, 45,
1155-1179.
Gim Chung, R. H., Kim, B. S., & Abreu, J. M. (2004). Asian American multidimensional
acculturation scale: development, factor analysis, reliability, and validity. Cultural
Diversity and Ethnic Minority Psychology, 10, 66.
Coffey, D. M., Marmol, L., Schock, L., & Adams, W. (2005). The influence of acculturation on
the Wisconsin Card Sorting Test by Mexican Americans. Archives of Clinical
Neuropsychology, 20, 795-803.
Coll, C. G., Crnic, K., Lamberty, G., Wasik, B. H., Jenkins, R., Garcia, H. V., & McAdoo, H. P.
(1996). An integrative model for the study of developmental competencies in minority
children. Child development, 67, 1891-1914.
Costigan, C. L., & Dokis, D. P. (2006). Relations between parent–child acculturation differences
and adjustment within immigrant Chinese families. Child Development, 77, 1252-1267.
Cummins, J. (1983). Bilingualism and special education: Program and pedagogical issues.
Learning Disability Quarterly, 6, 373-386.
Dana, R. H. (2005). Multicultural Assessment: Principles, Applications, and Examples. Mahwah,
NJ: Erlbaum.
Delvin, B, Daniels, M. & Roeder, K. (1997). The heritability of IQ. Nature, 388, 468-471
Delaney-Black, V., Covington, C., Ondersma, S. J., Nordstrom-Klee, B., Templin, T., Ager, J.,
& Sokol, R. J. (2002). Violence exposure, trauma, and IQ and/or reading deficits among
urban children. Archives of Pediatrics & Adolescent Medicine, 156, 280-285.
Denckla, M. B. (1996). A theory and model of executive function: A neuropsychological
perspective. In Reid, G. L., & Kransnegor, N. A. (Eds.) Attention, memory, and executive
function, (263-278). Baltimore, MD: Paul H Brookes Publishing.
Dickins, T. E., Sear, R., & Wells, A. J. (2007). Mind the gap (s)… in theory, method and data:
Re‐examining Kanazawa (2006).British Journal of Health Psychology,12, 167-178.
Downey, D. B., & Pribesh, S. (2004). When race matters: Teachers' evaluations of students'
classroom behavior. Sociology of Education, 77, 267-282.
Dynda, A. (2008). The relation between language proficiency and IQ test performance.
Unpublished manuscript, St. John's University, Oakdale, NY.
Eccles, J. S., Wong, C. A., & Peck, S. C. (2006). Ethnicity as a social context for the

131

development of African-American adolescents. Journal of School Psychology, 44, 407426.
Erickson, C. D., & Al-Timimi, N. R. (2001). Providing mental health services to Arab
Americans: Recommendations and considerations. Cultural Diversity and Ethnic
Minority Psychology, 7, 308.
Flanagan, D. P., Ortiz, S. O., & Alfonso, V. C. (2007). Essentials of cross-battery
assessment (Vol. 84). John Wiley & Sons.
Flanagan, D. P., & Harrison, P. L. (Eds.). (2012). Contemporary Intellectual Assessment:
Theories, Tests, and Issues. New York, NY: Guilford Press.
Fuster, J.M. (2008). The prefrontal cortex, fourth edition. London, England: Academic Press.
Gardner, H. (2011). The unschooled mind: How children think and how schools should teach.
Basic books.
Gardner, H. (1985). Frames of mind: The theory of multiple intelligences. Basic books.
Gil, A. G., Vega, W. A., & Dimas, J. M. (1994). Acculturative stress and personal adjustment
among Hispanic adolescent boys. Journal of Community Psychology, 22(1), 43-54.
Gladstone, R. (2012). Anti-American protests flare beyond the Mideast. New York Times.
Retrieved from http://www.nytimes.com/2012/09/15/world/middleeast/anti-americanprotests-over-film-enter-4th-day.html
Goodman, A. H. (2000). Why genes don't count (for racial differences in health). American
Journal of Public Health, 90, 1699.
Gordon, M. M. (1964). Assimilation in American life: The role of race, religion and
national origins. New York, NY: Oxford.
Graves, T. D. (1967). Psychological acculturation in a tri-ethnic community. Southwestern
Journal of Anthropology, 337-350.
Green, R. L., Hoffman, L. J., Morse, R. J., Hayes, M. E., & Morgan, R. F. (1964). The education
of children in a district without schools. East Lansing, MI: Michigan State University,
College of Education.
Gregg, G. S. (2005). The Middle East: A Cultural Psychology. Oxford: Oxford University Press.
Grigsby, J., Rosenberg, N. L., & Busenbark, D. (1995). Chronic pain is associated with deficits
in information processing. Perceptual and Motor skills, 81(2), 403-410.
Harris, J. G., Tulsky, D. S., & Schultheis, M. T. (2003). Assessment of the non-native English

132

speaker: Asssimilating history and research findings to guide clinical practice. In Tulsky,
D. S., Saklofske, D. H., Chelune, G. J., Heaton, R. K., Ivnik, Bornstein, R…& Ledbetter,
M. F. Clinical interpretation of the WAIS-III and WMS-III, (343-387). Sand Diego, CA:
Academic.
Heaton, R. K., Taylor, M., &Manly, J. (2003). Demographic effects and use of demographically
corrected norms with the WAIS-III and WMS-III. In Tulsky, D. S., Saklofske, D. H.,
Chelune, G. J., Heaton, R. K., Ivnik, Bornstein, R…& Ledbetter, M. F. Clinical
interpretation of the WAIS-III and WMS-III, (181-207). Sand Diego, CA: Academic.
Helms-Lorenz, M., Van de Vijver, F. J., & Poortinga, Y. H. (2003). Cross-cultural differences in
cognitive performance and Spearman's hypothesis: g or c?. Intelligence, 31, 9-29.
Herrnstein, R. J., & Murray, C. (1994). The Bell Curve: The Reshaping of American Life by
Differences in Intelligence. New York: Free Press.
Hernandez, D. J., Denton, N. A., & Macartney, S. E. (2007). Family circumstances of children in
immigrant families: Looking to the future of America. New York, NY: Guilford.
Jokeit, H., & Ebner, A. (2002). Effects of chronic epilepsy on intellectual functions. Progress in
Brain Research, 135, 455-463.
Kayyali, R. A. (2006). The Arab Americans. New York, NY: Greenwood Publishing Group.
Kennepohl, S. (1999). Toward a cultural neuropsychology: An alternative view and a
preliminary model. Brain and Cognition, 41, 365–380.
Knox, D. W. (1915). The Role of Doctrine in Naval Warfare. United States Naval Institute.
Kranzler, J. H., Flores, C. G., & Coady, M. (2010). Examination of the Cross-Battery Approach
for the Cognitive Assessment of Children and Youth From Diverse Linguistic and
Cultural Backgrounds. School Psychology Review, 39, 431-445.
Kugelmass, S., Lieblich, A., & Bossik, D. (1974). Patterns of intellectual ability in Jewish and
Arab children in Israel. Journal of Cross-Cultural Psychology, 5, 184-198.
Lewontin, R. C. (1993). The Doctrine of DNA: Biology as Ideology. Harmondsworth: Penguin.
Lieblich, A., & Kugelmass, S. (1981). Patterns of intellectual ability of Arab school children in
Israel. Intelligence,5, 311-320.
Lezak, M. D., Howieson, D. B., Bigler, D., & Tranel, D. (2012). Neuropsychological assessment,
5th edition. Oxford university press.
Lohman, D. F., Korb, K. A., & Lakin, J. M. (2008). Identifying academically gifted english-

133

language learners using language-reduced tests a comparison of the Raven, NNAT, and
CogAT. Gifted Child Quarterly, 52, 275-296.
Luria, A. R. (1976). Cognitive development: Its cultural and social foundations. Boston,
MA: Harvard University Press.
MacMillan, D. L., & Forness, S. R. (1998). The role of IQ in special education placement
decisions primary and determinative or peripheral and inconsequential? Remedial and
Special Education, 19, 239-253.
Manly, J. J. (2008). Critical issues in cultural neuropsychology: profit from diversity.
Neuropsychology Review, 18, 179-183.
Manly, J. J. (2005). Advantages and disadvantages of separate norms for African
Americans. The Clinical Neuropsychologist, 19, 270-275.
McDaniel, M. A. (2005). Big-brained people are smarter: A meta-analysis of the relationship
between in vivo brain volume and intelligence. Intelligence, 33, 337-346.
Mercer, J. R. (1974). A policy statement on assessment procedures and the rights of children.
Harvard Educational Review, 44, 125-141.
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D.
(2000). The unity and diversity of executive functions and their contributions to complex
“frontal lobe” tasks: A latent variable analysis. Cognitive Psychology, 41, 49-100.
Molfese, V. J., Molfese, D. L., Modglin, A. T., Walker, J., & Neamon, J. (2004). Screening early
reading skills in preschool children: Get ready to read. Journal of Psychoeducational
Assessment, 22, 136-150.
Mindt, M., Byrd, D., Saez, P., & Manly, J. (2010). Increasing culturally competent
neuropsychological services for ethnic minority populations: A call to action. The
Clinical Neuropsychologist, 24, 429-453.
Moody, A. (2014). Linguistic Factors and Language Assessment of Asians. In Neuropsychology
of Asians and Asian-Americans (pp. 27-54). New York, NY: Springer.
Naglieri, J. A., & Bornstein, B. T. (2003). Intelligence and achievement: Just how correlated are
they? Journal of Psychoeducational Assessment, 21(3), 244-260.
Neisser, U., Boodoo, G., Bouchard Jr, T. J., Boykin, A. W., Brody, N., Ceci, S. J., ... & Urbina,
S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 77.
Nell, V. (2000). Cross-cultural neuropsychological assessment: Theory and practice. Mahwah,
New Jersey London: Lawrence Erlbaum Associates Publishers.

134

Nieves-Brull, A. I. (2006). Evaluation of the Culture-language Matrix: A Validation Study of
Test Performance in Monolingual English Speaking and Bilingual English/Spanish
Speaking Populations. Unpublished doctoral dissertation, St. John’s University, Queens,
New York.
Ochoa, S. H., Riccio, C., Jimenez, S., de Alba, R. G., & Sines, M. (2004). Psychological
assessment of English language learners and/or bilingual students: An investigation of
school psychologists' current practices. Journal of Psychoeducational Assessment, 22,
185-208.
Ogbu, J. U. (Ed.). (2008). Minority status, oppositional culture, & schooling. New York, NY:
Routledge.
Ogden, J. A., & McFarlane-Nathan, G. (1997). Cultural bias in the neuropsychological
assessment of young Maori men. New Zealand Journal of Psychology, 26, 2-12.
Pollitt, E., Gorman, K. S., Engle, P. L., Martorell, R., Rivera, J., Wachs, T. D., & Scrimshaw, N.
S. (1993). Early supplementary feeding and cognition: effects over two decades.
Monographs of the Society for Research in Child Development, 58, i-118.
Razani, J., Burciaga, J., Madore, M., & Wong, J. (2007). Effects of acculturation on tests of
attention and information processing in an ethnically diverse group. Archives of Clinical
Neuropsychology, 22, 333-341.
Redfield, R., Linton, R., & Herskovits, M. J. (1936). Memorandum for the study of
acculturation. American Anthropologist, 38, 149-152.
Rhodes, R. L., Ochoa, S. H., & Ortiz, S. O. (2005). Assessing culturally and linguistically
diverse students: A practical guide. New York, NY: Guilford Press.
Richards, E. J. (2006). Inherited epigenetic variation—revisiting soft inheritance. Nature
Reviews Genetics, 7, 395-401.
Rohde, T. E., & Thompson, L. A. (2007). Predicting academic achievement with cognitive
ability. Intelligence, 35, 83-92.
Rogers, M. R., Ingraham, C. L., Bursztyn, A., Cajigas-Segrede, N., Esquivel, G., Hess, R., ... &
Lopez, E. C. (1999). Providing psychological Services to racially, ethnically, eulturally,
and linguistically diverse Individuals in the schools: recommendations for practice.
School Psychology International, 20, 243-64.
Romero, A. J., & Roberts, R. E. (2003). Stress within a bicultural context for adolescents of
Mexican descent. Cultural Diversity and Ethnic Minority Psychology, 9, 171.
Rudmin, F. (2009). Constructs, measurements and models of acculturation and acculturative
stress. International Journal of Intercultural Relations, 33, 106-123.

135

Rushton, J. P., & Jensen, A. R. (2005). Thirty years of research on race differences in cognitive
ability. Psychology, Public Policy, and Law, 11, 235-255.
Rushton, J. (1994). Sex and race differences in cranial capacity from International Labour Office
data. Intelligence, 19, 281-294.
Ryder, A. G., Alden, L. E., & Paulhus, D. L. (2000). Is acculturation unidimensional or
bidimensional? A head-to-head comparison in the prediction of personality, self-identity,
and adjustment. Journal of personality and social psychology, 79, 49.
Salvia, J., Ysseldyke, J. E., & Bolt, S. (2007). Assessment in special and inclusive education.
New York, NY: Houghton Miflin.
Sam, D. L., & Berry, J. W. (2010). Acculturation when individuals and groups of different
cultural backgrounds meet. Perspectives on Psychological Science, 5, 472-481.
Sattler, J. M. (2008). Assessment of children: Cognitive foundations. La Mesa, CA: JM Sattler.
Styck, K. M., & Watkins, M. W. (2014). Discriminant Validity of the WISC-IV CultureLanguage Interpretive Matrix. Contemporary School Psychology, 18(3), 168-177.
Hunter, J. E., & Schmidt, F. L. (Eds.). (2004). Methods of meta-analysis: Correcting error and
bias in research findings. New York, NY: Sage.
Salgado, J. F., Anderson, N., Moscoso, S., Bertua, C., De Fruyt, F., & Rolland, J. P. (2003). A
meta-analytic study of general mental ability validity for different occupations in the
European community. Journal of Applied Psychology, 88, 1068.
Shaheen, J. G. (2003). Reel bad Arabs: How Hollywood vilifies a people. The Annals of the
American Academy of Political and Social Science, 588, 171-193.
Shobris, J. G. (1996). The anatomy of intelligence. Genetic, Social, and General Psychology
Monographs, 122, 133-158.
Sotelo-Dynega, M., Cuskley, T., Geddes, L., McSwiggan, K., & Soldano, A. (2011). Cognitive
assessment: A survey of current school psychologists’ practices. In Poster presented at
the 2010 convention of the National Association of School Psychologists, San Francisco,
CA.
Springer, P. R., Abbott, D. A., & Reisbig, A. M. (2009). Therapy with Muslim couples and
families: Basic guidelines for effective practice. The family Journal, 17, 229-235.
Sternberg, R. J. (1985). Implicit theories of intelligence, creativity, and wisdom. Journal of
Personality and Social Psychology, 49, 607.

136

Sullivan, A. L. (2011). Disproportionality in special education identification and placement of
English language learners. Exceptional Children, 77, 317-334.
te Nijenhuis, J., & Dragt, J. (2010). Causes of group differences studied with the method of
correlated vectors: A psychometric meta-analysis of Spearman's hypothesis. Unpublished
manuscript, University of Amsterdam, the Netherlands.
te Nijenhuis, J., & Repko, J. (2011). Spearman's hypothesis tested with Raven's Progressive
Matrices: A psychometric meta-analysis. Unpublished manuscript, University of
Amsterdam, the Netherlands.
Tombaugh, T. N. (1996). TOMM, Test of Memory Malingering. New York: Multi-Health
Systems.
Triandis, H. C. (1995). Individualism & collectivism. New York, NY: Westview Press.
Turkheimer, E., Haley, A., Waldron, M., d'Onofrio, B., & Gottesman, I. I. (2003).
Socioeconomic status modifies heritability of IQ in young children. Psychological
Science, 14, 623-628.
Tzuriel, D., & Kaufman, R. (1999). Mediated learning and cognitive modifiability dynamic
assessment of young Ethiopian immigrant children to Israel. Journal of Cross-Cultural
Psychology, 30, 359-380.
U.S. Department of Education, National Center for Education Statistics. (2009). The Condition
of Education 2009 (NCES 2009-081), indicators 2, 3, 12, 13, and 20. Washington, DC:
U.S. Government Printing Office.
van der Linden, D., Frese, M., & Meijman, T. F. (2003). Mental fatigue and the control of
cognitive processes: Effects on perseveration and planning. Acta Psychologica
(Amsterdam), 113, 45–65.
van de Vijver, F. J. (2008). On the meaning of cross-cultural differences in simple cognitive
measures. Educational Research and Evaluation, 14, 215-234.
van de Vijver, F., & Tanzer, N. K. (2004). Bias and equivalence in cross-cultural assessment:
An overview. Revue Européenne de Psychologie Appliquée/European Review of Applied
Psychology, 54, 119-135.
van de Vijver, F. J., Helms-Lorenz, M., & Feltzer, M. J. (1999). Acculturation and cognitive
performance of migrant children in the Netherlands. International Journal of Psychology,
34, 149-162.
van de Vijver, F. J. (2008). On the meaning of cross-cultural differences in simple cognitive
measures. Educational Research and Evaluation, 14, 215-234.

137

Verderosa, F. A. (2007). Examining the effects of language and culture on the differential ability
scales with bilingual preschoolers. Published dissertation, ProQuest.
Vukovich, D., & Figueroa, R. A. (1982). The validation of the system of multicultural pluralistic
assessment: 1980-1982. Unpublished manuscript, University of California at Davis,
Department of Education.
Vygotsky, L. S. (1978). Mind and society: The development of higher mental processes. Boston,
MA, Harvard University Press.
Watkins, M. W., Lei, P. W., & Canivez, G. L. (2007). Psychometric intelligence and
achievement: A cross-lagged panel analysis. Intelligence, 35, 59-68.
Wechsler, D. (2011). Wechsler Abbreviated Scale of Intelligence--Second Edition. San Antonio,
Tx: The Psychological Corporation
Wrobel, N. H., Farrag, M. F., & Hymes, R. W. (2009). Acculturative stress and depression in an
elderly Arabic sample. Journal of Cross-Cultural Gerontology, 24, 273-290.
Yerkes, R. M. (1921). Psychological Examining in the United States Army: Edited by Robert M.
Yerkes (Vol. 15). US Government Printing Office.
Yoon, E., Langrehr, K., & Ong, L. Z. (2011). Content analysis of acculturation research in
counseling and counseling psychology: a 22-year review. Journal of Counseling
Psychology, 58, 83.
Zea, M. C., Asner-Self, K. K., Birman, D., & Buki, L. P. (2003). The Abbreviated
Multidimentional Acculturation Scale: Empirical validation with two Latino/Latina
samples. Cultural Diversity and Ethnic Minority Psychology, 9, 107.
Zogby, J. J., & Arab Thought Foundation. (2002). What Arabs Think: Values, Beliefs and
Concerns. Washington, DC: Zogby International.

138