mg.
3...»?
~ . g

tummy

 

 

 

 

 

“Lamﬂmm

a: In

.2: is.
éﬁﬁquhuvuﬁw. I.

w.- .3313...

 

I “gun...
to“.- l’i
.5 5.!
.3550”...
van»

9...

2.1
.1553}. .
: 1.1. £6.10...
‘3. :3na. 1.3.5!-
!l: .1. \\..1l:?
31...... Esta
. sandwui .9 l
7 {Yin .. «)3. 6
3 l. Vixilxhﬁu
‘ a}? .. .

. any...“

I: Inf:

:q....%.~rtr..:!1.§

. ..?.}.?.21?.!?
3.1.!- 3.55.0153.

1.0.! at... .
2:1 21.61. A...
(t2; .511...

5. ..

 

4..“ 1 [CK
1.

 

 

 

 

Z 7 LIBRARY
2009 Michigan ..tate
University

 

 

This is to certify that the
dissertation entitled

THE MULTILINGUAL/BILINGUAL DICHOTOMY:
AN EXPLORATION OF INDIVIDUAL DIFFERENCES

presented by

Amy S. Thompson

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in Second Language Studies

 

 

 

got/lat, [QM

Major Professor’s Signature

[(461.41 ‘11 7. 200 9
Date

MSU is an Aﬁnnative Action/Equal Opportunity Employer

-.oc---u— --.-

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOlD FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE DATE DUE

 

pm: o 3: MM»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K‘IProjIAcc8Pres/ClRC/DaleDue indd

 

 

THE MULTILINGUAL/BILINGUAL DICHOTOMY:
AN EXPLORATION OF INDIVIDUAL DIFFERENCES

By

Amy S. Thompson

A DISSERTATION
Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of
DOCTOR OF PHILOSOPHY
Second Language Studies

2009

ABSTRACT

THE MU LTILINGUAL/BILINGUAL DICHOTOMY:
AN EXPLORATION OF INDIVIDUAL DIFFERENCES

By

Amy S. Thompson

Bilingualism (Sanz, 2000), motivation (Pintrich, 1989), and language aptitude
(Grigorenko, Stemberg, and Ehrman, 2000) are crucial individual differences that
contribute to successful adult language learning. Since Gardner’s (1985) seminal work on
motivation, many studies have shown that motivation is dynamic and that it affects
language development in many ways (Domyei, 2005). We also know that bilingualism
has clear beneﬁts for successful third language (L3) acquisition (Bialystok, 2006, 2001;
Da Fonttma & Siegle, 1995; Swain et al., 1990). Yet still at issue is how much
bilingualism contributes to the success of L3 acquisition when taking into consideration
other individual differences, such as motivation and aptitude.

This study addresses the proﬁles of second language (L2) and third language (L3)
learners of English, using 128 participants from the Casa de Cultura Britanica ’3 English
language program, which is afﬁliated with the Universidade Federal do Cearci in
Fortaleza, Brazil. The participants completed the Portuguese version of the Motivated
Strategies for Learning Questionnaire (MSLQ), Questionario de Estratégias
Motivacionais para Aprendizagem (QEMA) (Brown et al., 2001), the Cognitive Ability
for Novelty in Acquisition of Language (Foreign) (CANAL-FT) (Grigorenko et al.

2000), the Michigan

State University English Language Test (MSUELT) (Cook, 2001), and a self-rated
proﬁciency scale based on the Common European Framework of Reference (CEFR)
(Hasselgreen, 2005). Analyses of the aforementioned factors, including AN OVAs,
correlations, factor analyses, and discriminant function analyses compare the bilingual
and multilingual participants, illustrating the differences between the groups.
Motivational proﬁles, language aptitude scores, performance on the independent measure
of English proﬁciency, and retention rates of the bilingual and multilingual participants
are discussed using the aforementioned analyses. There is also a content analysis of
interview data collected from a subgroup of the multilingual participants. One important
ﬁnding of this dissertation includes the statistically signiﬁcant results for the language
aptitude scores between the bilingual and multilingual participants. This dissertation also
supports the slowly growing body of research indicating that even a small amount of
previous language learning experience can affect the process of learning subsequent
languages. In addition, language learners who positively perceive interactions between

languages studied might have advantages in subsequent language learning experiences.

Copyright by
Amy S. Thompson
2009

This dissertation is dedicated to my friends and family who have helped me get this far.

ACKNOWLEDGEMENTS
This dissertation would have not been possible without the help and support of many
friends, family, colleagues, and professors. I would like to start by thanking the faculty
and students at the Casa de Cultura Britanica ’s English language program, especially
Sﬁmia Carvalho, the director of the Casa de Cultura Britanz'ca at the time of this
dissertation. I would also like to individually thank each member of my dissertation
committee who have each contributed to my degree completion in speciﬁc ways:
Cristina Sanz, who put up with my ceaseless e-mails and questions about
multilingualism; Debra Friedman, who given many hours to patiently explaining the
intricacies of content analysis; Debra Hardison, for her incredible number of letters of
recommendation over the past six years and support throughout the TESOL and SLS
programs; Susan Gass, who has valued my development as a student and who always
made time for me regardless of her other time commitments; and ﬁnally my chair, Paula
Winke, who has offered constant support, encouragement, and optimism. I would also
like to offer special thanks to Bob Frye (Fry with an “e”), who was the ﬁrst person to tell
me that I had what it took to complete my doctorate and who has been my inspiration
both as a scholar and a teacher. Several ﬁiends and colleges have helped me in my
analyses and revisions including Junkyu Lee, Shaofeng Li, Sara Hillman, Emma
Trentman, Ching-Ni Hsieh, Baburhan Uzum, Ryan Miller, and Kimi Nakatsukasa. I
would also like to thank my Grandpa Thompson, who I know has been watching over me
through this process. Finally, I would also like to thank my family, Mom, Dad, Laura
(a.k.a. Laurat or My Favorite Ski), and Memo who have been my encouragement not

only in this endeavor, but in all others. I love you all MTH and TTMAB.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................ ix
LIST OF FIGURES ........................................................................................................... xi
CHAPTER ONE — INTRODUCTION AND REVIEW OF RELATED LITERATURE ..1
Introduction ..................................................................................................................... 1
Review of Related Literature .............................................................................................. 3
Bilingualism and Multilingualism ................................................................................... 4
Language Aptitude ........................................................................................................ 19
Motivation ..................................................................................................................... 29
Conclusion ..................................................................................................................... 37
CHAPTER TWO: RESEARCH DESIGN AND METHODOLOGY .............................. 39
Context for the Study .................................................................................................... 39
Participants .................................................................................................................... 40
Materials .......................................................................................... r .............................. 4 l
Multilingualism ......................................................................................................... 42
English Language Proﬁciency .................................................................................. 47
Motivation ................................................................................................................. 47
Language Aptitude .................................................................................................... 48
Interviews .................................................................................................................. 53
Procedure ....................................................................................................................... 57
CHAPTER THREE: QUANTITATIVE RESULTS ......................................................... 60
Analysis: Levels of English .......................................................................................... 60
Analysis: The operationalization of a multilingual learner ........................................... 65
Analysis: Research Questions ....................................................................................... 7O
RQl: Do bilingual learners have the same motivational proﬁle as multilingual
learners? .................................................................................................................... 70
RQ2: Which group of participants has higher language aptitude, bilingual or
multilingual learners? ................................................................................................ 92
RQ3: Which group of participants has higher English language proﬁciency as
measured by the MSUELT, bilingual or multilingual learners? ............................... 94
RQ4: Which group has a higher retention rate in the progression of the English
courses, bilingual or multilingual learners? .............................................................. 96
Analysis: Summary of Quantitative results .................................................................. 97
CHAPTER FOUR: QUALITATIVE RESULTS ............................................................ 100
Analysis: Part one — Motivation for language learning .............................................. 103
Analysis: Part two — Language interaction ................................................................ 114

vii

CHAPTER FIVE: DISCUSSION AND CONCLUSION ............................................... 124

Discussion ................................................................................................................... 124
RQl: Do bilingual learners have the same motivational proﬁle as multilingual
learners? .................................................................................................................. 126
RQ2: Which group of participants has higher language aptitude, bilingual or
multilingual learners? .............................................................................................. 131
RQ3: Which group of participants has higher English language proﬁciency as
measured by the MSUELT, bilingual or multilingual learners? ............................. 132
RQ4: Which group has a higher retention rate in the progression of the English
courses, bilingual or multilingual learners? ............................................................ 135

CHAPTER SIX: CONCLUSION ................................................................................... 138
APPENDIX A ................................................................................................................. 144
BIBLIOGRAPHY ........................................................................................................... 147

viii

LIST OF TABLES

Table 1: Carroll’s 4-Constituent Model for Language Aptitude ............................. 20
Table 2: Description of the MLAT ............................................................... 23

Table 3: The 5 Knowledge Acquisition Processes for the CANAL Theory. . ..........27

Table 4: Languages of the Multilingual Participants Divided by Subgroups ............... 41
Table 5: Sample Questions from the CEFR ...................................................... 43
Table 6: Sample Question from the MSLQ ...................................................... 48
Table 7: Description and Sample Questions from the CANAL-FT .......................... 49
Table 8: List of Questions Asked in the Oral Interviews ............... - ........................ 54
Table 9: Summary of Data Collection Tools .................................................... 56
Table 10: Summary of Data Collection Timeline ............................................... 59
Table 11: Correlations between MSUELT Scores and English Semester ................... 60
Table 12: Descriptive Statistics for the Overall MSUELT Scores ........................... 61

Table 13: Signiﬁcance of Tukey Post-Hoe between Semesters for the MSUELT... . . . . .63

Table 14: Descriptions of Groups Based on MSUELT Scores ................................ 64
Table 15: Correlations between the MSUELT and CEF R Scores ............................ 66
Table 16: Categorizations and Numbers of Multilingual Participants ........................ 70
Table 17: Underlying Constructs of the MSLQ ................................................. 71
Table 18: Item Number Comparison for the MSLQ and QEMA ............................. 74
Table 19: Exploratory PCA: Eigenvalues and Variance Explained for the First 5

Factors ................................................................................................ 79
Table 20: Eigenvalues and Variance Explained ................................................ 81
Table 21: Pattern Matrix for the Factor Loadings .............................................. 82

ix

Table 22: Structure Matrix for the Factor Loadings ............................................ 86
Table 23: Component Correlation Matrix for the Conﬁrmatory PCA ....................... 90

Table 24: Results of the Discriminant Function Analysis: Any Previous Language
Experience ............................................................................................ 92

Table 25: Results of the Discriminant Function Analysis: Perceived Positive Language
Interaction ............................................................................................ 92

Table 26: AN OVA Results for the Bilingual and Multilingual Participant CANAL-FT
Scores ................................................................................................. 93

Table 27: AN OVA Results for the Bilingual and Multilingual Participant MSUELT

Scores ................................................................................................. 95

Table 28: Retention of L3 Learners Across Levels ............................................. 96
Table 29: Summary of the Results for Fisher’s Exact Test .................................... 97
Table 30: Quantitative Statistics: Summary of Research Questions, Analyses, and

Results ................................................................................................. 98
Table 31: Motivation for Studying Speciﬁc Languages ..................................... 104
Table 32: Interactions of Foreign Languages Studied ........................................ 115

Table 33: Interactions of Foreign Languages and Portuguese ............................... 119
Table 34: Summary of Standard Deviations of MSUELT Scores ........................... 134
Table 35: Participant Numbers within Groups ................................................. 135

LIST OF FIGURES

Figure 1: Group Divisions According to the MSUELT Scores ................................ 64

Figure 2: Scree Plot for the Exploratory PCA. .. .............................................................. 80

xi

CHAPTER ONE — INTRODUCTION AND REVIEW OF RELATED LITERATURE

Introduction

What are the factors that lead to successful language acquisition? Do those language
learners with a higher language aptitude or more motivation have an advantage in
language learning? Do bilinguals have an advantage in subsequent language acquisition
over monolinguals? What factors encourage students to continue with their language
studies? These are some of the issues that this dissertation addresses.

Current research documents a plethora of advantages to being bilingual. The
beneﬁts include metalinguistic awareness (e. g. Bialystok, 2001a, 2001b, 2006), more
efﬁcient learning strategies (e.g. Nayak et al., 1990; Wilson & Sperber, 2006),
heightened problem solving abilities (e. g. Bialystok & Majumder, 1998), better creative
thinking (e.g. Ricciardelli, 1992b), and less anxiety than monolinguals (Dewaele, 2002).
With regards to motivation and multilingualism, there have been a number of studies that
indicate multilinguals have higher motivation than bilinguals (e. g. Sanz, 2000). It has
also been found that biliteracy is an indication of how effectively the L3 will be learned
(e.g. Da Fontoura & Siegel, 1995; Rosier and F arella, 1976; Sanz, 2008). Concerning
the construct of language aptitude, there have been no speciﬁc measures or comparisons
among the language aptitude of monolinguals and the language aptitude of bilinguals
and/or multilinguals. However, there have been a few studies that have researched

language aptitude and bilingualism (e. g. Sanz, 2000). Sanz (2000), operationalizing

aptitude as verbal intelligence, did not ﬁnd a signiﬁcant difference in language aptitude
between monolingual and bilingual learners.

This dissertation applies the theories from the previous research to a new context.
In the previous studies involving multilingualism, motivation, and language aptitude, the
contexts have been in bilingual settings. The participants in these aforementioned studies
were part of a bilingual society; thus, they had access to the languages in question. This
dissertation seeks to examine if similar results can be found with participants in a
monolingual context—1 28 participants from the Casa de Cultura Britanica ’s English
language program, which is afﬁliated with the Universidade Federal do Ceard in
Fortaleza, Brazil. The participants completed a variety of language tests including the
Portuguese version of the Motivated Strategies for Learning Questionnaire (MSLQ),
Questionario de Estratégias Motivacionais para Aprendizagem (QEMA) (Brown et al.,
2001), the Cognitive Ability for Novelty in Acquisition of Language (Foreign) (CANAL-
F T) (Grigorenko et a1. 2000), the Michigan State University English Language Test
(MSUELT) (Cook, 2001), and a self-rated proﬁciency scale based on the Common
European Framework of Reference (CEFR) (Hasselgreen, 2005). There is also a content
analysis of interview data collected from a subgroup of the multilingual participants.

Using this battery of tests and the interview data, the question of whether those in
monolingual environments pattern the same way as the participants in bilingual or
multilingual environments is addressed. In addition, traditional means of deﬁning what
constitutes a multilingual learner are challenged. This dissertation argues that even a
small amount of language experience cannot be ignored and discusses an innovative way

of perceiving multilingualism—Perceived Positive Language Interaction. This sub-

categorization of multilingualism differentiates between language learners that perceive
positive interactions between foreign languages studies and those language learners who

do not.

Review of Related Literature

The organization of this literature review is threefold; the ﬁrst part is a discussion of the
literature on bilingualism and multilingualism, the second part is a discussion of the
literature on language aptitude, and the ﬁnal part is a discussion of the literature on
motivation.

The ﬁrst section on bilingualism and multilingualism starts with a discussion of
the advantages that have been found to be associated with language learning including
general intelligence, metalinguistic awareness, cognitive ﬂexibility and processing
mechanisms, more efﬁcient learning strategies, active brains with aging, the ability to see
in a different perspective, and less general anxiety. The section continues with a
discussion of additive and subtractive bilingualism and then embarks on a discussion of
the beneﬁts of biliteracy as opposed to bilingualism in the oral form. The section on
bilingualism/multilingualism concludes with a discussion on the proﬁciency level needed
in order to be considered bilingual or multilingual.

The section of the literature review dealing with language aptitude outlines the
main schools of thought within language aptitude, including the idea of language aptitude
as a dynamic construct as opposed to a static one. The main language aptitude tests are

also outlined, along with a detailed description of two of the main tests — the Modern

Language Aptitude Test (MLAT) and the Cognitive Ability for Novelty in Acquisition of
Language (Foreign) Test (CANAL-FT).

The section of the literature review dealing with motivation describes the three
main periods of motivation theory: the social psychological period, the cognitive-situated
period, and the process-oriented period. The constructs involved in the cognitive-
situated period are described in the greatest detail in this literature review since the tool
used in this dissertation was created during this period of motivation. Some of the
concepts that are crucial to this period of motivation are task value, attribution theory,
and self-determination theory. The literature review concludes with a brief discussion of

the gaps in the current literature and how this dissertation attempts to ﬁll these gaps.

Bilingualism and Multilingualism
The concept of being bilingual is complex, and the literatureon bilingualism uses dozens
of terms to describe the various types of bilingualism. These terms range from balanced
bilingual to describe someone whose two languages are roughly the same in terms of
proﬁciency, to incipient bilingual to describe someone who is just starting to learn a new
language (Wei, 2001). The implications of the varying definitions of bilingualism will
be discussed in the following literature review as well as in the experimental section of
this dissertation.

In the earlier days of bilingualism research, it was believed that knowing more
than one language interfered with cognitive development. Therefore, bilingual children
were thought to have intellectual and academic deﬁciencies (e. g. Barke, 193 3).

However, a seminal study by Peal and Lambert (1962) found that bilingual children were

not inherently at a disadvantage when other factors like socioeconomic status were taken
into consideration. In fact, their study was one of the ﬁrst to acknowledge the potential

advantages of being bilingual:

This study has found that bilinguals performed better than monolinguals
on verbal and nonverbal intelligence tests. These results were not
expected because they constitute a clear reversal of previously reported
ﬁndings. . .The picture that emerges of the French-English bilingual in
Montreal is that of a youngster whose wider experiences in two cultures
have given him advantages which a monolingual does not enjoy.
Intellectually his experience with two language systems seems to have left
him with a mental ﬂexibility, superiority in concept formation, and a

more diversiﬁed set of mental abilities (p. 20).

Peal and Lambert continued by saying that there is no question that a bilingual child is
intellectually superior to his or her monolingual counterpart. Since the appearance of this
1962 article, there have been numerous studies illustrating the inherent beneﬁts of
bilingualism. Besides the positive effect of bilingualism on general intelligence (Peal &
Lambert, 1962), there have been numerous studies on the relationship between
bilingualism or multilingualism and metalinguistic awareness (Ben-Zeev 1977;
Bialystok, 1987, 1988, 1991, 2001a, 2001b, 2006; Eviatar & Ibrahim, 2000; Galarnbos &
Goldin—Meadow, 1990;1anco-Worral, 1972; Jessner, 2005; Ricciardelli, 1992a, 1992b;

Thomas; 1988; Yelland et al., 1993), cognitive ﬂexibility and processing mechanisms

(McLaughlin & Nayak, 1989; Nation & McLaughlin, 1986; Nayak et al., 1990), and
more efficient leaming strategies (McLaughlin & Nayak, 1989; Nayak et al., 1990;
Pennington, 1999; Wilson & Sperber , 2006). There have also been studies supporting
the idea that bilingualism helps people’s brains stay young and active in old age
(Bialystok, 2006), that bilinguals can see the world through a different perspective
(Malakoff & Hakuta, 1991), are better at problem solving (Bialystok & Majumder,
1998), and have a more democratic disposition (Pandey, 1991). It has also been posited
that bilingualism has a positive effect on creative thinking (e.g. Ricciardelli, 1992b), and
even that multilinguals are less anxious than their monolingual counterparts (Dewaele,
2002). These studies support the notion that language learning is essential not only for
general education, but also for the development of further cognitive dimensions.

As stated in Sanz (2000), bilingualism has an effect on how subsequent languages
are learned. Metalinguistic awareness has been shown to play a role in successful
language acquisition. Ellis (1994) proposed the weak interface theory, stating that
metalinguistic knowledge acts as an advanced organizer, thus, speeding up the language
acquisition process. Several researchers, (Ben-Zeev 1977; Bialystok, 1987, 1988, 1991,
2001a, 2001b, 2006; Eviatar & Ibrahim, 2000; Galambos & Goldin-Meadow, 1990;
‘Ianco-Worral, 1972; Jessner, 2005; Ricciardelli, 1992a, 1992b; Thomas; 1988; Yelland
et al., 1993) illustrate that bilinguals have a better understanding of metalanguage, which
is the explicit knowledge of language information or the organization of language
systems. This explicit knowledge speeds up the learning process of subsequent
languages, and according to Thomas (1988, 1992) helps bilingual learners to analyze

language in a more structured system. Bialystok (2006) states the following: “It is

plausible that having two different language systems for examination may make
structural patterns more noticeable and hasten the child’s attention to the systematic
features of the language” (p. 579-5 80). This idea is supported by results from Nayak et
al. (1990) who found multilingual participants outperformed monolingual participants in
learning the syntax of an artiﬁcial language. Although the multilingual participants did
not outperform the monolingual participants in all aspects of language learning, Nayak et
al. concluded tha “. . .there is some evidence to suggest that more experienced language
learners show greater plasticity in restructuring their mental representations of the rules
governing linguistic input” (p. 242). In other words, if people have a concept of how
more than one language is structured, they may be cognitively more receptive to
accepting an additional language system into their cognitive repertoire. In addition,
Klein (1995) found that multilingual subjects in New York outperformed monolinguals
on a test of lexical learning and were able to use their newly acquired lexical items in
new contexts in a more effective way than the monolinguals. The multilingual
participants were also able to more quickly acquire the parameters of unfamiliar syntactic
structures, in this case preposition/verb collocations and preposition stranding. Bialystok
and Majumder (1998) found that these concepts of bilingualism could extend to non-
verbal advantages. Their study showed that balanced bilinguals outperformed
unbalanced bilinguals and monolinguals on non-linguistic tasks, supporting the idea that
people with a certain level of bilingualism have a more developed sense of selective
attention and control of processing.

There have also been several empirical studies (McLaughlin & Nayak, 1989;

Nation & McLaughlin, 1986; Nayak et al., 1990) that have suggested that when people

learn an L2, they learn the process of language learning; thus, people with previous
language experience have more resources to devote to the form of the new language.
According to VanPatten (1996), controlled processes take more space in a leamer’s
working memory than do automatic processes. For example, L2 speakers of Spanish
have already learned how to conjugate verbs in the subjunctive, so they know the process
of taking a root (the core meaning of the verb) and adding a speciﬁc morpheme (the part
that carries the tense and aspect). Although this process of verb conjugation begins as a
controlled process (explicit/declarative knowledge) requiring intense concentration to
ﬁgure out which morpheme goes where, the more the learner performs the conjugations,
the more automated (implicit/procedural knowledge) this process becomes When
bilinguals with an L2 of Spanish study Portuguese, the process of verb conjugation has
already been introduced, so the process of automatization may be more rapid. They can
focus on learning the intricacies of the verb mood, rather than the concept and pathways
of the verb conjugation itself. According to this theory, because the process of
conjugation becomes automatic more quickly, the learners can process the input more
effectively.

Along with the idea of the language learning process, there is evidence that
bilinguals and multilinguals take advantage of more efﬁcient learning strategies than
monolinguals (McLaughlin & Nayak, 1989; Nayak et al., 1990; Pennington, 1999;
Wilson & Sperber, 2006). Nayak et al. (1990) found that multilingual subjects
outperformed monolingual subjects on syntax tasks because of their ability to search for
underlying rules. They also found a signiﬁcant difference between multilinguals and

monolinguals in the use of mnemonic devices for memorization. Similarly, McLaughlin

and Nayak (1989) found multilinguals had a preference for mnemonic devices for
memorization and that multilinguals used a wider variety of strategies than did
monolinguals to help in rule discovery exercises; multilinguals were more willing to
change the language learning strategy if it was ineffective.

Both Cenoz and Valencia (1994) with their research in the Basque country, and
Sanz (2000, 2008) with her research in Catalonia cite evidence for the beneﬁts of
bilingualism with regard to third language (L3) acquisition. Cenoz and Valencia (1994)
studied 320 17-19 year olds, all of whom were in their last year of high school. Of the
participant pool 48% were monolingual Spanish, and the other 52% were Spanish-
Basque bilinguals. The goal of this study was to see if bilingualism had an effect on the
acquisition of English. The conclusion was clear, “bilingualism has a positive mediating
effect on third language learning.” These beneﬁts were not affected by “cognitive, socio-
structural, social psychological and educational variables” (p. 204). Similarly, Sanz
(2000) studied 201 English students in Catalonia (Northeastern Spain), 77 of whom were
monolingual Spanish and 124 of whom were Spanish-Catalan bilinguals. Like in the
Cenoz and Valencia study, these participants were also learning English. In Sanz’s study,
factors such as intelligence, age, motivation, attitudes towards British and American
language and culture, and number of languages spoken were all operationalized
variables. By performing a hierarchical multiple regression analysis, Sanz showed that
bilingualism results in higher scores on English proﬁciency tests. Sanz (2000) concluded
that “...immersion programs in the minority language, whether in the Basque Country or

in Catalonia, produce more efﬁcient L3 learners” (p. 34).

While oral bilingualism might have a positive mediating effect with regards to
subsequent language acquisition, literacy can have an even greater effect. The
advantages of biliteracy with regards to third language acquisition have been discussed
by a number of researchers (Ben-Zeev, 1977; Cenoz & Valencia, 1994; Cummins 1976;
Da Fontoura & Siegel, 1995; Diaz, 1985; Galarnbos & Goldin-Meadow, 1990;
Keshavarz & Astaneh. 2004; Ricciardelli, 1992a, 1992b; Sanz, 2008,; Thomas, 1988;
Valencia & Cenoz, 1992; Yelland et al., 1993). The concept of language literacy as
opposed to oral bilingualism is especially crucial when dealing with language policy and
minority languages. One of the ﬁrst studies to be done in this area was Rosier and
Farella (1976) with their work with the literacy rates of Navajo children. In the 19605,
the students at Rock Point Boarding school, a school whose students were of Navajo
origin, had grim test scores in English. The students’ L1 was Navajo, but they had no
literacy skills in this language and also had poor English language skills. The
implementation of an ESL program helped raise the test scores somewhat, but the
underlying problem of illiteracy in the L1 remained. Through funding by Title I, literacy
skills in Navajo were introduced, and it was found that Navajo children instructed in
Navajo performed better on English literacy tests than Navajo children who were
illiterate in Navajo. This trend of the importance of literacy for subsequent language
acquisition is not unique to these Navajo children; it is crucial for everyone wishing to
learn language. With few exceptions, the literature supports the idea that bilingualism
and biliteracy at all levels has a positive effect on acquiring subsequent languages (e. g.

Keshavarz & Astaneh. 2004; 2008; Swain et al., 1990; Valencia & Cenoz, 1992).

10

In addition to the example of the Navajo children, there are examples in Canada
of better success in language learning when L1 literacy is achieved (Swain, Lapkin,
Rowen, & Hart, 1990; Da Fontoura & Siegel, 1995). Many Canadian schools offer
bilingual education in French and English, so children of immigrants often speak one
language at home and receive instruction in two others at school. Though parents and
administrators previously thought that instruction in the ﬁrst language (L1) of immigrant
students would in fact impede the acquisition of the L2 and L3 (in this case, English and
French), two empirical studies (Swain, Lapkin, Rowen, & Hart, 1990; Da Fontoura &
Siegel, 1995) showed that literacy in the L1 did not impede, and in fact, was beneﬁcial to
learning subsequent languages.

In the Swain et al. (1990) study, immigrant children with a variety of language
backgrounds were tested to see if Ll literacy had an effect on the acquisition of French
and English. The results were clear: “. . . literacy in one’s mother tongue enhances third
language learning” and “. . . bilingual education programmes that promote ﬁrst language
literacy have an overall positive effect on the learning of other languages” (p. 78).
Similar results were obtained in Da Fontoura and Siegel’s (1995) study of bilingual
Portuguese-English Canadian children and the effects of bilingualism on reading: “. . .
bilingualism is clearly not an impediment to the development of reading, syntactic and
memory skills” (p. 149). In fact, bilingual children who are reading-disabled scored
signiﬁcantly higher on pseudoword reading and spelling tasks than did their monolingual
counterparts, supporting the idea that bilingualism aids in language acquisition on all

levels, not just with gifted learners.

ll

In Sanz’s 2008 study, she studies many factors affecting L3 acquisition such as,
effects of age of onset in L3 acquisition, order of acquisition of the majority and minority
languages, and the degree of proﬁciency in the two languages. The students were
bilingual to varying levels in Spanish and Catalan and were all enrolled in English
language courses. For the ﬁrst research question of the individual differences found to
predict L3 acquisition, a signiﬁcant effect was found for language use — language used
when speaking to the mother, to friends and while shopping. In addition, the balance of
proﬁciency correlated signiﬁcantly with written achievement. A regression analysis
showed signiﬁcant effects for both exposure and motivation, and when these two
variables were eliminated from the analysis, there was a signiﬁcant effect for attitudes
towards British English. No signiﬁcant results were found with age of L2 onset or order
of acquisition of the majority or minority languages. As for the question of balanced L1
and L2 and the relationship of L3 acquisition, it was found that literacy in both the L1
and the L2 aided in the more efficient acquisition of structures in the L3. The
implications of this study are that actual language use (using the language with ﬁ'iends,
while shopping, etc.) had a signiﬁcant effect. Thus, inquiries about the use of the
languages in question, as opposed to simply proﬁciency measures, are crucial in contexts
that involve minority languages. Another crucial point to this study is that biliteracy, as
opposed to oral bilingualism, is essential to L3 acquisition.

Shifting contexts, bilingualism acquired in a formal context also has an effect on
learning the vocabulary of subsequent languages, as shown in Keshavarz and Astaneh
(2004). There were three groups of participants — two bilingual groups (30 Turkish-

Persian and 30 Armenian-Persian) and one Persian monolingual group — and the

12

researchers were comparing the acquisition of English vocabulary. In this study, the
bilingual participants outperformed the monolingual participants with English
vocabulary acquisition. According to the authors, “The results showed that bilingualism
has a more positive effect on third language’s vocabulary achievement when the ﬁrst two ,
languages are taught formally” (p. 300), providing additional support for the beneﬁts of
biliteracy.

One possible reason for these strong ﬁndings could be that if children learn to
read and write in their L1, a language they already know, they are able to better
understand the need and purpose of reading and writing (Hudelson, 1987) and already
understand the “discourse structure of stories and [. . .] the symbolic system used to
encode that language” (Bialystok, 2006, p. 589). In addition, if the writing systems of the
two languages are similar, the children can transfer what they already know about form-
meaning connections (the semantic value of the letters or words) in their L1 to their L2.
However, reading skills not only transfer across languages, but across writing systems as
well as evidenced by Geva, Wade-Woolley, and Shany (1997) who investigated the
transfer of the reading skills by bilingual Hebrew-English children. In their study, 66 L1
English six-year olds were learning to read in Hebrew, and although the process in
learning to read in the L2 (Hebrew) was not identical, the initial L1 English literacy was
thought to help in gaining the L2 Hebrew literacy, although there was no control group
for comparison. In further support of the importance of biliteracy, Sanz (2000) argued
that “biliteracy rather than ‘oral bilingualism’ to be the factor determining cognitive
abilities” (p. 38), thus facilitating language acquisition, and according to Cenoz (2001),

“linguistic distance plays an important role in cross-linguistic transfer” (p. 16), although

13

Elli

va

lli

sometimes the perceived possibility of transfer can be more inﬂuential than the actual
linguistic distance (Kellerman, 1977, 1978, 1979, 1983). Numerous researchers have
documented that for aid in learning a new language, multilingual learners often rely on
languages they know that have linguistic similarities to the new language at hand (Cenoz,
2001; Ecke, 2001; Sikogukira, 1993). However, it is important to remember that even
unrelated languages can have an effect on subsequent language acquisition, as
emphasized by Ringbom’s (2003) presentation entitled: “If you know Finnish as an L2,
there will be no major problem learning Swahili.” According to Ringbom, although
Swahili is a Bantu language and Finnish is a Finno-Ugric language, open-minded
learners can ﬁnd a relationship between the two.

With the seemingly boundless support indicating that bilingualism contributes
positively to subsequent language acquisition, what could account for the studies that
demonstrate no relationship between bilingualism and subsequent language acquisition
(e. g. Wagner, Spratt, and Ezzaki (1989), Sanders and Meijers (1995), and van Gelderen,
Schoonen, de Glopper, Hulstijn., Snellings, Sirnis, and Stevenson (2003)? Though it is
possible that the materials used or the design of the study could play a role, the most
salient feature accounting for the lack of beneﬁts of bilingualism is the idea of additive
and subtractive bilingualism. According to the deﬁnitions found in Wei (2001), an
additive bilingual is “someone whose two languages combine in a complementary and
enriching fashion” and a subtractive bilingual is “someone whose second language is
acquired at the expense of the aptitudes already acquired in the ﬁrst language” (pp. 6-7).
The aforementioned studies indicating positive aspects of bilingualism were primarily

carried out in additive bilingualism contexts, thus contributing to positive results. As

14

Cenoz (2003) elucidates, “Most studies on general aspects of proﬁciency indicate that
bilingualism has a positive effect on third language acquisition when L3 acquisition takes
place in additive contexts and bilinguals have acquired literacy skills in both their
languages” (p. 83).

Three illustrative examples of this subtractive bilingualism phenomenon are
Wagner et al. (1989), Sanders and Meijers (1995), and van Gelderen et al. (2003).
Wagner et al. (1989) takes place in Morocco where three languages exist simultaneously:
Berber, Arabic, and French. This study was part of the Morocco Literacy Project, a ﬁve
year longitudinal study of Moroccan elementary school children to see if Ll Berber
literacy made the acquisition of L2 Arabic literacy more efficient. The question
proposed was the following: Was instruction in the minority and socioeconomically
disadvantaged language necessary to the speakers of this language? The results show
that though the L1 Arabic participants had advantages during the initial stages of the
study, the L1 Berber students eventually caught up to the L1 Arabic participants in terms
of Arabic literacy. The conclusion of this study was that learning literacy in the L2 as
opposed to the L1 posed no disadvantage to the students.

The second two studies by Sanders and Meijers (1995) and van Gelderen et al.
(2003) both take place in Holland with elementary and middle school aged children. The
participant groups comprised of L1 Dutch speakers and L2 Dutch speakers with and an
L1 of Turkish, Arabic, or a language from Surinam. Sanders and Meijers (1995) had 46
bilinguals (15 Arabic speakers and 31 Turkish speakers) who were matched with 46 L1
Dutch students to compare their acquisition of English over a two-year period. The

results indicated that there was no difference in English acquisition by the monolingual

15

and bilingual elementary school age children. Van Gelderen et al. (2003) had a sample
of 397 grade 8 Dutch children, 281 of whom were Ll Dutch and 57 of whom were L2
Dutch and came from Turkey, Morocco, or Surinam. The task was to investigate if one
of the groups, monolingual or bilingual, had an advantage in English reading
comprehension. The results indicated that there was not a difference in English reading
comprehension between the monolingual and bilingual groups, though the authors stated
that one reason for this could have been socioeconomic status and the fact that the
bilingual group as a whole had lower scores in reading comprehension to begin with.
The common link in these three studies is the L1 language status of the participants. In
all cases, the participants had learned their L2s at the expense of the Lls because of the
esteemed status of their L2s in their countries of residence. The ﬁndings that contradict
the studies in which beneﬁts of bilingualism were found can thus partially be attributed
to the situation of subtractive bilingualism. 1

There are other concepts that are crucial to the discussion regarding bilingualism
and multilingualism: The advantage or disadvantage of knowing multiple foreign
languages, the proﬁciency of a foreign language needed for that language to make a
difference in subsequent language acquisition, and the idea of perceived language
distance. Evidence has been found for advantages of knowing multiple languages on
learning another language, but sometimes if the languages are too similar, lexical
confusion can occur. Gibson and Hufﬁsen (2003) investigated the relationship between
knowing multiple foreign languages and the ability to translate a text in an unknown
foreign language. The participants were residing in Germany at the time of the study and

were enrolled either in EFL (English as a foreign language) or GSL (German as a second

16

language) courses. All of the participants had knowledge of both English and German
and knew up to four languages. The results were that the more languages the participants
knew, the better they could translate Swedish text. Gibson and Hufﬁsen concluded the
following: “There were several robust indications that not only are learners about to
exploit their foreign language knowledge when approaching a ‘new’ foreign language,
but that this exploitation is facilitated by the sheer number of previous languages” (p.
102). The study also indicated that the more languages the person knew, the better they
were at making use of their metalinguistic knowledge. Kemp (2001) also found similar
results with his participants in Spain; the more languages the participants knew, the
easier it was for them to learn Basque.

In the literature there is a question of what level of proﬁciency is needed in order
for a non-native language to have an inﬂuence on the subsequent learning of another
language. Most researchers monitor the learners’ linguistic background only if the learner
has a high proﬁciency in another language (e.g., Clyne, 1997; Clyne & Cassia, 1999;
Nayak et al. 1990; Ringbom, 1987; Williams & Hammarberg, 1998), although Ringbom
(1987) also found inﬂuence at the lexical level even with a very low level of proﬁciency.
Cummins (1976, 1979) also proposed the “Threshold hypothesis,” indicating that a
certain level in the non-native language had to be reached to avoid cognitive
disadvantages, especially when dealing with language learning in children

There are also researchers who have found other types of effects for even very
low levels of a non-native language (De Angelis 1999, 2005a, 2005b; Selinker &
Baumgartner-Cohen, 1995). De Angelis (2005a) showed a clear example of the effect of

very little non-native language exposure. Participants were either L1 English or L1

17

Spanish, and some from each L1 group had prior knowledge of French. The third person
masculine singular subject pronoun in French is il, in Italian, it is lui, and in Spanish it is
el. Participants who had studied French and those who had not were compared in their
performance of overt subject insertion in Italian. In both Spanish and Italian, overt
subjects are optional according to the context, whereas in French, overt subjects are
essential. Those who had studied French had not done so extensively; however, this
minimal knowledge of French signiﬁcantly affected the overt subject insertion in Italian.
Those with even a minimal amount of French were making more overt subject insertions
than those with no French experience. Though there is not a consensus about the
proﬁciency level of a non-native language needed to see an effect in subsequent language
acquisition, it is clear that researchers cannot ignore those participants who have even

6

low levels of additional non-native languages. ‘. . .as little as one or two years of formal
instruction in a non-native language can affect the acquisition of another non-native
language to a signiﬁcant level” (De Angelis, 2007z6).

The concept of perceived language distance as proposed by Kellerman (1977,
1978, 1983) states that language transferability depends on the learner’s perception of the
language distance of the languages in question. If the learner perceives the languages to
be related, then they are more likely to transfer form and meaning; if the learner does not
perceive a closeness of the languages, then the learner will be less likely to transfer form
and meaning. Kellerman’s (1987) study of Dutch L1 - German L2 learners led her to
propose the notion of perceived language distance and found that “. . .the less

representative of the prototypical meaning a usage of a given form is, the lower its

transferability” (Kellerman, 1987, p. 65). This concept will be important in the

18

methodology section with the discussion of the deﬁnitions of the multilingual learners

for this study.

Language Aptitude

In addition to bilingualism as a possible factor in successful subsequent language
acquisition, language aptitude could also play a role. In order to discuss this idea of the
relationship of language aptitude to language acquisition, another, more basic question
must ﬁrst be answered: When talking about measuring language aptitude, what construct
is being measured? One answer is Skehan’s (1990) explanation that aptitude consists of
two separate factors: a person’s innate ability for processing language combined with
environmental factors, such as parental language background and literacy.

One of the ﬁrst scholars to study language aptitude was Carroll. Carroll (1981) in
his seminal work reviewing 25 years of language aptitude research carefully
distinguished language aptitude from general intelligence or “G” (as measured by IQ
tests). Though the two constructs are related, high intelligence does not denote a high
aptitude and vice versa. He was also careful to note the difference between language
aptitude and achievement: “Aptitude as a concept corresponds to the notion that. . .the
individual may be thought of as possessing some current state of capability of
leaming. . .Achievement, on the other hand, corresponds to the notion that the individual
can have acquired certain speciﬁed capabilities of actual performance” (p. 84). In other
words, a language learner’s language aptitude could be average, but because of other
factors (such as motivation), the learners could have a high level of achievement. On the

other hand, a high level of aptitude does not guarantee a high level of achievement (see

19

also Gardner, 1985, and Skehan 1986). Another of Carroll’s contributions to the study of

language aptitude was the idea of aptitude as a stable feature, remaining unchanged

regardless of the language learner’s added experiences.

Carroll’s (1981) model of language aptitude is comprised of four constituent

abilities: phonetic coding ability, grammatical sensitivity, rote learning ability, and

inductive language learning ability. See Table 1 for descriptors of these constituents.

Table 1

Carroll’s 4-Constituent Model for Language Aptitude

 

Constituent

Descriptor

 

Phonetic coding ability

Grammatical sensitivity

Rote learning ability

Inductive language learning

ability

The ability of identifying a number of sounds and
then being able to link them with the associated
symbol. This skill involves being able to store the
sounds into long-term memory for later use.

An awareness of grammatical relationships
presented in the input, even those never seen
before. The learner should be able to recognize
patterns and functions of the various parts of the
input.

The ability to memorize, store, and recall large
amounts of foreign language material.

The ability to make inferences about language rules
aﬁer being presented with a sample of that

language.

 

20

More recently, the concept of language aptitude as a stable feature has been
challenged by Robert Stemberg who believes language aptitude to be dynamic in nature.
Stemberg (2002) argues that a language learner’s language aptitude can ﬂuctuate
depending on the experience (or lack thereof) of that particular language learner. He
describes his triarchic theory of human intelligence and insists on the fact that the way

that language aptitude is viewed should be changed:

First, we need to test for creative and practical language-acquisition
abilities, not just for memory and analytical ones. Second, we should
recognize the need for a more reﬁned analysis that does not just give a
general language-aptitude score, but sub-scores that suggest appropriate
forms of instruction. Finally, we should consider the use of dynamic
testing, whereby testing and instruction occur at the same time so that the

examiner can assess the ability to learn in real time. (p. 42).

Several tests exist to measure the construct of language aptitude, perhaps the most
familiar being the Modern Language Aptitude Test (MLAT, Caroll & Sapon, 1958). The
Pimsleur Language Aptitude Battery (PLAB, Pimsleur, 1966), the Army Language
Aptitude Test (ALAT, Home, 1971), the Defense Language Aptitude Battery (DLAB,
Petersen & Al-Haik, 1976), and the VORD (V ORD - the artiﬁcial word used for word in

the test) (Parry & Child, 1990) are several well-known tests. More recently, a new

21

aptitude test called the Cognitive Ability for Novelty in Acquisition of Language
(Foreign) Test (CAN AL-FT) was developed based on Stemberg’s (2002) aforementioned
triarchic theory of human intelligence (Grigorenko et al., 2000). In order to illustrate two
of the competing theories of language aptitude, below are detailed descriptions of both
the MLAT and the CAN AL-FT.

The MLAT was ﬁrst developed in 1959 by Carroll and Sapon. With the
development of the communicative language approach to teaching, the MLAT has
become less popular because many language educators and researchers feel that the
subtests of the MLAT measure aptitude as it relates to a more audiolingual pedagogical
ﬁ'amework (Stansﬁeld & Reed, 2004). Upon closer examination, however, has been
found that regardless of the teaching style of the classroom, the MLAT can still predict
success in a foreign language classroom (Ehrrnan, 1998). There have been many other
studies that validate the use of the MLAT to measure language aptitude, such as Carroll
(1962), Gardner and Lambert (1965), Gajar (1987), and Parry and Child (1990).

Language aptitude is operationalized as comprising four skills, according to
Carroll (1981): phonetic coding ability, grammatical sensitivity, rote learning ability,
and inductive language learning ability. Phonetic coding deals with the ability to
recognize sounds, remember them, and associate those sounds with the words, whereas
grammatical sensitivity is the ability to recognize syntactic functions of words in a
sentence. Rote learning ability is the measure of how effectively a person can learn
words and retain them, and inductive language learning ability is how well someone can
infer rules from data sets. The MLAT consists of ﬁve parts which sample these four

underlying abilities: number learning, phonetic script, spelling clues, words in sentences,

22

and paired associates (D6rnyei, 2005). Table 2 below provides an explanation of the ﬁve
parts of the MLAT with sample questions for each part. (Note that all of the sample
questions below are taken with permission from Dr. Charles Stansﬁeld (President of
SLTI) from the SLTI website: http://www.2lti.com/htm/Test_cb_mlat.htm#13)

Table 2

Description of the MLAT

 

Part of the test Description

 

Part I: The participants hear audio information about numbers in a new
Number language. For example, ba is one, baba is two, dee is three. They are
Learning then asked to listen to numbers in the new language and write the

English equivalent. The numbers become more complicated, such as tu
is twenty and ti is thirty. T u-ba is twenty-one because tu is twenty and
ba is one. After this brief training, the participants hear a series of
numbers, such as ti-ba, ti-dee, etc. and have about ﬁve seconds to write

down the English equivalent to the numbers.

Part II: The participants learn phonetic script for English. They are presented
Phonetic with four possible transcriptions for each question, and the recording
Script pronounces each of the syllables during a training period. Here are

some examples: 1. bot but bok buk and 2. bok buk bov bof. After the
training period, the participants have to listen to the speaker on the

recording who says only one of the words previously presented. The
participants have to choose the written syllable that corresponds with

the word on the recording.

23

Table 2 con’t
Part III:
Spelling

Clues

Part IV :
Words in

Sentences

The participants are presented with a word that is not spelled correctly.
They then have to pick an answer from the four choices presented that
most closely corresponds with the word presented. For example, the
participant would be presented with a word such as, kloz and would
have to choose between A. attire, B. nearby, C. stick, and D. giant.
The participants must identify the part of the sentence that has the same
function as a previously indicated part.
For example:
1. Jill fell down A_N_D_ Jack came tumbling after.

Nil/.1 you may wait o_ut there, g you may come back g Friday if you

wish.

A B C D E

24

Table 2 con’t
Part V: Paired The participants must memorize Mayan-English word pairs - 2 minutes
Associates for 24 words. They do a practice exercise and then answer the

questions. A sample questions would be like the following:

Vocabulary

Maya -- English
c?on gun
bat ax

pal son

They would then be asked to identify the meaning of the words given,

such as bat - A. animal, B. stick, C. jump, D. ax, or E. stone.

 

Though there have been some doubts about using the MLAT as a measure of
language aptitude, these doubts do not affect part four, words in sentences, as DeKeyser

(2000) clearly states:

Carroll and Sapon’s (1959) Modern Language Aptitude Test is usually
considered the best verbal aptitude test in terms of its predictive value
for L2 learning. Some of the minor technical problems discovered

over the years do not affect the “Words in Sentences” part (Carroll,

25

1990), which is speciﬁcally aimed at measuring grammatical
sensitivity and therefore should be the best predictor of grammar

learning (p. 509).

Because of this, part IV, Words in sentences, has been used in previous studies as the
sole measure of language aptitude (e. g. Otto, 1996) to the extent that when Otto (1996)
created the Hungarian version of the MLAT, he chose only to develop a words in
sentences section to measure language aptitude.

The Cognitive Ability for Novelty in Acquisition of Language (Foreign) Test
(CANAL-FT) is based on a second language aptitude theory that encompasses ﬁve
language acquisition processes: selective encoding, accidental encoding, selective
comparison, selective transfer, and selective combination, all of which take place at the
lexical, morphological, semantic, and syntactic levels. Table 3 below describes the ﬁve

knowledge acquisition processes.

26

Table 3

The 5 Knowledge Acquisition Processes for the CANAL Theory

 

 

Process Description

1. Selective encoding Decide the relative relevance within the
stream of incoming data

2. Accidental encoding Encoding of secondary information to

understand the context — important for
language learning because it refers to the
knowledge absorbed outside of the
learner’s primary focus

3. Selective comparison The ability to determine the usefulness of
prior knowledge for the task at hand

4. Selective transfer The application of learned or inferred rules
to new situations

5. Selective combination The synthesis of selective and accidental
encoding along with old structures to form

a new knowledge base.

 

In addition, this theory posits that language leaming takes place with a combination of
visual and oral input and that uptake (language that is successfully acquired) manifests
itself in both immediate and delayed recall settings (Grigorenko, Stemberg, & Ehrman,

2000), thus the test is designed with these aspects in mind. The test is different in that it

27

deals with an aspect of language learning that the other aptitude tests have not — that
language learners must cope with novelty and ambiguity (e. g. Ehrman, 1996; Ehrman &
Oxford, 1995). This test simulates a language learning situation in which the participants
are learning a new (artiﬁcial) language, Ursulu. Since the language is artiﬁcial, the
learners have no previous experience with it, which controls the factor of incidental
experience with the language. Ursulu is introduced gradually through a series of written
and oral stimuli, so that by the end of the test, the learners should be able to handle a
short text in the new language. There are ﬁve sections to the test: Learning meanings of
neologisms from context; Understanding the meaning of passages; Continuous paired-
associate learning; Sentential inference; and Learning language rules. All of the sections
are comprised of both immediate and delayed recall questions with the exception of the
last section, Learning language rules. Because this is the last section of the test, only
immediate recall is possible. Since this is the language aptitude test used in this study,
the table with the descriptions and sample questions is provided in the material
description chapter of this dissertation.

Between the time of the development of the MLAT and the CANAL-F T,
relatively little was done in the way of language aptitude research in comparison to other
individual learner variables such as motivation and learning strategies (Skehan, 2002). In
his 2002 article “Theorising and updating aptitude,” Skehan provides an overview of the
former aptitude research and sheds new light on the relationship of language aptitude to
SLA. In this article, he correlates between the SLA processing stages and the aptitude

components in an attempt to bring language aptitude research, “. . .to come in from the

28

rather isolated position it has occupied in applied linguistics for the last half-century” (p.

93).

Motivation

Another factor central to that of language learning is that of motivation. The research in
this area is proliﬁc; it is helpful to separate the trends into three main categories. The
social psychological period of motivation research was from about 1959 to 1990 with
Gardner and his colleagues in Canada at the forefront. The cognitive-situated period
popular during the 19903 was based largely on cognitive theories from the ﬁeld of
educational psychology, and the most recent trend is the process-oriented period from
2000 to the present, which is based on Domyei’s concept of the dynamic nature of
motivation (Domyei, 2005).

In the social psychological period, one of the ﬁrst motivational constructs
developed was Gardner’s (1985) theory of integrative motivation. The language learner
who has high integrative motivation identiﬁes himself or herself with the language
learning community and is also able to be positive and accurate when evaluating the
language learning situation. Within the context of the socio-educational model of
motivation, integrative motivation can be summarized as follows: 1) integrative
motivation combines attitude, goal-orientation, and motivational variables; 2) integrative
motivation assumes that the learner is striving for native-like proﬁciency; 3) positive
attitudes to the language culture are necessary for ultimate attainment (Gardner, 2001).
Because his research takes place in Canada, which is a largely bilingual society,

Gardner’s model of integrative motivation assumes that there is contact with native

29

speakers of the target language (in this case, French). Therefore, it is not as generalizable
to situations in which the students might never have contact with native speakers of the
target language.

The cognitive-situated period is largely inﬂuenced by Crookes and Schmidt
(1991), though many other researchers contributed as well (e.g. Pintrich, 1989). The
main concepts of motivation research were to show how thinking about tasks and
personal capabilities are crucial to the concept of motivation and to approach motivation
as it affects learning situations, such as classrooms (DOmyei, 2005). This period of .
motivation is based on the belief that studen , “. . .need to acquire not only factual
knowledge and basic skills but also ‘critical thinking’ skills which will enable them to
evaluate new ideas and concepts” and that motivation and cognitive ﬁmctions cannot be
separated (Pintrich, 1989, p. 118). The cognitive-situated period comes from social
cognitive theory, which states that both learning and performance are greatly affected by
the motivational process (Schunk, 1995). There are three processes inherent in the social
cognitive theory — self observation, self judgment, and self-reaction (Bandura, 1986;
Kanfer & Gaelick, 1986; Schunk, 1994; Schunk & Zimmerman, 2003; Zimmerman,
1990). Self observation involves behavioral self-reﬂection on the part of the learner,
self-judgment is a comparison of the leaner’s present abilities with his ultimate goal, and
self-reactions are the learner’s responses to self-judgments. If the learner believes that he
or she is making sufficient progress towards the goal, self-efﬁcacy will be improved.
(For a complete description of these three processes, see Schunk, Pintrich, & Meece,

2008)

30

One of the instruments created during this period was the Motivated Strategies
for Learning Questionnaire (MSLQ) (Pintrich, 1989), which currently is considered
“...the best known instrument in this area in educational psycholOgy” (Domyei, 2005, p.
178). The MSLQ was developed by Pintrich and his colleagues at the University of
Michigan and has undergone a series of rigorous internal reliability measures (e. g.
McKeachie, Pintrich, & Lin, 1985; Pintrich, 1986a,b). In addition, the MSLQ has been
translated into Chinese (Rao & Sachs, 1999) and Portuguese (Brown, Cunha, Frota, &
Ferreira, 2001) to be used in other educational settings, adding to the breadth of its
applications. The entire MSLQ has 85 questions: 35 questions for the motivation section
and 50 questions for the learning strategies section. In the motivation section, there are
three major categories of questions: value components, expectancy components, and
affective components. The value components are made up of questions dealing with
three sub-sections: intrinsic motivation (K=4), extrinsic motivation (K=4), and task value
(K=6). The expectancy components are made up of questions addressing control beliefs
(K=8), self-efficacy (K=5), and expectancy for success (K=3). The affective components
section includes questions about text anxiety (K=5). In the learning strategies section,
there are two main themes for questions: cognitive strategies (K=31) and resource
management strategies (K=19). Though both the motivation section and the learning
strategies section make up the MSLQ, it is also possible to use the two sections
separately as independent measures of the construct of motivation and of learning
strategies.

There are several studies that have come out of the cognitive-situated period of

motivational research (e. g. Donitsa-Schmidt, Inbar, & Shohamy, 2004; Domyei &

31

Csizéi
Nikola
how p
pm

m

can 5
panic
their 2

merit

I
‘71 5 .~
Malls:

metix

Csizér, 2002; Inbar, Donitsa-Schmidt, & Shohamy, 2001; Nikolov, 2001; Pintrich, 1989).
Nikolov’s (2001) qualitative study of unsuccessful language learners in Hungary showed
how perceptions of language classes can affect successful language learning. The
participants were 94 young adults of varying educational levels who all started to study a
foreign language between the ages of six and nine. The results showed that negative
experiences with the earliest language classes greatly affected their language learning
process even as adults. Many of the participants felt threatened by the grammar drills
and forced memorization, described by one participant as a “death camp” (p. 156), and
even subsequent positive experiences could not override these negative experiences. The
participants had negative attitudes towards the language learning process as a result of
their attitudes towards their language learning context of the classroom, and their
motivation decreased over the years.

Also in Hungary, Domyei and Csize’r (2002) investigated how the sociocultural
changes that took place in Hungary in the 1990s affected language learners’ attitudes and
motivation. The two groups of participants totaling 8,593 ﬁlled out questionnaires on
two occasions, in 1993 and 1999, thus allowing both the dynamic change in motivation
and the effect of the sociocultural changes on the learners to be studied simultaneously.
Though the ﬁndings in the enormous study were many, one of the most important
ﬁndings relating to the current topic was that of the relationship of active engagement in
language learning and motivation: “It was found that at a national level, an active
engagement in the study of an L2 exerts a positive inﬂuence on the learners’ L2 attitudes

and motivation” (p. 455). Those students who were actively pursuing a language through

32

a classroom context tended to have higher motivation and better attitudes towards
language learning.

Pintrich (1989) examines a different population of learners — those who were
enrolled in various courses in three different universities in Michigan. He did not
speciﬁcally study how motivation related to language learning, but instead, how it related
to learning in general. The 224 participants were enrolled in English, Biology, and
Psychology courses and were given the Motivated Strategies for Learning Questionnaire
(MSLQ) at the beginning of the term. A series of correlations were performed with the
data to ﬁnd if the participants’ motivation (the ﬁrst part of the MSLQ) had a relationship
with the cognitive and metacognitive learning strategies (the second part of the MSLQ).
The students’ course performance was also examined with relationship to the
motivational and cognitive aspects measured by the MSLQ. The motivational
components positively correlated with the cognitive components and students with high
task value for the course itself also performed better on the assessment tasks within the
course. In sum, the results showed that, “. . .task value and expectancy management
components can combine with cognitive, metacognitive, and effort management
components to inﬂuence student performance” (p. 153). In other words, perceptions of
the task necessity combined with learning strategies greatly affect the outcome of
learning in all classrooms, not just language classrooms.

Donitsa-Schmidt, Inbar, and Shohamy (2004) and Inbar, Donitsa-Schmidt, and
Shohamy (2001) examined how the perceived quality of Arabic instruction in classrooms
in Israel affected the students motivation to continue studying the language and how the

study of Arabic affected attitudes towards native speakers of Arabic and the Arabic

33

culture. The Donitsa-Schmidt, Inbar, and Shohamy (2004) article illustrated the effect of
changes on the classroom Arabic teaching environment in Israel on students’ motivation
to study the language. Through self-reporting questionnaires to both elementary school
students (N =692) and parents (N =3 62), the ﬁndings indicated that those students who
studied spoken Arabic rather than Modern Standard Arabic had a better attitude towards
the people and the culture and were also more motivated to learn the language. The
ﬁndings also reinforced the idea of parental inﬂuence in language attitudes of students.
Inbar, Donitsa-Schmidt, and Shohamy (2001) studied the motivation of 1,690 seventh
grade students with regards to their motivation to study Arabic, a language “. . .associated
with and representing the political conﬂict between Jews and Arabs” (p. 307). This
paper addressed whether the study of Arabic in schools could assuage some of the
conﬂicting feelings of the language and the speakers of the language as well as motivate
the learners to study the language in the future. Of the groups, those students who
studied Arabic in school and those who did not, the group who studied Arabic had
overall, “. . .higher motivations for instrumental, cultural, political, and parental reasons
and higher motivation to study Arabic in the ﬁrture” (p. 307). Within this group, one of
the highest predictors for future Arabic study was the degree of satisfaction with the
current Arabic course.

Many research projects have been carried out with successful language learners,
but Nikolov (2001) took a different perspective in his paper that looked at unsuccessful
language learners. In Nikolov’s study, only 12% of the participants claimed to speak a
foreign language while 88% claimed to speak nothing but their L1.1 Through interview

questions in this qualitative study, it was found that the majority of the participants were

34

de

de

la

CC

demotivated by negative classroom experiences ﬁnding, “. . .oral and written assessment
practicing with an insistence on accuracy and memorization threatening, the focus on
grammar drills and rote learning of texts boring and useless” (p. 167). The ﬁndings of
this study are important insofar as they illustrate the potentially devastating effects of
classroom language instruction that does not meet students’ expectations.

Within the cognitive-situated period of motivation research, there are several
important concepts including the following: task value, attribution theory, and self-
determination theory. According to Pintrich (1989), there are three components within

this model of motivation theory (see the article for a full description):

(1) students’ beliefs about the importance and value of the task (value
components), (2) students’ beliefs about their ability or skill to perform
the task (expectancy components), and (3) students’ feelings about

themselves of their emotional reactions to the task (affective components)

(p. 120).

Task value is one of the aspects of the value component set, and has been shown to
greatly affect motivational outcomes. Task value is composed of the person’s perception
of the importance of the task, the intrinsic interest in the task, and the usefulness of the
task for future goals (Pintrich, 1989).

Attribution theory is the idea that humans can make decisions rationally and
consciously (Weiner, 1992) and makes two general assumptions. The ﬁrst assumption is

that “. . .humans are motivated by a goal of understanding and mastering the environment

35

and themselves” (Schunk et al., 2008, p. 81). The second assumption is that, “. . . people
are nai've scientists, trying to understand their environments and, in particular, trying to
understand the causal determinants of their own behaviors as well as the behaviors of
others” (Schunk et al., 2008, p. 81). The main idea of this theory is that all humans
attribute success and failure to both internal and external factors. This approach starts
with some completed event (such as success or failure of an exam) and the reaction that
occurs at the end of this event (Weiner, 2000). These reactions will determine ﬁrture
motivation and learning behaviors of language students.

Self-determination theory is the idea that Deci and colleges developed to
distinguish self-determination ﬁ'om will (Deci, 1980; Deci & Moller, 2005; Deci & Ryan,
1985, 1991; Reeve, Deci, & Ryan, 2004; Ryan & Deci, 2000). Deci (1980) deﬁnes will
as “the capacity of the human organism to choose how to satisfy its needs” and self-
determination as “the process of utilizing one’s will” (p. 26). In other words, self-
detennination is a sort of control over a person’s intrinsic motivation, and there are many
environmental factors that can decrease a person’s self-deterrnination (see Schunk et al.,
2008, for a complete description). The impact of this theory in SLA research has been on
comparing intrinsic and extrinsic motivational orientations. Noels and colleagues have
done several studies using this theory of motivation. (e. g. Goldberg & Noels, 2006;
Noels, 2005, 2003; Noels, Pelletier, Clement, & Vallerand, 2000) Both attribution theory
and self-determination theory are crucial for understanding people’s motivation within
the framework of the cognitive situation period of motivation.

The process oriented period of motivation introduced the concept that motivation

is dynamic and should be measured accordingly. That is, one can argue that motivation

36

has three stages: preactional (motivation is initiated); actional (motivation is
maintained); and postactional (past events are reﬂected upon - motivation retrospection)
(Domyei, 2005). A few empirical studies have been carried out in this area (e. g. Lim,
2002; D6myei, Csizér, & Németh, 2006), although not as many as in the other

motivational periods.

Conclusion

In the literature review, three main topics were addressed: the beneﬁts of bilingualism,
language aptitude, and motivation. This dissertation draws on research from all of these
areas of individual differences. The inherent advantages of multilingualism will be
discussed in relation to the data collected for this project. In addition, arguments will be
made for varying deﬁnitions of multilingual learners. The motivation component of this
study draws on the cognitive-situated period of motivation and uses a tool developed in
this time period—the QEMA, the Portuguese version of the MSLQ. As for language
aptitude, the idea that aptitude is dynamic is central to this project; the CANAL-FT is
used to measure the language aptitude of the participants.

As can be seen from the above literature review, there is copious research
outlining the beneﬁts of bilingualism, although there are few studies that compare the
language aptitude proﬁles of monolingual and bilingual learners or those of bilingual and
multilingual learners (c.f. Sanz, 2008). In addition to the other aspects of this project,
this dissertation compares bilingual and multilingual learners in terms of language

aptitude in an attempt to ﬁll the gap in the current literature on this topic.

37

Research Questions

In this study, a variety of factors will be examined to study the multilingual/bilingual
dichotomy and the effect that it has on language learning. Variables in this study are
additional language experience, English proﬁciency, motivation, and language aptitude.
The results of the data involving the accuracy of self-rated language proﬁciency, English
proﬁciency, motivation, and language aptitude will be discussed in detail. The research

questions and hypotheses are as follows:

RQl: Do bilingual learners have the same motivational proﬁles as multilingual learners?
H1: Bilingual and multilingual learners have distinct motivational proﬁles.
RQ2: Which group of participants has higher language aptitude, bilingual or multilingual
learners?
H2: Multilingual learners have a higher language aptitude than bilingual language
learners.
RQ3: Which group of participants has higher English language proﬁciency as measured
by the MSUELT, bilingual or multilingual learners?
H3: Multilingual participants have higher English language proﬁciency scores as
measured by the MSUELT
RQ4: Which group has a higher retention rate in the progression of the English courses,
bilingual or multilingual learners?
H4: Multilingual learners have a higher retention rate in the English

language courses.

38

C
o.._‘ " ‘— ‘ ‘
‘ h a. If: .
‘ r
~ Q .‘ . ‘ l i .- . ,a_
—— , ‘
a. Euo'g I- '-I

 

..—

3:-epgsag

 

_=.....:" Etfu‘ﬁl’ I L- 39"
u "

wan".

O

u — -0 ‘.-‘u ’ "
--- o .A- V‘- -I. ..
. .

‘ o¢,~
454-.

--
- .. . sna,

; ‘vv
.z .
- -- ~-
2 ....._- v
. -
.- .- n—
” . ‘ '
.-A -'
inn I- .-
a
" a
‘O -' " ’
an. 1 0-1d—J’ . *-
‘ .

 

 

CHAPTER TWO: RESEARCH DESIGN AND METHODOLOGY

Context for the Study

Participants in this study are students enrolled in the Casa de Cultura Brita‘nica (CCB
affiliated with the English Language Program of the Universidade Federal do Ceara
(UF C - the Federal University of Ceara, a northern province of Brazil) in Fortaleza,
Brazil. Besides the English program, the university also has language programs in
several other languages: Casa de Cultura Alema' (German), Casa de Cultura F rancesa
(French), Casa de Cultura Hispa‘nica (Spanish), Casa de Cultura Italiana (Italian), and
Casa de Cultura Portuguesa (Portuguese). In total, these language programs have 5,600
students, with the English program the largest with 2,400 students.

The CCB was founded in 1964 and was the third of the language programs or
“houses” to be formed under the department called the Casas de Cultura Estrangeira
(the houses of foreign culture), the German language program being the ﬁrst. The Casas
as a whole function like a department of UFC; however, the students taking these
language classes do not take them for credit towards a degree. There is a different
department within the university called Lettras, which is similar to a College of Arts and
Letters in the American university system that offers degree programs in various
languages. The students taking classes at the Casas are doing so for personal enrichment
or to supplement their university studies. Not all of the students taking language classes
are enrolled in a different department at the university; a number of them are working

people who are taking a language course on the side. That being said, getting into a

39

language program at the Casas is quite competitive. A government subsidized program,
the cost is minimal (the equivalent of about $30.00 a semester for one language), but for
students to get into the program they must take a placement test based on an estimate of
their level. There are not enough slots for everyone who wants to take a class, so the
enrollment is limited to those who score the highest on the individual level placement

tests.

Participants

Participants in this study are all enrolled in the curso basico, the ﬁrst set of English
courses at CCB consisting of seven levels. Out of the 350 original volunteers, 128
completed the questionnaires and the aptitude test, and 79 completed the entire project.
Thirty of the 79 multilingual participants who completed the entire project were also
interviewed after all of the tests had been completed. Of the 128 participants, there were
64 L2 learners and 64 L3 learners, based on the ﬁrst categorization of multilingualism
provided in this dissertation. The L2s of the multilingual participants varied, though the
most common L2 was English. For the 79 participants who completed the entire project,
there were 39 L2 learners and 40 L3 learners. Similar to the larger sample, the most
common L2 for the multilingual participants was English, whereas the L3s varied
considerably. See Table 4 for a summary of the linguistic backgrounds of the
participants in the study. In this table, note that the information is provided solely for the
multilingual participants. For the bilingual participants in this study, all of their Lls

were Portuguese and their L2s English.

40

12

C0

Table 4

Languages of the Multilingual Participants Divided by Subgroups

 

 

 

L2 L3 L4

All Subset All SubSet All Subset

participants participants participants

(N=128) =79) (N=128) (N=79) (N=128) (N=79)
English 41 29 17 9 2 2
Esperanto 1 1 1 1 0 0
French 4 2 16 12 6 5
German 2 1 6 6 2 1
Greek 0 0 1 0 0 0
Italian 1 1 2 2 O 0
Japanese 0 0 2 2 3 1
Latin 0 0 2 2 1 1
Spanish 1 1 6 12 7 3 3
unknown 4 0 5 0 0 0
Total 64 40 64 40 l7 l3

 

Note: The subset of 79 are those participants who completed all parts of the study

Materials

Variables in this study are additional language experience, English proﬁciency,
motivation, and language aptitude. All of the materials used were translated into
Portuguese, with the exception of the MSUELT, the English level test. The participants

completed the background questionnaire, the motivation questionnaires, the self-

41

assessment questionnaires, the language aptitude test, and the optional interviews in
Portuguese. The ﬁrst part of the data collection was the background questionnaire,
which collected basic information about the participants as well as information about

their language learning experience (see Appendix A).

Multilingualism

Additional language experience, or multilingualism, was measured by the Reading and
Listening sections of the Common European Framework of Reference (CEFR). There
are 29 listening questions and 46 reading questions. All the questions are in a yes/no
format of “yes, I can do the skill described or no, I can’t do the skill described. Both the
listening and the reading questions test the levels of A1 , A2, B1, B2, C1, and C2. In

Table 5 below, there are sample questions from each of the levels tested.

42

Table 5

 

 

Sample Questions from the CEF R
Listening Place a check
mark here if you
can do this.
A1 1 I can understand when someone speaks very slowly to me

B1

B2

C1

C2

11

17

23

29

and articulates carefully, with long pauses for me to
assimilate meaning.

I can understand what is said clearly, slowly, and directly
to me in simple everyday conversation; it is possible to
make me understand, if the speaker can take the trouble.

I can follow clearly articulated speech directed at me in
everyday conversation, though I sometimes have to ask
for repetition of particular words and phrases.

I can understand in detail what is said to me in standard
spoken language even in a noisy environment.

I can follow extended speech even when it is not clearly
structured and when relationships are only implied and not
signaled explicitly.

I have no difficulty understanding any kind of spoken
language, whether live or broadcast, even when delivered
at fast native speed, provided I have time to get familiar

with the accent.

43

 

Table 5 con’t

Reading Place a check
mark here if you

can do this.

 

A1

B1

BZ

C1

C2

17

25

33

41

I can understand information about people (place of
residence, age, etc.) in newspapers.

I can identify important information in news summaries or
simple newspaper articles in which numbers and names
play and important roles and which are clearly structured
and illustrated.

I can understand the main points in short newspaper
articles about current and familiar topics.

I can rapidly grasp the content of the signiﬁcance of news,
articles, and reports on topics connected with my interests
or my job, and decide if a closer reading is worthwhile.

I can understand fairly long demanding texts and
summarize them orally.

I can recognize plays on words and appreciate texts whose

real meaning is not explicit (for example irony, satire).

The concept of previous language exposure is crucial in determining successful

subsequent language acquisition. One of the problems in studying multilingual learners is

44

the deﬁnition of what qualiﬁes as a multilingual learner (Wei, 2001). Once a deﬁnition
has been decided upon, the next complication involves classifying the participants in a
study according to the deﬁnition. Although an independent measure of proﬁciency of the
languages studied by the multilingual learner is usually considered the most accurate
measure of the language skills, oftentimes administering these languages tests is
logistically impossible. Is there an alternate way to assess language ability? Though the
idea is still somewhat controversial and debated in applied linguistics research, using
self-assessment scores is a way to assess the language ability of multilingual participants.
Although self-rating proﬁciency scores have a variety of uses and advantages,
“. . .there has been surprisingly little discussion of the value of self-assessment to more
expensive and logistically viable approaches to proﬁciency and achievement assessment,
particularly in the area of second and foreign language testing” (Ross, 1998, p. 1). There
have, however, been some studies discussing the accuracy of self-assessment tools.
Peirce, Swain, and Hart (1990) found that self-assessment only correlates weakly with
actual language ability in Canadian French immersion programs. Similarly, Brantrneier
(2006) found that the self-assessment of reading ability by advanced learners of Spanish
entering into the university as freshmen does not correlate strongly with actual abilities in
reading. Contrastingly, Brantmeier (2005) found advanced learners of Spanish to assess
quite accurately their own L2 reading ability. According to Blanche and Merino (1989),
although self—assessment scores can vary depending on the tools used as well as other
factors, the relationship between self assessment scores and actual ability is “good to
quite good” (p. 315). In Blanche and Merino’s article, the authors provide a meta-

analysis of 11 articles investigating the reliability of self-assessment in reading, listening,

45

and speaking. The participants’ countries of origin and Llls differed, as well as the
languages that they were studying, though the most common of the target languages was
English as a second language (ESL). Across the articles in the meta-analysis, the most
correlations between the self-assessment scores and the actual abilities of the participants
were found in the skill of reading, followed by listening. Though there were also
correlations found in the area of speaking, the overall results showed that students are
less accurate in assessing their own speaking abilities. Thus, students prove to be more
accurate in self-assessing the two receptive skills of reading and listening over the
productive skill of speaking. In the article, there was no mention of the other productive
skill, writing. There are many other articles that ﬁnd a positive correlation between self-
assessment scores and placement tests and ability. (e. g. Birckbichler et al., 1993;
Brantmeier, 2005; Hargan, 1994; Heilenman, 1991; von Elek, 1987). In addition, Harris
( 1997) supports the use of self-assessment in formal language situations by showing that
self-assessment in language classes helps students become more active learners and gives
them motivation when they see their progress over the course of the language class.
Blanche and Merino (1989) suggest two reasons that self-assessments can vary.
One reason is that the participants may interpret the questions differently. The second
reason is that if a language learner has never experienced a situation mentioned in the
self-assessment questionnaire, it might be difficult for the learner to self-estimate his or
her ability to do it. One common tool used for self-assessment is the Common European
Framework of Reference (CEF R) (Schneider & North, 2000). An example question for
intermediate proﬁciency in the CEFR is “I can understand in detail what is said to me in

standard spoken language even in a noisy environment.” If language learners have never

46

been in a noisy environment trying to understand the language in question, how do those
language learners know if they would be able to understand the foreign language in

question?

English Language Proﬁciency

The independent measure of English language proﬁciency is the Michigan State
University English Language Test (MSUELT). This test consists of two parts: listening
and reading. The listening section comes ﬁrst with 53 questions and 25 minutes to
complete. There are four main parts to this listening section. The ﬁrst part is a short oral
presentation on which the students must take notes as in a lecture class. The second part
consists of three short listening passages followed by a list of questions. The third part
has conversations with two or more people followed by questions. The fourth part has
questions about the short oral presentation presented in part one. The students can use
their notes that they took in part one to complete part four. The reading section is
second, with 53 questions and a time limit of 45 minutes. Part one, questions 1-25, is
reading comprehension. This part consists of passages followed by multiple choice
questions based on the content of the reading passages. Part two, questions 26-50, is a
multiple choice vocabulary test. There is a sentence with a blank and the participant

must choose the word that best ﬁts into the blank from four choices.

Motivation
Motivation was measured by the ﬁrst part of the Questionario de Estratégias

Motivacionais para Aprendizagem (QEMA), a Portuguese translation of the Motivated

47

Strategies of Learning Questionnaire (MSLQ) (Pintrich, 1989). The QEMA consists of
35 items with a Likert scale of seven. Please see Table 6 below for a sample question

ﬁom the English version of the questionnaire.

Table 6

Sample Question from the MSLQ

 

1 2 3 4 5 6 7

 

Absolutely false <— —> Deﬁnitely true

 

1. In aclass like this, I prefer course material that really 1 2 3 4 5 6 7

challenges me so I can learn new things.

 

Language Aptitude

Language aptitude is measured by the Cognitive Ability for Novelty in Acquisition of
Language (Foreign) Test (CANAL-FT) (Grigorenko et al., 2000), which has 83 items.
The test is a mix of oral and written stimuli, increasing in difﬁcultly as the test
progresses. Aptitude is measured by how well participants can learn a new imaginary
language, Ursulu, by the end of the test. The test contains ﬁve types of questions. See
Table 7 below for a description and sample questions from each of the ﬁve types of

questions asked. (see Grigorenko et al., 2000 for a full description with examples).

48

Table 7

Description and Sample Questions from the CANAL-F T

 

Section

Description and sample questions

 

1. Learning
meanings of
neologisms from
context
(immediate and

delayed recall)

This section consists of oral and written paragraph stimuli with
unknown words in Ursulu inserted and was designed to assess the
ﬁve cognitive processes mentioned above. The oral paragraphs are
generally shorter than the written ones; the oral passages are read
once before the questions are seen and once after the questions are
seen. As the test progresses, the paragraphs have an increasing
number of unknown words in Ursqu for the participants to
understand. The understanding of the new vocabulary is tested by
two multiple choice questions following the paragraph input.
Example Item (immediate recall) (partial text):

Rising tuition costs and increasingly large loans aren’t the only
ﬁnancial issues facing mukulu nafe-de, the latest threat to Yuve- Y uve
ya-pama-de pocketbooks comes from mandatory twok-dc. One laka
will require entering freshmen fru hujuk a mukulu-speciﬁed laptop
twok at a cost of $3,000.

Questions:

Fru hujuk most likely means: (a) to arrange; (b) having; (0)
carrying; (d) to purchase; (e) to rent.

Mukulu most likely means: (a) schools; (b) student; (c) parent; (d)

universities; (e) college.

49

Table 7 con’t

2. Understanding
the meaning of
passages
(immediate and

delayed recall)

The presentation of the unknown lexical items of Ursulu within the
paragraphs is similar to those in section one in terms of density of
the unknown words. The difference is that in this section, the
participant’s general comprehension of the passage is measured, as
opposed to the understanding of speciﬁc lexical items as in section
one. The four skills measured in the questions asked are
comprehension of gist, comprehension of main ideas,
comprehension of details, and inference and application.

Example Item (immediate recall) (partial text):

The wealthy hunting femo-de of late glacial Europe might have
maintained or even enriched culture, or unta-u erto to stagrrate ik
decline: Yuve could hardly have advanced erto to a higher form of
civilization, for the environment neunta-u erto. But Y ave-Yuve
future cutta-u not left in Yuve-Yuve own sima-de.

Question: (delayed recall):

The author of the passage about the hunting society apparently
believes that levels of civilization are determined by: (a) economic
luck; (b) a balance of solar energy; (c) the ambitions of the people;

((1) a piece of magic; (e) climatic conditions.

50

Table

pairs
leam
(imrr

dela)

Table 7 con’t

3. Continuous
paired-associate
learning
(immediate and

delayed recall)

This section of the test is comprised of 60 word pairs, half of which
are presented orally and half of which are presented visually. The
task of learning the words differs from rote memorization in that the
participants are intended to understand the underlying morphology
of the Ursula words presented, thus aiding in memorization. The
participants are tested at irregular intervals on the new words.

Example pairs: (immediate recall):

kiSS = lutik

maki smelano = ﬂoweret
to oppose = fru prostoto
threerish = two

to luxuriate = fru shikta
unteriapremu = fairytale
to learn = fruumbrad
juk-de = ﬁngers
Yellow = hukoi

pj ze_min-de = workers
Questions:

In Ursulu, ‘ﬂoweret’ most likely means (a) maki smelano; (b)
ummake; (c) lutik; (d) pjze_min; (e) makijuk.
ﬁu umbrad most likely means: (a) to eat; (b) to go; (c) to learn, (d)

to kiss; (e) to dream

51

Table 7 con’t
4. Sentential
inference
(immediate and

delayed recall)

5. Learning
language rules

(immediate

For this section, the participants either see or hear three to ﬁve
sentences in Ursulu and the translations in the L1. The questions
consist of a new sentence either in the L1 or Ursulu, and the
participants must choose which of the ﬁve choices provided most
accurately represents the new sentences by using the information
learned by the previously translated sentences.

Example item: (immediate recall):

Panlin-u Sumu T wah chuck means I handed a stick to him.
Panlin-u Y at Two dozz means He handed an umbrella to me.
Panilcos-u Yut T wa ﬂexta means He handed a piece of paper to
me.

Panleh-u Sumu Twah chuchu means I handed a rope to him.
Question:

The sentence: Panilcos-u Sumu T wah otikum most likely means:
(a) He handed a rod to me;

(b) I handed a cord to him;

(0) I handed a postcard to him;

(d) I handed a waterhose to him;

(e) I handed a tree-branch to her.

This is the ﬁnal section of the test and is the culmination of the
participant’s understanding of the Ursulu language. In addition to

the language knowledge already gained in the course of the test, this

52

Table 7 con’t

recall only) section provides some vocabulary, some grammar, and some
examples of the inner workings of Ursulu. The twelve questions in
this section are designed to measure the participant’s lexical,
semantic, morphological, and semantic understanding of Ursulu.
Example item: (immediate recall):
In Ursulu, ya-bum baqlo means ‘the chief’s mule,’, ya being the
possessive and ya-bum the modiﬁer of the noun baqlo ‘mule.’

Match the corresponding pairs:

ya-ﬁtama pokka corresponds to
preumma chicca-de corresponds to
ya-xori gazza corresponds to
prebrutama tepla-de corresponds to
ya-ayama xrosyo corresponds to
preuntam rutuma corresponds to

(a) monkey’s smile; (b) alligator gloves; (c) sheep wool; (d) cat’s

tail; (e) gigantic tiger; (f) wife’s book.

Interviews
The intent of the interviews was to gain deeper insight into the language backgrounds of
the multilingual participants; thus, only the multilinguals were asked to volunteer for the

interviews, and 29 of the multilingual participants did so. The participants were given

53

the choice of which language they preferred for the interviews, and all of the participants

chose to conduct the interview in Portuguese. For the interviews, a semi-structured

interview format was used with a preliminary list of questions. The times for the

interviews ranged from 10 minutes to about one hour, depending on how much each

individual participant had to say in response to the questions. All of the interviews were

recorded on tape and then transcribed. Table 8 has a list of the basic questions used in

the interviews.

Table 8

List of the Questions Asked in the Oral Interviews.

 

l.

2.

Do you enjoy studying languages? Why or why not?

What languages have you studied?

Why do/did you want to learn English?

Why do/did you want to learn the other languages you have previously
studied/are studying?

How do you plan to use the languages you have studied/are studying in the
future?

You have studied more than one language. Do you think that your prior
language experience affected the process of learning subsequent languages?
Please give detailed examples.

Do you think that studying other languages has affected the way you understand

Portuguese? Please give detailed examples.

54

Table 8 con’t

8. Please describe your English classes, past and present. You might want to
include class activities, teaching style, pace of the course, exams, etc.

9. What is your approach to language learning? Besides going to class, what do
you do to improve your target language?

10. What effect do you think the teacher has on the language learning experience?
What about the other students in the class? Please provide speciﬁc examples.

11. Do you have any other comments?

 

In the previous pages of material description, each data collection tool was explained in
detail. See Table 9 below for a summary of all of the data collection tools used in this

dissertation.

55

Table 9

Summary of Data Collection Tools

 

 

Tool Construct it measures Scoring
CEF R can-do statements measure of self assessed - listening: 29
level of English and any - reading: 46
other language studied - “yes” or “no” answers
(“yes,” I can do the skill

described or “no,” I can’t

 

 

do the skill described)
MSUELT independent measure of - listening: 53 questions
English language - reading: 50 questions
proﬁciency - multiple choice
QEMA measure of motivation - 35 questions

- 7-point Likert scale

 

 

CANAL-FT measure of language - 83 questions
aptitude
Interview questions to provide more details - taped interviews

about the language learning - content analysis

experience

 

Note: all of the tools were in Portuguese with the exception of the MSUELT.

56

Procedure
Part 1 — Recruitment of the participants

The entire data collection process took place during a period of three months in
the spring of 2008. During the ﬁrst two weeks of the project (weeks one and two),
participants were recruited for the project. To recruit students, 57 English classes were
visited: ﬁve semester one classes, eight semester two classes, ten semester three classes,
nine semester four classes, seven semester ﬁve classes, nine semester six classes, and
nine semester seven classes. During the class visits, 350 students initially volunteered,
though as explained above, that number diminished over time. Those students who
volunteered in the class were given a packet of materials including the consent form, the
background questionnaire, the QEMA, and a CEF R for English and any other foreign
language that they had studied. The volunteers were asked to sign up for a time to take
the CAN AL-FT and were also asked to bring back the questionnaires when they came to

take the language aptitude test.

Part 2 — (Language aptitude test — CANAL-F T)
During the next three weeks (weeks three through six), the participants were
given the Cognitive Ability for Novelty in Acquisition of Language (Foreign) Test
(CANAL-FT). Upon ﬁnishing the aptitude test, they were asked to sign up for a time to

take the MSUELT.

57

Part3-

llSl'E
This I:
studen
level t
comp:
Sever

delve

Patti

panic

thorc

acqu

aliec

ind

Part 3 - (Independent measure of English language proﬁciency —— MSUELT)

After a one week spring break (week seven), the testing resumed with the
MSUELT (weeks eight through ten), as the independent measure of English proﬁciency.
This test has high measures of reliability as discussed in Cook (2001). Because the
students were currently taking English language classes, it was crucial that the English
level test be administered in a relatively short period of time so that the scores would be
comparable. The MSUELT was the last required part of the data collection process.
Several of the multilingual participants were asked to schedule an additional interview to

delve more deeply into their language acquisition processes.

Part 4 — (Interview questions)

During the last two weeks (weeks eleven and twelve), 29 of the multilingual
participants were interviewed. These interview questions were designed to gather a
thorough understanding of how the participants view their process of language
acquisition, including questions about how their additional language proﬁciency was

affecting their acquisition of English.

The previous procedural description, the process of data collection was explained

in detail. See Table 10 below for a summary of the entire procedure for this dissertation.

58

Table 10

Summary of Data Collection Timeline

 

Week(s) Description of data collection

 

1-2 Participant recruitment and questionnaire distribution

(background questionnaire, CEFR, QEMA)

3-6 CAN AL-FT
7 Break

8-10 MSUELT

1 1-12 Interviews

 

59

CHAPTER THREE: QUANTITATIVE RESULTS

Analysis: Levels of English

Before answering the speciﬁc research questions, some background information must be
addressed. The ﬁrst concept to be addressed is the idea of grouping the participants
according to the MSUELT scores and the semesters in which they are enrolled for their
English classes at CCB. The question that needs to be addressed is whether the
participants can be divided into subgroups when discussing the level of English.

To answer this question, correlations were done comparing the MSUELT total scores, as
well as the sub-skills, with the participants’ English semesters. As shown by the above
correlation matrix, the total MSUELT scores, as well as the sub-skills correlate
signiﬁcantly with the English semester of the participants; however, the correlation is not
strong enough to suggest that one measure can substitute for another (Bachman 1990,

2004). See Table 11 for a summary of the correlations.

Table 11

Correlations between MSUELT Scores and English Semester

 

 

MSUELT: MSUELT: MSUELT: English

Total Listening Reading Semester
MSUELT: T 1 .897“ .936M .544“
MSUELT: L 1 .693“ .439**
MSUELT: R 1 .532";
English Sem. 1

 

*"' Correlation is signiﬁcant at the .001 level.

60

Perhaps the reason for this is because the scores for some of the sequential semesters are
very close to each other, and there are subgroups within the English semesters whose
scores are close enough to be considered as one group. See Table 12 for the descriptive

statistics by semester of the overall MSUELT scores.

Table 12

Descriptive Statistics for the Overall MSUELT Scores

 

 

 

English Semester Number of Participants Mean Standard Deviation
1 11 46.36 13.13

2 8 46.62 7.76

3 16 54.19 14.25

4 8 54.63 15.23

5 15 66.87 15.93

6 10 68.60 12.43

7 11 68.36 9.87

TOTAL 79 58.58 15.62

 

To address the topic of proximity in sequential semesters, a one-way AN OVA with
Tukey post-hoe tests were performed with the participants’ overall MSUELT scores and
English semesters. The ANOVA was signiﬁcant F(6, 72) = 5.94, p < .0001, so the post-
hoc tests were performed to see where the signiﬁcance was exactly. The results of the

Tukey post-hoe test indicated that semester 1 signiﬁcantly differed from semester 5 (p =

61

.005.

.006,

of 0

.004, MD = -20.50, 95% CI upper = -36.52, 95% CI lower = -4.49 ), semester 6 (p =
.005, Nﬂ) = -22.23, 95% CI upper = -39.86, 95% CI lower = -4.61), and semester 7 (p =
.006, MD = -22.00, 95% CI upper = -39.20, 95% CI lower = -4.80). Semester 2
signiﬁcantly differed from semester 5 (p = .014, MD = -20.24, 95% CI upper = -37.90,
95% CI lower = -2.58), semester 6 (p = .014, MD = -2l.98, 95% CI upper = -41.11, 95%
CI lower = -2.83), as well as from semester 7 (p = .013, MD = -21.74, 95% CI upper =
-40.48, 95% CI lower = -2.99). Semesters 3 and 4 did not signiﬁcantly differ from any

of the other semesters. See Table 13 below for a summary of the post-hoe results.

62

Table 13

Signiﬁcance of Tukey Post-Hoc between Semesters for the MSUELT

 

 

Semester Semester Semester Semester Semester Semester Semester

1 2 3 4 5 6 7
Semester] 1.00 1.00 .743 .832 .004" .005" .006"
Semester 2 1.00 .844 .891 .014“ .014" .013”
Semester 3 1.00 1.00 .126 .116 .108
Semester 4 1.00 .362 .300 .296
Semester 5 1.00 1.00 1.00
Semester 6 1.00 1.00
Semester 7 1.00

 

** Correlation is signiﬁcant at the .01 level.

As can be seen from the signiﬁcance levels of the Tukey post-hoe tests, the English

semesters can be clearly divided into three groups: semesters 1-2, semesters 3-4, and

semesters 5-7. The MSUELT scores for the participants from semesters 1 and 2 differ

signiﬁcantly from those of the participants in semesters 5, 6, and 7. However, for the

group in the middle, semesters 3 and 4, their scores differ signiﬁcantly from neither the

group below, semesters 1 and 2, nor ﬁ'om the group above, semesters 5, 6, and 7. This

result is also illustrated by Figure 1 below.

63

Figure 1
Group Divisions According to the MSUELT Scores

 

701

Mean of MSUELT Score

 

 

 

English Semester

 

 

 

These groups will henceforth be named groups one, two, and three. See Table 14 for a

description of these groups.

Table 14

Descriptions of groups based on the MSUELT scores

 

Group 1 Semesters 1 and 2
Group 2 Semesters 3 and 4
Group 3 Semesters 5, 6, and 7

 

64

This merging of participants is important when discussing the research questions
involving English proﬁciency. It would not be fair to judge the English proﬁciency of a
ﬁrst semester participant against that of a seventh semester participant: instead, within
group comparisons will be made using the three aforementioned groups, making the

semester of study the independent measure of proﬁciency.

Analysis: The operationalization of a multilingual learner

Since the concept of multilingualism is central to this study, the tools and criteria used to
deﬁne a multilingual learner for the purpose of this study need to be addressed. The ﬁrst
question is as follows: Are the CEF R self-rating scores an accurate way to assess
language proﬁciency? Although it is generally believed that independent measures of
language proﬁciency are preferred over self-assessment scores (see literature review for a
discussion of this issue), there are times when administering such tests are impossible.
For some languages, these proﬁciency tests are not easily available, and if they are, the
results they produce might not be comparable. For this reason, the reliability of self-
assessment measures becomes an important question. Is it possible to substitute self-
assessment measures for independent measures of language proﬁciency? To address this
topic for the purpose of this study, correlations between the participants’ MSUELT
scores and their English CEFR scores were carried out. Not only were the overall scores
compared, but the sub-skill scores of reading and listening were also compared. See

Table 15 for a smnmary of these results.

65

Table 15

Correlations between the MSUELT and CEF R Scores

 

MSUELT CEFR: MSUELT: MSUELT: CEFR: CEFR:

 

Total Total Listening Reading Listening Reading
MSUELT:T 1 .585“ .897M .936" .505“ .588"
CEFR: T 1 .485* * .569" .918“ .976"
MSUELT: L 1 .693" .449M .472"
MSUELT:R 1 .471 ** .584“ *
CEFR: L 1 .810"
CEFR: R ' 1

 

*" Correlation is signiﬁcant at the .001 level.

As can be seen from the above table, all of the correlations were Signiﬁcant; however, as
discussed in the previous section, for testing purposes, a correlation of 0.9 or higher is
needed for one measure to be able to replace the other measure (Bachman 1990, 2004).
Therefore, although the correlations were signiﬁcant, the self-assessment measure (the
CEF R) cannot be used as a substitution for an independent measure of language
proﬁciency for this group of participants.

This leads to the next item at hand, which is the criteria that were used to deﬁne a
multilingual learner for this study. There were two operational categorizations of a
multilingual learner for this study; those participants with any previous language
experience and a subcategory of those with previous language experience, including only
those participants who perceived a positive language interaction among previous foreign

languages studied. For all of the statistical analyses that follow, the tests will be
66

performed separately for both categorizations. Categorization 1 is any previous language
experience and Categorization 2 is Perceived Positive Language Interaction (PPLI).

For the ﬁrst categorization, the learners were separated into groups based on the
information given on the background questionnaire. If the participant indicated that he or
she had ever studied a language other than English, regardless of the length of study, then
the learner was classiﬁed as a multilingual participant according to the ﬁrst
categorization. The second categorization of a multilingual participant is a bit more
complicated. The idea stems ﬁ'om Kellerman’s (1977, 1978, 1979, 1983) work on
perceived language distance (see the literature review for a full discussion of this issue).
To be classiﬁed as a multilingual learner for the second categorization, “Perceived
Positive Language Interaction,” the participant had to perceive a positive language
interaction between the non-native languages studied, as measured by the following
question on the background questionnaire: “If you have studied other languages in the
past, do you think that this has helped or hindered your ability to learn subsequent
languages? Please provide speciﬁc examples where appropriate.” These answers were
then classiﬁed into groups based on whether or not a positive interaction was indicated.
The initial classiﬁcation was second rated, and the second rater agreed 100% with the
initial rating. The following is an example of an answer that qualiﬁed the participant as

an L3 learner in the Perceived Positive Language Interaction category:

67

Example 1

Acho que ajudou, pois quando vocé tem cantata com outros idiomas, diferentes do seu,
ﬁca maisfdcil assimilar a gramcitica e aprender a pronucia.

I think that it helps because when you have contact with other languages that are
different than yours, it is easier to assimilate the grammar and learn the pronunciation.

(Participant 52)

This participant acknowledged the fact that learning a language other than the L1 opens
the door to a grammar and pronunciation that is not present in the ﬁrst language; thus the
learner perceived a positive interaction between the foreign languages studied. The
following is an example of an answer that qualiﬁed the participant as an L2 learner in the

perceived positive interaction category:

Example 2

N60 acredito que o estudo do outro idioma tenha influenciado de alguma maneira na
aprendizagem do inglés.

I don’t think that studying another language inﬂuenced me when I learned English in any

way. (Participant 20).

This participant did not recognize the interaction of foreign languages studied; thus was
classiﬁed as an L2 learner for the second categorization of Perceived Positive Language

Interaction.

68

It is important to clarify a point here. The two categorizations mentioned above
are potential categorizations of L2 and L3 learners. The ﬁrst way categorized
participants in the traditional way of L2, meaning someone who has studied one
language other than the L1, and L3 meaning someone who has studied two or more
languages other than the L1. The second categorization of Perceived Positive Language
Interaction does not follow traditional categorization of L2 and L3 learners. The L2
learners according to this categorization are those who are truly L2 learners in the
traditional sense as well as some L3 learners who do not perceive a positive language
interaction between previous foreign languages studied. The L3 learners according to
this categorization are those L3 learners who perceive a positive interaction between
foreign languages studied. To avoid confusion in the discussion of the research
questions, the groups will henceforth be deﬁned as the following: the ﬁrst categorization
will use the traditional labels of L2 and L3 when talking about the groups of learners,
while the second categorization will take on new labels. The L2 group will be referred to
as NPPLI (No Positive Perceived Language Interaction) and the L3 group will be
referred to as PPLI (Perceived Positive Language Interaction). It is mportant to note that
the second categorization is not a deﬁnition of multilingualism per se, but instead is a
new way of conceptualizing the factors involved in being multilingual. This
categorization is a sub-set of the participants who have studied multiple languages. The
following table is a summary of the two categorizations of multilingualism, as well as

how many participants were classiﬁed into each category.

69

Table 16

Categorizations and Numbers of Multilingual Participants

 

 

Categorization 1 Categorization 2

Any previous language experience Perceived Positive Language Interaction
L2: N=39 NPPLI: N=58

L3: N=40 PPLI: N=21

 

Analysis: Research Questions

RQI : Do bilingual learners have the same motivational proﬁle as multilingual learners?
Before answering this research question, an analysis of the Portuguese version of the
MSLQ, the QEMA must be discussed. The ﬁrst aspect for discussion concerns the
factors that emerge from the QEMA and how these factors compare to the MSLQ.

This discussion will start with a description of the original English version of the
Motivated Strategies for Learning Questionnaire (MSLQ). The MSLQ and all
subsequent translations are divided into two main parts: motivation scales and learning
strategies scales. For this study, only the motivation scales were used; henceforth, only
this part will be discussed. The motivation section of the MSLQ has 31 questions. The
underlying construct consists of six factors: three value components, two expectancy
components, and one affective component (see Pintrich et al., 1991 for a full
description). The value components consist of intrinsic motivation, extrinsic motivation

and task value. The expectancy components are control of learning beliefs and self-

70

efﬁcacy for learning and performance; the affective component is test anxiety. See Table

17 below for the complete categorization of the factors for the MSLQ.1

Table 17

Underlying Constructs of the MSLQ

 

Factor 1 — Value Component: Intrinsic Goal Orientation

Item

1. In this class, I prefer course material that really challenges me so I can learn new
things.

16. In a class like this, I prefer course material that arouses my curiosity, even if it is
difficult to learn.

22. The most satisfying thing for me in the course is trying to understand the content as
thoroughly as possible. A

24. When I have the opportunity in this class, I choose course assignments that I can

learn ﬁom even if they don’t guarantee a good grade.

 

Factor 2 — Value Component: Extrinsic Goal Orientation

Item

7. Getting a good grade in this class is the most satisfying thing for me right now
11. The most important thing for me right now is improving my overall grade point

average, so my main concern in this class is getting a good grade.

 

 

1
All of the information presented about the MSLQ comes from A Manual for the Use of the Motivated
Strategies for Learning Questionnaire (MSLQ).

71

 

Table 17 con’t
13. If I can, I want to get better grades in this class than most of the other students.
30. I want to do well in this class because it is important to show my ability to my

family, ﬁiends, employer, or others.

 

Factor 3 — Value Component: Task Value

Item

4. I think I will be able to use what I learn in the course in other courses.
10. It is important for me to learn the course material in the class.

17. I am very interested in the content area of the course.

23. I think the course material in the class is useful for me to learn.

26. I like the subject matter of this course.

27. Understanding the subject matter of this course is very important to me.

 

Factor 4 — Expectancy Component: Control of Learning Beliefs

Item ‘

2. If I study in appropriate ways, then I will be able to learn the material in the course.
9. It is my own fault if I don’t learn the material in this course.

18. If I try hard enough, then I will understand the course material.

19. If I don’t understand the course material, it is because I didn’t try hard enough.

 

Factor 5 — Expectancy Component: Self-Efﬁcacy for Learning and Performance

Item

5. I believe I will receive an excellent grade in the class.

6. I’m certain I can understand the most difﬁcult material presented in the reading for

this course.

 

72

 

Table 17 con’t

12. I’m conﬁdent I can understand the basic concepts taught in this course.

15. I’m conﬁdent I can understand the most complex material presented by the
instructor of this course.

20. I’m conﬁdent I can do an excellent job on the assignments and tests in this course.
21. I expect to do well in the class.

29. I’m certain I can master the skills being taught in this class.

31. Considering the difﬁculty of this course, the teacher, and my skills, I think I will do

well in this class.

 

Factor 6 —- Affective Component: Test Anxiety

Item

3. When I take a test I think about how poorly I am doing compared with other students.
8. When I take a test I think about items on other parts of the teSt I can’t answer.

14. When I take tests I think of the consequences of failing.

19. I have an uneasy, upset feeling when I take an exam.

28. I feel my heart beating fast when I take an exam.

 

For the Portuguese version of the MSLQ, the Questionario de Estratégias Motivacionais
para Aprendizagem (QEMA), there are 35 questions for the motivation section. Four
questions about the proper attribution for the success in the course were added to this
Portuguese version. These questions were interspersed within the original questionnaire
so that the numbering does not coincide exactly. For clariﬁcation purposes, see Table 18

for a comparison of the item numbers of the MSLQ and the QEMA.

73

Table 18

Item Number Comparison for the MLSQ and QEMA

 

MSLQ item numbers

QEMA item numbers

 

1. In a class like this, I prefer course
material that really challenges me so I can

learn new things.

2. If I study in appropriate ways, then I
will be able to learn the material in the

course.

3. When I take a test I think about how
poorly I am doing compared with other
students.

n/a. If I don’t assimilate the course

content, it’s because it is too difﬁcult.

4. I think I will be able to use what I learn
in the course in other courses.

5. I believe I will receive an excellent
grade in this class.

6. I’m certain I can understand the most

difﬁcult material presented in this course.

1. Num curso coma este, dou preferéncia a
um tipo de material dida’tica que apresente
desaﬁos para que me permitam aprender
coisas novas.

2. Se eu estudar de forma apropriada, serei

capaz de assimilar a canteudo deste curso.

3. A0 fazer um teste, penso no quanta estau
me sainda mal camparada cam outros
alunas.

4 — new question: Se eu ndo assimila o
canteuda do curso, é parque ele é diﬁcil
demais.

5. Acredita que serei capaz de aplicar o
que aprender neste curso em outros.

6. Acredita que vau tirar uma nota
excelente neste curso.

7. Estau certa que sou capaz de assimilar o
canteuda mais diﬁcil apresentado na

bibliograﬁa deste curso.

74

Table 18 con’t

7. Getting a good grade in this class is the
most satisfying thing for me right now.

8. When I take a test I think about items

on other parts of the test I can’t answer.

9. It is my own fault if I don’t learn the
material in this course.

10. It is important for me to learn the
course material in this class.

11. The most important thing for me right
now is improving my overall grade point
average, so my main concern in this class
is getting a good grade.

12. I’m conﬁdent I can learn the basic
concepts taught in this course.

13. If I can, I want to get better grades in

this class than most of the other students.

n/a. If I assimilate the course content, it’s

because of the instructor.

8. T irar uma boa nota neste curso é a
maiar prémia para mim neste momenta.

9. Quando faco uma prova, penso sabre
questaes que na'o sei responder em autras
partes do teste.

10. A culpa é toda minha se ndo assimilar
a conteudo neste curso.

11. E importante para mim aprender o
canteuda ensinado neste curso.

12. A caisa mais importante para mim no
momenta é melhorar minha me’dia geral;
partanto, minha preacupacdo neste curso e'
tirar uma baa nota.

13. T enha certeza que posso aprender as
canceitas bdsicos ensinados neste curso.
14. Se depender de mim, vau tirar natas
melhores do que a maiaria dos outros
alunas.

15. — new question: Se eu assimilar a
canteudo deste curso, é basicamente

devida aa professor.

75

Table 18 con’t

14. When I take tests I think of the
consequences of failing.

15. I’m conﬁdent I can understand the
most complex material presented by the
instructor in this course.

16. In a class like this, I prefer course
material that arouses my curiosity, even if
it is difﬁcult to learn.

17. I am very interested in the content
area of this course.

18. If I try hard enough, then I will
understand the course material.

19. I have an uneasy, upset feeling when I
take an exam.

20. I’m conﬁdent I can do an excellent job
on the assignments and tests in this
course.

21. I expect to do well in this class.

22. The most satisfying thing for me in
this course is trying to understand the

content as thoroughly as possible.

16. Quando faco provas, penso nas
cansequéncias de ser repravada.

l7. Tenha certeza de que passa assimilar o
contezida mais complexa apresentado pela
professor neste curso.

18. Num curso coma este, prefira material
diddtica que desperte minha curiasidade,
mesmo que seja dificil de assimilar.

19. Estau muita interessado no canteuda
deste curso.

20. Se eu me esforcar o suﬁciente, vau
assimilar a canteuda deste curso.

21. T enha uma sensaca'a negativa e
desagradavel quando faco uma prava.

22. Tenha certeza que passa ter um étima
desempenho nos trabalhas e provas neste
curso.

23. Espero ter um barn desempenho neste
curso.

24. O que me traz maiar satisfaca'a neste
curso e' tentar assimilar a canteuda da

maneira mais detalhada passive].

76

Table 18 con’t

23. I think the course material in this class
is useful for me to learn.

24. When I have the opportunity in this
class, I choose course assignments that I
can learn from even if they don’t
guarantee a good grade.

25. If I don’t understand the course
material, it is because I didn’t try hard
enough.

26. I like the subject matter of this course.
27. Understanding the subject matter of
this course is very important to me. I
n/a. When I don’t assimilate the content

of this course, it’s the instructor’s fault.

28. I feel my heart beating fast when I
take an exam.

29. I’m certain I can master the skills
being taught in this class.

n/a. If I assimilate the content of this

course, it’s because it’s easy.

25. Acho que a material deste curso é util
para o meu aprendizado.

26. Quando tenha apartunudade, escalha
as trabalhas que possam me trazer
canhecimenta, mesmo que niio me
garantam uma baa nota.

27. Quando na'a assimila a contezida deste
curso, e' parque ndo me esforcei a
suﬁciente.

28. Gasto do assunta deste curso.

29. E muita importante para mim
compreender o programa deste curso.

30 — new question: Quando ndo assimila o
canteudo deste curso, é par causa do
professor.

31. Sinto o meu coracdo bater forte
quando estaufazenda uma prava.

32. T enha certeza que passa daminar as
habilidades desenvolvidas neste curso.

33 - new question: Se assimila a

conterido deste curso é parque ele é fdcil.

77

Table 18 con’t

30. I want to do well in this class because 34. Quero ser bem sucedida neste curso

it is important to show my ability to my parque é importante mostrar meu patencial
family, friends, employer, or others. para a minhafamilia, amigos, patra'o, etc.
31. Considering the difficulty of this 35. Tendo em vista 0 nz’vel de dificuldade
course, the teacher, and my skills, I think I deste curso, a professor e o meu patencial,

will do well in this class. acha que serai bem sucedida neste curso.

 

For the following discussion about the QEMA, the item numbers for the QEMA will be
used, which differ ﬁ'om the item numbers of the MSLQ. To start the analysis of the
motivation questionnaire, an exploratory Principal Components Analysis (PCA) using
principle component analysis and a direct oblirnin rotation was performed from the
questionnaire data from the 128 participants who completed the initial stage of the
project. An exploratory PCA was performed with the data collected from the QEMA.
The sample size was adequate, which is shown by a KMO of .767 and signiﬁcance of
.000. To increase the strength of the PCA, all of the questions that did not load onto the
ﬁrst ﬁve factors that emerged were eliminated. These items were eliminated based on
the combination of the analysis of the structure matrix and the total variance explained,
as well as the visual analysis of the scree plot. In addition, all of the item retained had an
eigenvalue greater than one. The amount of variance explained after factor ﬁve decreased
dramatically; the ﬁrst ﬁve factors could account for 48% of the variance with each
additional factor increasing this value by only about 4%. The combination of the

eigenvalues, the percent of variance explained, the number of questions loading on the

78

factors, and the visual analysis of the scree plot were used in tandem for the decision to
use only the ﬁrst ﬁve factors of the explanatory PCA. See Table 19 for the eigenvalues

for the ﬁrst ﬁve factors as well as the percentage of the variance explained:

Table 19

Exploratory PCA - Eigenvaules and Variance Explained for the First 5 Factors

 

 

Factor Eigenvalue % of Variance Cumulative % of
Variance

1 6.82 19.47 19.47

2 3.91 11.16 30.63

3 2.57 7.33 37.96

4 2.02 5.78 43.75

5 1.50 4.29 48.04

 

In addition, the majority of the questions loaded onto the ﬁrst ﬁve factors. See Figure 2

below for the scree plot, the visual representation of the factor loadings of the PCA.

79

Figure 2

Scree Plot for the Exploratory PCA

 

Eigenvalue
4?

N
l

 

0-

 

 

 

IIrtttrlrlIrllrrlItllrlrtllrlrrrlrt
l 20 25

l 0
Component Number

A second PCA using principal component analysis and a direct oblirnin rotation was run
with the remaining 24 questions. The KMO for this second PCA was .844 with the
signiﬁcance level at .000, indicating an adequate sample size. As found in Brown et al.
(2001), the reliability of the QEMA is strong. For each of the ﬁve factors found, the
Cronbach’s a was of a satisfactory level. Kline (1999) states that a value of .8 is an
acceptable value for all tests, whereas .7 and above is acceptable for cognitive-type tests.
For those tests dealing with a physiological construct (like motivation), Cronbach’s a. can
be even a bit lower and still be reliable. In addition, Cortina (1993) notes that it is more

likely to get a high value of Cronbach’s a with a higher number of items. In this case,

80

the factors which loaded a higher number of questions were more likely to have a higher
Cronbach’s a. That being said, Cronbach’s a for all of the factors was within adequate
range. Two of the factors, factor one and factor three, had more questions loaded than
the other factors. Factor one (eight items) had Cronbach’s a of .88 and factor three (ﬁve
questions) had Cronbach’s a of .79. The other three factors had fewer questions, but still
had Cronbach’s (1 scores of acceptable ranges: factor two (3 items), Cronbach’s a of .70,
factor four (three items), Cronbach’s a of .60, and factor ﬁve (4 items), Cronbach’s a of
.58. From this second PCA, a clear construct of ﬁve factors emerged, all of the questions
loading on these ﬁrst ﬁve factors having Eigenvalues of greater than 1. See Table 20 for

the eigenvalues for the ﬁve factors as well as the percentage of the variance explained:

Table 20

Conﬁrmatory PCA - Eigenvalues and Variance Explained

 

 

Factor Eigenvalue % of Variance Cumulative % of
Variance

1 6.79 29.54 29.54

2 2.60 11.31 40.85

3 1.77 7.69 48.54

4 1.55 6.72 55.26

5 1.19 5.15 60.41

 

81

The ﬁve factors found are illustrated below in Tables 21 and 22 followed by a discussion
comparing them with the underlying constructs of the MSLQ. Note that in the following
tables, the questions are not listed in numerical order but instead in the order of strength
that they contributed to the factors. Though Tables 21 and 22 seem similar, the factor
loading are slightly different, making it important to report both. Table 21 is the Pattern
Matrix for the factor loading, which is easily interpreted but can have some values
suppressed because of the relationship between the factors. Table 22, the structure
matrix, is a combination of the pattern matrix and the correlation coefﬁcients between

factors found in the PCA.

Table 21

Pattern Matrix for the Factor Loadings

 

QEMA Item number Factors

 

Factor I : Extreme confidence in abilities and l 2 3 4 5
success in the course (29.5% of variance,
eigenvalue = 6. 79)
1 I’m conﬁdent I can understand the most .84
7 complex material presented by the
instructor in this course.
7 I’m certain I can understand the most .83
difﬁcult material presented in the readings

for this course.

82

Table 21 con’t

13 I’m conﬁdent I can learn the basic

concepts taught in this course.
I’m certain I can master the skills being
taught in this course.
If I can, I will get better grades in this
class than most of the other students.
I’m conﬁdent I can do an excellent job on
the assignments and tests in the course.
Considering the difficulty of this course,
the teacher, and my skills, I think I will do
well in this class.
I believe I will receive an excellent grade

in this class.

F actor 2: Success with hard work and course

usefulness (11.3 % of variance, eigenvalue =

2. 60)

2

If I study in appropriate ways, then I will
be able to learn the material in this course.
If I try hard enough, then I will understand
the course material.

I think the course material in this class is

useful for me to learn.

83

.74

.73

.68

.61

.58

.47

.82

.69

.57

-.32

Table 21 con’t

F actor 3: Negative attitudes towards the
course (7. 7% of variance, eigenvalue = 1. 77)
1 It is important for me to learn the course

1 material in this class.

2 Understanding the subject matter of this

9 course is very important to me.

1 I am very interested in the content area of
9 this course.

2 I expect to do well in this class.

2 I like the subject matter of this course. .30
8
Factor 4: Test Anxiety (6. 7% of variance,
eigenvalue = 1.55)
1 When I take tests, I think of the
6 consequences of failing.
9 When I take a test, I think about items on
other parts of the test I can’t answer.
2 I have an uneasy, upset feeling when I

1 take an exam.

84

-.84

-.71

-.71

-.69

-.48

.82

.75

.57

Table 21 con’t

F actor 5: Challenges promote learning —

intrinsic motivation (5.2% of variance,

eigenvalue = 1.19)

1

In a class like this, I prefer course material
that really challenges me so I can learn
new things.

When I have the opportunity in this class,
I choose course assignments that I can
learn from even if they don’t guarantee a
good grade.

In a class like this, I prefer course material
that arouses my curiosity, even if it is
difficult to learn.

The most satisfying thing for me in the .55
course is trying to understand the content

as thoroughly as possible.

.80

.61

.56

.48

 

85

Table 22

Structure Matrix for the Factor Loadings

 

 

QEMA Item number Factors hz

Factor I .' Extreme conﬁdence in abilities and 1 3 4

success in the course (29. 5% of variance,

eigenvalue = 6. 79)

7 I’m certain I can understand the most .82 .69
difficult material presented in the readings
for this course.

17 I’m conﬁdent I can understand the most .80 .66
complex material presented by the instructor
in this course.

32 I’m certain I can master the skills being .77 .63
taught in this course.

22 I’m conﬁdent I can do an excellent job on .74 -.45 -.42 .69
the assignments and tests in the course.

13 I’m conﬁdent I can learn the basic concepts .73 .61
taught in this course.

35 Considering the difficulty of this course, the .73 -.51 .65
teacher, and my skills, I think I will do well
in this class.

14 If I can, I will get better grades in this class .67 .46

than most of the other students.

86

Table 22 can’t

6

I believe I will receive an excellent grade

in this class.

F actor 2: Success with hard work and course

usefulness (11.3 00 of variance, eigenvalue = 2. 60)

2

20

25

F actor 3: Negative attitudes towards the course

If I study in appropriate ways, then I will be

able to learn the material in this course.

If I try hard enough, then I will understand

the course material.
I think the course material in this class is

useful for me to learn.

(7.7% of variance, eigenvalue = 1. 77)

ll

19

29

23

28

It is important for me to learn the course
material in this class.

I am very interested in the content area of
this course.

Understanding the subject matter of this
course is very important to me.

I expect to do well in this class.

I like the subject matter of this course.

87

.45

.80

.74

.67

.41

-.41

-.80

-.78

-.75

-.70

-.58

.53

.66

.67

.55

.68

.72

.71

.63

.44

Table 22 con’t

F actor 4: Test Anxiety (6.7% of variance,

eigenvalue = 1.55)

1

6

2

1

When I take tests, I think of the consequences
of failing.

When I take a test, I think about items on other
parts of the test I can’t answer.

I have an uneasy, upset feeling when I take an

exam.

F actor 5: Challenges promote learning — intrinsic

motivation (5. 2% of variance, eigenvalue = 1.19)

1

In a class like this, I prefer'course material that
really challenges me so I can learn new things.
In a class like this, I prefer course material that
arouses my curiosity, even if it is difficult to
learn.

When I have the opportunity in this class, I
choose course assignments that I can learn

from even if they don’t guarantee a good
grade.

.80

.72

.63

.77

.63

.62

.67

.55

.48

.65

.56

.40

 

88

Table 22 con’t .55 .56 .59
24 The most satisfying thing for me in the
course is trying to understand the content

as thoroughly as possible.

 

Factor one corresponds very closely with the expectancy component: self-
efﬁcacy for learning and performance. Out of the eight questions that loaded onto this
factor, only question 14 is not part of this section on the MSLQ, but instead is in the
category of the value component: extrinsic motivation. Factor 2 has components of both
the value component: task value with item 25 as well as the expectancy component:
control of learning beliefs with items 2 and 20. This factor corresponds closely with the
value component: task value with item numbers 11, 19, 29, 28. Item 23 is part of the
expectancy component: self-efﬁcacy for learning and performance. It is important to
note that all of the items load negatively onto this factor, indicating that the participants
don’t agree with the items in this factor. For example, item 28 states that “I like the
subject matter of this course.” The negative factor loading would indicate that the
participant is thinking, “I DO NOT like the subject matter of this course.” This is true
for all of the items in factor 3, indicating a very low task value attributed to the
participants’ English courses. Factor 4 corresponds exactly with the affective
component: test anxiety on the MSLQ. Factor ﬁve corresponds exactly to the value

component: intrinsic goal orientation. It is also interesting to note the correlations

89

between factors. The component correlation matrix shows this relationship, and is

illustrated in Table 23 below.

Table 23

Component Correlation Matrix for the Conﬁrmatory PCA

 

 

Component 1 2 3 4 5

l 1 .20 -.30 -.21 .26
2 1 -. l 9 .09 .24
3 l .03 -.15
4 1 -.06
5 1

 

Although the correlations were not strong between any of the factors, there was some
relationship of the constructs presented by the QEMA. The fact that the correlations
were not strong indicates that each of the factors found in the PCA represents a construct
in of itself.

From this PCA, it is interesting to note that with only a few exceptions, the
constructs found using the QEMA with this group of participants correspond well with
the constructs deﬁned in the original MSLQ. This indicates not only that the original
constructs were well thought out and well grounded, but also that the Portuguese
translation is very successful because it measures the intended constructs of the MSLQ.
It must be pointed out that the MSLQ had 6 factors, whereas this PCA of the QEMA

produced only 5 factors. It is interesting to note that extrinsic motivation did not play a

90

strong role in this PCA, which was the factor that was included in the MSLQ and not this
version of the QEMA. This can perhaps be explained by the similar language
backgrounds of the participants (see the discussion section for further commentary on
this issue). Also interesting that the four new questions that were added to the QEMA
did not load on the ﬁve factors, indicating that the addition of these questions was
perhaps not necessary.

After the PCA was completed, and to ﬁnd the answer to the ﬁrst research
question: “Do bilingual learners have the same motivational proﬁle as multilingual
learners?” a discriminant frmction analysis was performed to ascertain if the bilingual
and multilingual participants in this study have distinct motivational proﬁles. The
dependent variable for analysis was the L2/L3 group divisions, and the independent
variable was the numerical average (from the Likert scale in the QEMA) of the responses
for the questions loading on each factor. The result is that for this population of language
learners, there is not a difference in motivational proﬁles. Both categorizations of the
multilingual learners that were previously mentioned were tested, and as seen in the
tables below, neither of the categorizations yielded signiﬁcant results for group
prediction with the discriminant function analysis. See tables 24 and 25 for the results of

the discriminant function analysis.

91

Table 24

Results of the Discriminant Function Analysis: Any Previous Language Experience

 

 

 

Function Eigenvalue Wilks’s Chi-Square % Variance p
Lambda

1 .033 .968 2.392 100 .793

Table 25

Results of the Discriminant Function Analysis: Perceived Positive Language Interaction

 

 

Function Eigenvalue Wilks’s Chi-Square % Variance p
Lambda
1 .030 .970 2.234 100 .816

 

As seen from the above results, none of the possible L3 divisions even approached
signiﬁcance, indicating that the motivational proﬁles of the bilingual and multilingual
participants in this study are very similar. See the discussion section for possible

explanations for this unlikely phenomenon.

RQ2: Which group of participants has higher language aptitude, bilingual or
multilingual learners?

To answer this research question, one-way ANOVAs were performed with each
categorization of multilingualism. The dependent variable for this analysis was the
language aptitude scores, and the independent variable was the L2/L3 group divisions.

Both categorizations yielded signiﬁcant results. Table 26 below illustrates these ﬁndings.

92

Table 26

ANOVA Results for the Bilingual and Multilingual Participant CANAL-FT Scores

 

L3 Categorization Number of participants AN OVA output for CAN AL-F T

for L2 and L3 groups scores

 

Categorization 1: Any L2: N=39 F (1, 77) = 9.274, p = .003*
L3 experience (M = 38.87, SD = 7.89)
L3: N=40

(M = 44.30, SD = 7.94)
Categorization 2: NPPLI: N=58 F (l, 77) = 18.082, p = .000"
Perceived Positive (M = 39.45, SD = 7.74)
Language Interaction PPLI: N=21

(M = 47.62, SD = 6.97)

 

Note: * indicates a p—value of at least .01; ** indicates a p-value of at least .001

The results of the language aptitude test are interesting on many levels. Firstly, the
results support the idea that even a small amount of exposure to a language can impact a
learner’s language aptitude, as seen by the results for Categorization 1. Though this is an
incredibly interesting ﬁnding and is one that contradicts much of the existing literature
on cross-linguistic inﬂuence, an even more interesting ﬁnding is that relating to
Categorization 2. In this case, a signiﬁcant difference was also found, despite the
extremely uneven distribution of the groups. Statistically speaking, with unbalanced
groups, it is even more difficult for a test to reach signiﬁcance. The indication of this is

that the concept of Perceived Positive Language Interaction is a very strong indicator of

93

language aptitude of a language learner. See the discussion section for further discussion

of this issue.

RQ3: Which group of participants has higher English language proﬁciency as measured
by the MSUELT, bilingual or multilingual learners?

To answer this research question, it is necessary to refer back to the ﬁrst part of the
results section for the discussion relating to English levels. (see Table 13 and Figure 1 for
details). Because the participants’ English semester is the strongest predictor of their
performances on the MSUELT, to explore the question of the dichotomy of the bilingual _
and multilingual participants, it is necessary to examine the participants within the
smaller groups found in the results of research question number three (see Table 14 for a
description of these groups). From the previously explained results, Group 1’s
(semesters 1 and 2) MSUELT scores signiﬁcantly differed from Group 3’s (semesters 5,
6, and 7) MSUELT scores, while Group 2’s (semesters 3 and 4) MSUELT scores did not
signiﬁcantly differ from either Group 1 or Group 3. Because of the wide range of
abilities according to the semesters of English, an overall participant analysis of the
performance on the MSUELT cannot be done. Therefore, according to the group
divisions made in the ﬁrst part of the analysis section, The MSUELT scores of Group 1
and Group 3 will be analyzed individually, and the results will be compared. Group 2’s
MSUELT scores will not be analyzed because that group did not differ signiﬁcantly ﬁ'om
Group 1 or group 3. Similarly to the previous analyses, both categorizations of
multilingual learners will be taken into consideration. The purpose of this analysis is to

see which group, bilinguals or multilinguals, performed better on the measure of English

94

language proﬁciency. See Table 27 for the results of the AN OVAs to compare the

bilingual and multilinguals’ scores on the MSUELT.

Table 27

AN OVA Results for the Bilingual and Multilingual Participant MSUELT Scores

 

L3 Categorization Group Number of participants AN OVA output for

number for L2 and L3 groups MSULET scores

 

Categorization 1: 1 L2: N=12 F(1, 17) = 3.307, p = .087
Any L3 experience (M = 49.75, SD = 11.86)
L3: N=7

(M = 40.86, SD = 6.44)
3 L2: N=14 F(1, 34) = .058, p = .812
(M = 67.14, SD = 17.88)
L3: N=22

(M = 68.23, SD = 9.19)

Categorization 2: l NPPLI: N=17 F(1, 17) = .158, p = .696
Perceived (M = 46.82, SD = 11.22)
language PPLI: N=2
interaction (M = 43.50, SD = 10.61)
3 NPPLI: N=23 F(1, 34) = .146, p = .705

(M = 67.17, SD = 14.88)
PPLI: N=l3

(M = 68.92, SD = 9.33)

 

95

None of the results are signiﬁcant, indicating that there is no difference between bilingual
and multilingual learners in the performance on the independent measure of English

proﬁciency for this group of learners.

RQ4: Which group has a higher retention rate in the progression of the English courses,
bilingual or multilingual learners?

This research question arose from the idea that if L3 learners are more successful
language learners, then they would naturally be more inclined to continue with the
language courses as the difficulty level increased. See Table 28 below for the

percentages of the L3 learner retention with the two categorizations of L3 learners.

Table 28

Retention of L3 Learners Across Levels

 

L3 Categorization Group Number of participants Percentage of L3 learners

number for L2 and L3 groups

 

Categorization l: 1 L2: N=12 36.84%
Any L3 experience L3: N=7
3 L2:N=14 61.11%
L3: N=22
Categorization 2: 1 L2: N=l 7 10.53%
Perceived L3: N=2
language 3 L2: N=23 36.11%
interaction L3 : N=1 3

 

96

The raw data percentages show that for both L3 categorizations there are clearly more L3
learners in the higher levels than in the lower levels. In addition to the percentages,

F isher's exact test was run with the data for both categorizations of L3 learners to see if
there is a signiﬁcant difference between the number of L3 learners in the lower levels
and the higher levels. Categorization 2 of L3 learners has a value that is approaching
signiﬁcance. Although neither of the values is signiﬁcant, a pattern that shows more L3
learners in the higher levels of English classes can be interpreted by using the
percentages and the p-values from the Fisher’s exact test. See Table 29 for a summary of

the results.

Table 29

Summary of the Results for Fisher's Exact Test

 

 

L3 Categorization p-value
Categorization 1: Any L3 experience p = 0.0992
Categorization 2: Perceived Positive p = 0.0582
Language Interaction

 

Analysis: Summary of Quantitative results
The preceding tables and descriptions illustrated the results of the quantitative statistical
analysis used for this dissertation. Because of the length of this section, a summary

(Table 30) is provided below to summarize the results of the quantitative analysis.

97

Table 30

Quantitative Statistics: Summary of Research Questions, Analyses, and Results

 

 

Research Questions Analysis Summarized Results

RQl: Do bilingual learners Discriminant No signiﬁcant difference in the
have the same Function Analysis motivational proﬁles of the
motivational proﬁle as with the factor bilingual and multilingual learners
multilingual learners as scores found by the

measured by the QEMA? PCA as well as the
original constructs
in the MSLQ
RQ2: Which group of one-way AN OVA Statistically signiﬁcant results for

participants has higher with the CANAL- L3 Categorizations l and 2, with

language aptitude, FT scores of the the strongest being with
bilingual or multilingual participants. Categorization 2. This indicates
learners? that L3 learners have a higher

language aptitude than L2 learners.
RQ3: Which group of one-way AN OVA No signiﬁcant differences found
' participants has higher with the MSUELT
English language scores of Group 1
proﬁciency as measured and Group 3 (as
by the MSUELT, bilingual deﬁned for RQ3)
or multilingual learners? analyzed

individually.

98

Table 30 con’t

RQ4: Which group has a
higher retention rate in the
progression of the English
courses, bilingual or

multilingual learners?

Raw percentages
calculated. Fisher's
exact test used for
the numbers of
participants in
Group 1 and Group

3 (as deﬁned by

RQ3).

99

For both L3 categorizations, the
percentages indicate that there is a
higher number of L3 learners in the
higher levels. However, the
Fisher's exact test did not yield
signiﬁcant results, although the
results are approaching

signiﬁcance.

CHAPTER FOUR: QUALITATIVE RESULTS

Language learning cannot be explained entirely by a battery of tests. Participants are
individuals with many different facets that create their language learning processes. The
purpose of this qualitative analysis it to provide additional support for some of the
ﬁndings presented in the quantitative section. Two main themes emerged from the
interview data that directly related to this dissertation — motivation for language learning
and the interactions between languages. The methodology for the content analysis as well
as the detailed results for these two themes will be explained in this chapter.

This qualitative analysis is comprised of the results of a content analysis
performed with the data from 29 interviews. The participants for these interviews were a
subset of the multilingual students who completed all three previous parts of the project
as a whole. The intent of the interviews was to gain deeper insight into the language
backgrounds of the multilingual participants; thus, only the multilinguals were asked to
volunteer for the interviews. The participants were given the choice of which language
they preferred for the interviews, and all of the participants chose to conduct the
interview in Portuguese. For the interviews, a semi-structured interview format was used
with a preliminary list of questions. The times for the interviews ranged ﬁom 10 minutes
to approximately one hour, depending on how much each individual participant had to
say in response to the questions. All of the interviews were recorded on tape and then
transcribed by a native speaker of Brazilian Portuguese. These transcripts were

subsequently veriﬁed by the principle researcher.

100

Although all of the answers to the questions proved insightful, certain questions
in the interview are more pertinent to the study at hand; thus, only a subset of the
questions asked was used for analysis. The content analysis results provided below are
divided into two parts: motivation for language learning and interaction between
languages. The interview questions dealing with motivation for learning both English
and the other languages studied are directly related to the ﬁrst research question of this
dissertation: Do bilingual learners have the same motivational proﬁle as multilingual
learners? In looking at the detailed responses about the reasons for learning the different
languages, results ﬁ'om the ﬁrst research question regarding the unexpectedly similar
motivation proﬁles of the L2 and L3 learners can be explained further. As for the
questions dealing with the interactions between languages, this theme relates to the idea
of reconceptualizing a multilingual learner. The interview questions inquire about the
perceived interactions of the foreign languages studied as well as the perceived
interactions between the foreign languages and the L1 (Portuguese). For the illustrative
quotations provided below, the original Portuguese statements are provided along with
the English translations. The ﬁrst part dealing with the motivation for learning languages

is comprised of the answers for the following two questions:
1. Why did you want to learn English?

2. Why did you want to learn the other languages you have previously

studied?

101

The second part of the analysis dealing with the perceived interaction between languages

is with the answers to the following two questions:

1. If you have studied other languages in the past, do you think that this
has helped or hindered your ability to learn other languages? Please '
provide speciﬁc examples where appropriate.

2. Do you think that studying other languages has affected the way you

understand Portuguese? Please give detailed examples.

The content analysis was done by a single rater (the principal researcher of this
dissertation) for several reasons. Firstly, the interview data were all in Portuguese, and a
second Portuguese-speaking applied linguist familiar with content analysis was not
available at the time of the study. Translating the interview data into English for the
content analysis would have resulted in the loss of some of the intricacies of the
interviews. In addition, the principle investigator was the most familiar with the
interview data and the context in which the interviews were conducted, being the one
who conducted the interviews in Brazil.

Although there was only one person doing the coding of the interview data
,reliability was enhanced by applying rigorous standards to the content analysis. The
categories that emerged from the analysis were evaluated by the model proposed in
Richards (2003) used for evaluating categories. Richards (2003) stated that categories
should be analytically useful, empirically relevant, practically applicable, and

conceptually coherent (p. 276). The coding schemes in this dissertation follow his

102

model. When deciding on the categories for the analysis, all of the interviews were read
several times for familiarity of the answers. Then, a primary coding scheme was
developed and the answers were classiﬁed according to this scheme. In order for a
category to stand alone, at least three responses from one of the questions had to ﬁt into
this category, making the coding schemes both practically applicable and empirically
relevant. After the initial coding was complete, the researcher re-evaluated the responses
to ensure the accuracy of the coding. The coding schemes are analytically useful in that
they lead to an understanding of the ﬁrst research question dealing with motivation as

well as the categorization of the multilingual participants.

Analysis: Part one — Motivation for language learning

This analysis addresses the following two questions together “Why did you want to learn
English?” and “Why did you want to learn the other languages you have previously
studied. For both of these questions, the following themes emerged: universal
language/useful, work, study, travel, general enjoyment (e.g. thought it was interesting),
culture, and esthetics. These themes were then divided into two overarching categories:
practical and personal. The practical category included any response that mentioned the
usefulness of the language in any way (e. g. using the language to DO something). The
personal category encompassed responses involving language for personal enrichment or
enjoyment. See Table 31 below for a summary of responses for both of the interview
questions. It is important to note that in Table 31 below, as in all tables in this section,
the numbers in the table represent the number of responses for each category. In other

words, each number listed does not represent a single participant but instead a comment

103

made by a participant. A single participant could have several comments classiﬁed in

various categories for the same question.

Table 31

Motivation for Studying Speciﬁc Languages

 

English Language other than English

 

Practical Personal Practical Personal
U W S T Enj C Es U W S T Enj C Es
12 8 8 5 10 1 2 2 6 12 l 8 3 5

Total Practical: 32 Total Personal: 13 Total Practical: 21 Total Personal: 16

 

Note: U=Universal, W=work, S=study, T=travel, Enj=enjoyment, C=culture, Es=Esthetics

As can be seen by the above table, the motivation for studying English differs somewhat
from the motivation for studying other languages. Firstly, there was a great spread
between English and the other languages with the idea of practicality or of being a
universal language. Although Brazil is located in South America, a predominantly
Spanish-speaking continent, only one participant mentioned the idea of Spanish being a
practical or universal language, although this participant did not study Spanish. On the
other hand, roughly half of the participants mentioned something about the need to study

English because of its universal nature.

104

Example 1:

0 Inglés parque é a lingua universal e que toda lugarfala lnglés.

English because it is a universal language and is spoken everywhere
(Participant 20).

Example 2:

Inglés parque toda munda cansidera bdsico.

English because everyone considers it necessary (Participant 38).

There were two comments about the general usefulness of a language other than English,
both including French and one also including Spanish. One was a general comment

about the necessity of French in conjunction with liking it:

Example 3:

0 Frances é mais par gasta, parque eu acha legal 0 som e par
necessidade.

French is more for desire because I like the sound of it, but also for need

(Participant 70).

The second comment mentioned the importance of both English and Spanish for the CV
and how the study of French came out of studying English:

Example 4:

Primeiro parque e' importante para a curricula e Inglés e Espanhal sdo as

duas linguas impartantes, ai eu me interessei pela Inglés e depais gostei

105

do estudo de linguas, entdaﬁz a F rancés. Mas no inicia era s6 pela
impartcincia mesmo e para a curricula.

First because it is important for your CV, and English and Spanish are the
two important languages for that, so I studied English. Then I liked
studying languages, so I studied French. But in the beginning, it was just
because studying language was important, and for my CV (Participant

79).

Another interesting trend is the dichotomy of language study for working and studying.
For the question involving English, there were an equal number of participants, eight in
both cases, mentioning the practicality for working and studying. In the case of necessity
for working, many of the participants mentioned the need to know English for getting a
job in general, not speciﬁcally for a job they already had or even a speciﬁc job they had
in mind. Regarding study, several of the participants mentioned English as important for
having access to texts needed for their studies or the desire to study abroad. Below is an
example of a comment made about the usefulness of English for work, followed by an

example of the useﬁrlness of English for study.

Example 5:

Inglés, parque é pré-requisita, quase todas as empregas sc’ia super
impartantes.

English, because it’s a pre-requisite, it is considered very important for

almost any job (Participant 30).

106

Example 6:

Eu faca F ilasaﬁa e mesmo na minha area deﬁlasaﬁa é uma lingua
importante. Aprendenda inglés eu passa ter acessa a bem dizer todas as
abras defilasoﬁa que eu quiser.

I study philosophy, and English is an important area for philosophy. By
studying English I can have access to all the philosophical works that I

want (Participant 54).

When dealing with languages other than English, there was not a balance in the
perceived need of language for purposes of work and study. In this case, there were twice
as many participants who talked about the usefulness of the language for studying,
twelve participants, than for working, six participants. The imbalance of the total
numbers for work verses study indicates that although languages other than English are
useful in terms of schooling, the more useful language is English because it would be
used in the workplace, even after all schooling has been completed. For the participants
who mentioned the usefulness of a language other than English for work, it was often the

case that that person had a speciﬁc job in mind, such as the following:

Example 7:

0 Italiana fai necessidade de trabalha, parque eu trabalha cam turisma e
é preciso e aqui em F artaleza precisa saber Italiana.

Italian was a necessity for work because I work in tourism, and it’s

necessary; here in Fortaleza, you have to know Italian (Participant 51).

107

For the case of studying, there was an aspect to these answers that did not exist in the
answers for studying English. Several of the participants mentioned studying Spanish
because it was the easiest language to study to pass the foreign language section of the

vestibular, the test at the end of high school needed to get into university:

Example 8:

0 espanhal é uma questda de facilidade na hora da prava do vestibular,
samente para também ter mais canhecimentas de uma nova lingua.

I chose to study Spanish because it was the easiest option when I had to
take the vestibular, and also to have knowledge in one more language

(Participant 6).

There were also comments regarding study abroad to ﬁnish school or to do an advanced
degree and comments about the general usefulness of a language for a speciﬁc area of
study. Interestingly, there was no mention of needing the language to access texts for

studying:

Example 9:

para poder viajar para terminar minhafaculdade em outra pais e coma

eu estudo ﬁ'ancés e inglés eu na'a sei para onde deva ir.

108

...in order to go to another country to ﬁnish college, and because I’ve
studied French and English, I don’t know where'I should go (Participant
2).

Example 10:

0 F rancés é muita zitil na minha praﬁssa'a, parque estudo Histéria...
French is very useful in my profession because I study History...

(Participant 67).

The category in which the idea of English as a universal language rose again was in the
category of using language for travel. Although no speciﬁc countries were mentioned
along with the mention of language use for travel, ﬁve people mentioned that English

would be useful for traveling in general:

Example 11:

Eu acha que ajuda muita no trabalha e também nas viagens. Coma um
dia pretenda viajar para algum pois, vai me ajudar muita.

I think it’s useful for work and also for traveling. One day I plan on
traveling to some country, so I think it will help me a lot (Participant 52).
Example 12:

...falar comﬂuéncia, se quiser viajar, fazer turisma...

. . .to speak ﬂuently, in order to travel as a tourist (Participant 65).

109

For the question of the usefulness of a language other than English regarding travel, there

was only one comment:

Example 13:

Interesse pessoal, aprender mais, ensinar as pessaas de outros paises e
quem sabe futuramente viajar.

For personal interest, to learn more, to teach people from other countries,

and who knows, travel in the ﬁrture (Participant 69).

When looking at the category of usefulness of language study for personal reasons, the
numbers seem to indicate that the reasons for studying English (13 comments) or a
language other than English (16 comments) are almost the same. However, upon closer
examination, the responses differ in content. There was only one comment about

studying English for cultural reasons:
Example 14:
Devido a cultura mesmo.

Due to the culture itself (Participant 74).

However, for the answers about studying a language other than English, there were three

people who stated they studied the language because they liked the culture. For example:

110

Example 15:

Japanes, parque eu gasta da cultura, desenhas, das revistas em quadrinha
e também parque me atrai a cultura oriental.

Japanese because I like the culture, drawings, comic books, and also
because I’m attracted to oriental culture in general (Participant 12).
Example 16:

Eu gasta muita da cultura (espanhal), enta'afoi par issa.

I like the Spanish culture a lot, so it was because of that (Participant 57).

The participants also mentioned that they enjoyed studying languages for esthetic
reasons, although there were more comments with regards to the languages other than
English for this category. For English, there were two comments about the esthetic

attractiveness of English. For example:

Example 17:

Parque é uma lingua muita bonito...

Because it’s a very beautiful language (Participant 33).
Example 18:

Parque eu tanta acha bonito quemfala inglés...

Because those who speak English are so beautiful (Participant 6).

With regards to the languages other than English, there were ﬁve comments with regards

to the esthetics:

111

Example 19:

0 Frances é mais par gasta, parque eu acha legal 0 som...

French because I like it — I like the sound of it (Participant 70).
Example 20:

Oﬁancés parque eu simpatizei e achei uma lingua bastante sofisticada,
coma assim — chique.

French because I liked it and I thought it was a sophisticated language —

chic (Participant 31).

Besides the personal reasons for studying regarding culture and esthetics, there were
eight comments about studying English for general personal preference. Some offered

little explanation such as the following:

Example 21:

Inglés, parque, eu na'o sei, escalhi e sempre desde pequena que eu gasta
da lingua.

English because, I don’t know, I chose it, and ever since I was small I

liked the language (Participant 55).

There were other, more speciﬁc comments, such as the following:

112

Example 22:

Parque acha que abre as horizontes para o mundo, sair um pouco das
ﬁonteiras do meu pais e daqui do local.

Because I think that it opens horizons for me to the world so that I can
cross the borders of my country and of my local environment (Participant

45).

There were also eight comments about studying a language other than English just
because they liked it for general reasons. These reasons varied quite a bit; some were
intrigued by some sort of incidental contact, and some studied one language and liked it,
so they wanted to study other languages. The following quote is an example of the

aforementioned ideas:

Example 23:

0 Alema'a e mais hobby e coma gasta de linguas, ﬁcou nessa ordem: em
primeira a Inglés, segundo 0 Frances e terceiro o Alema'o parque um
ajuda o outro.

German was a hobby, and because I like languages, it was in this order:
fu'st English, second French, and third German because one helps the

other (Participant 72).

As can be seen from the above discussion, although the practical and personal

motivations for studying English and languages other than English are similar in number,

113

upon closer look, the motivations are not identical. The participants were more likely to
state more practical reasons for wanting to learn English, such as work, study, travel, and
the universality of English. Languages other than English emerged as important for
studying, but less crucial to have while entering the job market. Questions about
language other than English also generated comparatively more responses regarding the
culture and general fondness for the language. See the discussion section for a
comparison of the quantitative and qualitative measures of motivation and how they

interact.

Analysis: Part two - Language interaction

This section of the/content analysis examines the participants’ perceptions of language
interaction from two angles. The ﬁrst angle is the interaction of the foreign languages
studied; the second angle is the interaction of the foreign language with Portuguese, the
L1 of all the participants. In this analysis the “gene ” category signiﬁes an
acknowledgement of a relationship between languages with no speciﬁc example given
about what kind of relationship. Below Table 32 summarizes the comments made about

the relationship of foreign languages studied.

114

Table 32

Interactions of Foreign Languages Studied

 

 

 

Yes No/negative affect
General Lexical Grammar/structure Strategies

5 13 4 6 6

Total for YES: 22 Total for NO: 6

 

Some of the participants stated that yes, there was a positive interaction between

languages studied, but no speciﬁc examples were illustrated:

Example 24:
A experiéncia de um idioma ajudou...

The experience of a language helped...(Participant 69).

The most common interaction seen was that of a lexical interaction. There were

some general comments about lexical interactions such as the following:

Example 25:

Uma complementa a autra, parque tern muitas palavras parecidas na
escrita e no signiﬁcada.

One complements the other because there are many words that are similar

in spelling and in meaning (Participant 30).

115

There were other comments about speciﬁc lexical interactions. Some of the perceived
lexical interactions were expected, such as the transfer of lexical items to languages in

the same family or the noticing that many English words come from Latin roots:

Example 26:

Primeiro eu acha assim: Se vacé fala uma lingua latina como a portugués
é muito mais facil para vocé falar a espanhal. Quando vocé fala o
espanhal, vocé tem uma base para falar italiano. Do italiano para falar
francés, do francés para falar catala'o que é um dialeta da Espanha Mas
se vocé aprende inglés, coma sendo uma lingua secundaria é melhor
parque tem muitas palavras latinas.

This is what I think: If you speak a Romance language such as
Portuguese, it’s much easier for you to speakSpanish. When you speak
Spanish, you have the base to speak Italian, ﬁom Italian to French, from
French to Catalan, which is a dialect of Spain. But when you learn
English as an additional language, it’s better because it has a lot of Latin-

based words (Participant 62).

There were other comments about lexical interactions that were not expected, such as the

interaction of French, English and Japanese vocabulary:

116

Example 27:

Ofrancés e japonés passuem muitas coisas parecidas, principalmente a
japonés ele usa muitas palavras vindas diretamente do inglés. (31)
French and Japanese have many similar things; furthermore, Japanese

borrows a lot of words from English (Participant 31).

In addition, there were a few participants who commented on grammatical interactions

between languages studied:

Example 28:

Eu acha que todas as linguas, elas tém alga em comum. Assim como a
portugués e o espanhal, o inglés e a alema'o, principalmente em algumas
estruturas gramaticais.

I think that all languages have something in common. For example,
Portuguese and Spanish, English and German, mostly in some of the

grammatical structures (Participant 39).

Several participants also stated that learning multiple languages helped with their overall

language learning strategies:

117

Example 29:

Parque a gente acabafazendo analogias, acaba tracando um paralela
entre um idioma e autra, a que em minha opinia'ofacilita.

Because we end up making analogies, sketching a parallel between
languages, and in my opinion, this makes thing easier (participant 2).
Example 30:

Acho que ajudou muita, parque ajudou a entender devida as tematicas
dos exercicias dos livros , ja aprende a escutar com maisfacilidade a
outra lingua que nc’io seja a sua.

I think it helps a lot because it helps to understand homework in the sense
of the types of questions in the workbooks, and also the fact that you learn

to listen with more ease a language that is not your own Participant 26).

However, while the vast majority of responses to this question reported some kind of
interaction between languages, there were also those participants who did not see any

interaction with any languages:

Example 31:
N50. 0 alema'o na‘a me ajudou muita na'o.

No, German didn’t help me [with English] (Participant 33).

The question regarding the interaction of the foreign languages studied and Portuguese

also yielded interesting results. The categories that emerged in this part of the analysis

118

were similar to the categories for the interactions of the foreign languages, with the
exception of the category about learning strategies. Since Portuguese is the ﬁrst
language of the participants, no explicit strategies are needed to learn this language.

. Table 33 below is a summary of the types of comments regarding the interactions of the

foreign languages to Portuguese.

Table 33

Interactions of Foreign Languages and Portuguese

 

 

 

Yes No

General Lexical Grarnmar/ structure

6 6 9 7

Total for YES: 21 Total for NO: 7

 

Several participants commented on general ways that studying foreign languages helped
them understand Portuguese better. The following quotation is an example of how other

languages helped this participant think more rationally in Portuguese.

Example 32:

T ecnicamente, eu nunca pensei nisso. So se fai naforma de raciocinar
que me ajudou. T alvez eu tenha melhorado o Portugués na técnica do
raciocinio, parque eu nuncafiz um curso de Portugués...

I never really thought about it. It was in the way of rationalizing that
helped me. Perhaps I have improved my rationalizing skills in Portuguese

because I have never taken a Portuguese course. ..(Participant 45).

119

There were also several participants who mentioned that by studying other languages,

they realized how difficult Portuguese was:

Example 33:

Olha, eu acha que sim. Parque dd para perceber o quanta a lingua
portuguesa é complicado.

Look, I think so because it helps to understand how complicated

Portuguese is (Participant 65).

As far as the lexical items were concerned, several participants commented that studying

other languages improved their vocabulary in Portuguese:

Example 34:

Ajuda parque tern algumas palavras em Espanhal que sa'o cognatas de
palavras portuguesas intelectualizadas, fala mais bonito.

It helps because there are some words in Spanish that are cognates of

intellectual words in Portuguese, more beautiful language (Participant 4).

The area that had the most effect according to the participants was the area of
grammatical interaction. Several of the participants noted that through studying other
languages, they learned more about Portuguese grammar or that they were motivated to

learn more about their own language:

120

Example 35:

Principalmente na parte da grama'tica onde vocé tem que estudar coisas
coma verbos transitivos e intransitivos. E essa parte da gramdtica que
ajudou, parque eu tive que saber o que é que eu tinha que estudar na
gramdtica mesmo na parte estrangeira. Entcio primeira eu tinha que ver
na minha gramcitica, fazer comparacdes.

Mostly in the part of grammar where you have to study things like
transitive and intransitive verbs. It’s in this part of the grammar that
helped because I had to know what it was that I had to study in the foreign
language grammar. So, ﬁrst I had to learn it in my own grammar so that I
could compare the two (Participant 67).

Example 36:

A gramdtica e a conjugaca'o dos verbos dos dais idiomas sda bem
diferentes e isso me incentiva em canhecer os da minha, estudar mais a
minha gramatica para comparar mais.

The grammar and the verb conjugation of the two languages are very
different and this encouraged me to learn them in my language, to study

my grammar more to be able to compare more (Participant 33).

There were also those participants who did not see an interaction between the foreign

languages and Portuguese:

121

Example 37:

Eu acha queﬁca igual. Assim, eu acha que vocé acrescenta autras
coisas, portuguésﬁca limitado aa portugués, ai vocé acrescenta outras
informagﬁes no inglés e no alemc‘io.

I think that it stays the same. Even though I think that you augment other
areas, Portuguese stays limited to Portuguese, but you increase your

knowledge in English and German (Participant 39).

It is interesting to note that in the case of both questions, more participants found
interactions than those who did not. The most common interaction found between
foreign languages was that of lexical origin, and the most common type of interaction
between the foreign language and Portuguese was grammatical. Overall, the participants
deemed the interactions between the various languages in question to be positive in
nature.

The purpose of this content analysis was to delve deeper into the minds of
language learners to investigate ideas that cannot be measured by quantitative measures
alone. This content analysis has shown that the participants have a variety of
motivations for learning foreign languages, and these reasons can differ depending on the
language in question,. The participants had differing motivation for studying English
and the other foreign languages, and their motivations for studying English may have had
an effect on the motivational proﬁles with regards to the English language classes that

were elicited as part of this research (see the discussion section for more on this issue).

122

In addition, many of the participants found interactions between languages, both
between foreign languages studied and between foreign languages and their ﬁrst
language — Portuguese. The questions about language interactions greatly adds to the
understanding about what kinds of relationships the participants see between languages,
leading to a more profound exploration of the Perceived Positive Language Interaction

operationalization of multilingualism.

123

CHAPTER FIVE: DISCUSSION AND CONCLUSION

Discussion
Before discussing the results related to the research questions, a discussion about the
nature of deﬁning multilingualism is essential. How to deﬁne a multilingual learner is an
extremely difficult question. Historically, researchers have deﬁned people as
multilingual only when they have a considerably high proﬁciency in all of the languages
in question. Recently, however, other researchers, notably De Angelis (1999, 2005a,
2005b) and Selinker and Baumgartner-Cohen (1995), have found a multilingual effect
even with very limited proﬁciency in the L3 (see the literature review for a complete
discussion of this issue). The issue at hand is complex; several factors come into play
when attempting to deﬁne a multilingual learner including proﬁciency in the L3,
language choice/family, the relationship to the L1 or other languages studied, and the
context of language acquisition, to name a few. It is also problematic that as the criteria
for deﬁning a multilingual learner become more stringent, the pool of potential
participants inevitably decreases, especially when dealing with a classroom setting for
recruiting purposes. However, it is generally assumed that as the deﬁnition for an L3
learner becomes stricter, the effect size will be greater once the analyses are completed.
That being said, what happens if the criteria become so strict that the L2/L3 groups
become unbalanced? The key is to ﬁnd at what point the L3 effect starts to occur so that
valuable participants are not excluded ﬁom the inevitably smaller L3 group.

In this study, all statistical tests were performed twice using 2 different

categorizations of L3 learners. The ﬁrst categorization covered any L3 experience, the

124

second Perceived Positive Language Interaction. Signiﬁcant differences were found for
both categorizations tested, with the strongest effect often being in Categorization 2.
These results bring up several interesting points. First, a signiﬁcant effect of
multilingualism appears even among those participants with a very low level of previous
language experience. The idea behind the second categorization of multilingualism,
Perceived Positive Language Interaction, is that the multilingual participants can be
divided into two groups: those that are cognizant of the interactions between the
languages that they have studied and those who are not. In other words, some
multilinguals are self-aware of the beneﬁts of having studied multiple languages. It was
the case that there were signiﬁcant results for this categorization even though this group
was quite unbalanced in terms of numbers. This point is most strongly illustrated by
Table 26, which shows the results for the language aptitude test. Categorization 2 for L3
learners shows a signiﬁcant difference despite having extremely unbalanced groups of 58
L2 participants and 21 L3 participants.

To summarize this point, it must be said that there is indeed an effect for even
very limited experience in a previous language. With this idea of L2/L3 learner
classiﬁcation, can a learner truly be considered multilingual if he or she does not
recognize the inherent interaction that comes with learning multiple languages? Keeping
these two ideas in mind — the effect of a small amount of language learning and the
perceived interaction between languages — researchers should be cautious when

classifying participants as L2 or L3 as this classiﬁcation could greatly affect their results.

125

RQI : Do bilingual learners have the same motivational profile as multilingual learners?

HI : Bilingual and multilingual learners have distinct motivational proﬁles.

To answer this research question, a PCA was performed on the QEMA, the Portuguese
version of the MSLQ. The factors revealed in the conﬁrmatory PCA were very similar to
the factors on the original MSLQ, indicating that the translation of the MSLQ into
Portuguese retained the intended meaning of the English version. This result supports
the idea that this assessment tool can possibly be used across various cultures and is
perhaps not culturally sensitive to the country in which it was written. This is an
important ﬁnding not only for this Portuguese version but also for versions in other
languages that have been used in various contexts.

It is interesting to note that the only factor that was not the same as the original
MSLQ was that of extrinsic motivation. In the QEMA, the questions intended for this
factor did not load strongly on any of the new factors found. In fact, all but one question
dealing with extrinsic motivation were eliminated from the analysis with the exploratory
PCA due to low factor loadings. It is possible that the cause of this is the very high
intrinsic motivation of these participants, such that their success does not depend on
outside concerns, but instead more on internal concerns.

Another point of discussion concerns the four new questions added to the QEMA
that were not in the original MSLQ. These questions dealt with possible reasons for
success or failure in the language classroom, such as the difficulty level of the course and
the effect of the professor. It is interesting to note that not one of these four questions
were included in any of the ﬁnal ﬁve factors because of low factor loadings on the

exploratory PCA. It can perhaps be said that these questions are unnecessary to the

126

QEMA and do not provide any further insights than do the original questions on the
MSLQ.

This hypothesis of the distinct motivational proﬁles of bilingual and multilingual
learners was not proven correct. Results from the discriminant function analysis
demonstrated that there were no distinctions between the bilingual and multilingual
participants in terms of their motivational proﬁles. By exploring the backgrounds of the
participants in this study, the perceived universal nature of English, and the differing
contexts for the study, perhaps a reason for this discrepancy can be found.

To begin with, all of the participants are enrolled in a very competitive language
program that does not give them university credit. The government heavily subsidizes
this program so that that the cost for a one semester course is about $30.00. To be
accepted into this program, the students have to take a language placement test and only
the students who score the highest on this placement test are admitted into the language
school. The fact that no university credit is given to these students who complete the
classes, means that all of the students, whether they are L2 or L3 learners, could
inherently have a relatively high level of motivation to begin with. This is quite a
different context from the typical classroom language learning setting where students
enter for a variety of reasons, some merely for a graduation requirement or the like. The
same tool used on such a population of students may yield different results.

The content analysis of the interview questions can also shed light on this result.
The motivation questionnaire asks questions about a speciﬁc language course, as
opposed to motivation for language learning in general. For example, question one

states: “In a class like this, I prefer course material that really challenges me so I can

127

learn new things.” The participants were all instructed to answer the questions on the
QEMA thinking of the English course, as opposed to the other language courses. This
was done because all of the participants were currently enrolled in English language
courses, but not all of the participants had taken languages other than English. The
content analysis shows some interesting trends with regards to motivation for taking
English courses as opposed to taking other language courses. Only 29 of the multilingual
participants were interviewed about their language learning process; however, these
interviews provide a sample of the differing language attitudes towards English and the
languages other than English. Perhaps one of the most important ﬁndings of this
qualitative analysis is the fact that many of the participants regard English as a universal
language, one that is essential for advancement in work or for study. This result
contrasts sharply with the perceived utility of other languages studied - not one of the
participants stated that any other language was “universal” in nature. Not even Spanish
was labeled as universal, a surprising ﬁnding coming from people living in a country
surrounded by Spanish-speaking countries.

The interviews were conducted with the multilingual participants only, so this
next part of the discussion requires some speculation. Considering the sampled
multilingual participants’ language attitudes towards English versus the other foreign
languages, it could be assumed that the bilingual participants might have the same
attitude towards the study of English—that is English’s universal nature. However, while
it is possible that the reason for the similarities between the bilingual and multilingual
learners in their motivational proﬁles could be explained by similar attitudes regarding

the necessity of studying English to be competitive in today’s global market, the results

128

cannot completely be explained by this presumed “universality” of English. In the
aforementioned studies of multilinguals in bilingual societies, the language in question
for the L3 was also English. The universality of English did not affect the differing
motivational proﬁles of the bilingual and multilingual learners in these cases. In fact, in
Sanz (2000), the factor of motivation was so strong that it had to be taken out by using a
hierarchical regression model so that the effect of the other variables could be seen. Thus,
although the universality of English was clearly elaborated by the participants, this alone
cannot fully explain the similarities found in the motivational proﬁles.

It is clear that the contexts in for previous studies dealing with motivation and the
context for the current study differ greatly, which can also lend to interpreting the results
for the ﬁrst research question. In addition to the highly competitive entry to this language
school, the participants shared other similar traits. For example, with the exception of
one participant who worked as a tour guide in a different town, they had all learned their
languages in a classroom setting. Only one participant had traveled abroad, so their
cultural knowledge was obtained primarily through the language classes, and for some,
music and television. Fortaleza is an inherently unilingual context with very few people
growing up hearing more than one language at home. None of the participants grew up
in a bilingual setting, making their context for language learning very similar. In
addition, the participants stated that they had little to no contact with native speakers of
the languages that they were studying. This group of multilingual participants differs
greatly from previous studies with multilingual learners in which the multilinguals were
part of an inherently multilingual society. For example in Sanz’s studies (e.g. 2000,

2008), the participants were from Catalonia, a northeastern region of Spain that is

129

bilingual in nature. In the Catalonia context, some members of society are
Spanish/Catalan bilinguals, whereas some of the members of the society are Spanish
monolingual (these days it is much less common to ﬁnd Catalan monolinguals). The
same concept applies for Cenoz’s work in the Basque country as well as the multilingual
studies that take place in F rench-speaking Canada. In these studies, most of the
participants grew up bilingually and learned other languages at a later age. In contrast,
the participants in this study were raised monolingual and learned all foreign languages
in a classroom setting. These contexts in multilingual societies differ greatly from the
context for the current study, which is an inherently monolingual society.

As a result, it is possible to infer that motivation for L2 and L3 learning is context
speciﬁc.

In most situations, this type of participant homogeneity is an ideal situation for a
researcher, but in this case, it resulted in non-signiﬁcant results for the motivational
proﬁle part of this study. Although surprising, this result is not necessarily to be
considered negative. In many cases, motivation is the factor that dominates the results of
studies involving individual differences. Because of the lack of differentiation in the
motivational proﬁles of the bilingual and multilingual learners, the other factors may be
more salient and account for more in terms of variation among the independent factors

investigated in this study.

130

RQ2: Which group of participants has higher language aptitude, bilingual or
multilingual learners?

H2: Multilingual learners have a higher language aptitude than bilingual

language learners.

For both categorizations of multilingualism used in this study, hypothesis 2 proved to be
correct—the multilingual participants had signiﬁcantly higher language aptitude than did
the bilingual participants. In fact, the answer to this research question is one of the
clearest answers in the entire study; for Categorization 1 of L3, the p-value is .003 and
for Categorization 2, the p-value is .000, illustrating that multilingualism corresponds
with higher language aptitude.

The question is whether a high language aptitude encourages multilingualism
because of the relative success in language acquisition, or on the other hand, whether
studying multiple languages can in fact increase one’s language aptitude. As discussed
in the literature review, language aptitude has traditionally been treated as a stable
cognitive feature, much like IQ. However, part of the theory behind the CANAL-FT is
that language aptitude is in fact dynamic, able to change depending on the experience
that a language learner undergoes. With this idea in mind, are the aptitude scores
influenced by the linguistic backgrounds of the participants, or do their high language
aptitudes help shape their linguistic backgrounds? Such a cyclical question should not
be answered lightly, but an extensive review of the literature on language aptitude and
multilingualism found no other articles discussing the relationship of bilingualism or
multilingualism to language aptitude. There are many articles discussing the inherent

beneﬁts that bilinguals have over monolinguals or even that multilinguals have over

131

 

bilinguals (e.g. Ben-Zeev 1977; Bialystok, 1987, 1988, 1991, 2001a, 2001b, 2006;
Bialystok & Majumder, 1998; Dewaele, 2002; Eviatar & Ibrahim, 2000; Galarnbos &
Goldin-Meadow, 1990; Ianco-Worral, 1972; Jessner, 2005; Malakoff & Hakuta, 1991;
McLaughlin & Nayak, 1989; Nation & McLaughlin, 1986; Nayak et al., 1990; Pandey,
1991; Pennington, 1999; Ricciardelli, 1992a, 1992b; Sanz, 2000, 2008; Swain et al.,
1990; Thomas; 1988; Yelland et al., 1993; Wilson & Sperber, 2006). Despite the
copious research on bilingualism and multilingualism, there is not research comparing
the language aptitude of monolinguals and bilinguals or of bilinguals and multilinguals.
To see the result of language learning on a person’s language aptitude, a different
type of study would have to be designed wherein monolinguals or bilinguals could be
tested before having the learning experience of an additional language. After a period of
study with an additional language, an additional measure of language aptitude could be
taken to measure the possible change in language aptitude as a result of studying an
additional language. The difficulty of this design if using the CANAL-FT as the measure
of language aptitude would be that the participants would have already been exposed to
Ursulu before, potentially causing a test effect. This is an area ripe for exploration and

potential research projects.

RQ3: Which group of participants has higher English language proficiency as measured
by the MSUELT, bilingual or multilingual learners?
H3 : Multilingual participants will have higher English language proficiency

scores as measured by the MSUELT.

132

The analysis of this research question provided some rather interesting results.
For this question, the MSUELT scores of Group 1 (semesters 1 and 2) and Group 3
(semesters 5, 6, and 7) were compared. None of the results were signiﬁcant, meaning that
the bilinguals and the multilinguals were performing similarly on the MSUELT, the
independent measure of English proﬁciency. This result can partly be explained by the
extremely weak eta squared values for this analysis. For an effect size to be considered
large, the eta squared value must be above 0.5. However, the eta squared values for this
analysis were nowhere near this value. For the ﬁrst categorization, the eta squared value
for Group 1 was 0.163 and for Group 3 was .002. For the second categorization, the eta
squared value for Group 1 was .009 and for Group 3 was .004. As can be seen from
these values, the effect size was very small; thus the analysis was asking too much of the
sample. Because of this, it is not surprising that no signiﬁcant results were found.

Although the results are not signiﬁcant, it is informative to look at the standard
deviation scores of this analysis. In all cases, the standard deviations are larger for the
lower-level English learners. Table 34 illustrates the standard deviations for all groups

with both L3 categorizations:

133

Table 34

Summary of Standard Deviations of MSUELT Scores

 

 

L3 Categorization Group number SD
L2 L3
Categorization 1: Any L3 experience 1 11.86 6.44
3 17.88 9.19
Categorization 2: Perceived Positive Language 1 11.22 10.61
Interaction 3 14.88 9.33

 

This wider range of the standard deviation at the L2 level could indicate an
inconsistency in how participants at this level of English are performing on the
MSUELT. Some of the bilingual participants are performing quite well on the
independent measure of English proﬁciency, while others are not performing as well.
The data suggest a trend that the L3 learners are scoring within a tighter range, indicating
a- relative amount of consistency within the L3 group.

Another important point in this part of the analysis is that the MSUELT scores of
the bilingual and multilingual participants could not be compared across all levels,
resulting in the division of the participants into smaller groups. For example, for Group
1, Categorization 1, the L2 group had only 12 participants, and the L3 group only had 7
participants. The numbers are similarly small for the other divisions in the analysis (see

Table 35 for details).

134

Table 35

Participant Numbers within Groups

 

 

L3 Categorization Group number Participants
L2 L3
Categorization 1: Any L3 experience 1 N=12 N=7
3 N=14 N=22
Categorization 2: Perceived Positive Language 1 N=17 N=2
Interaction 3 N=23 N=l3

 

Using such small ntunbers for analyses is far from ideal, and it may be that these
numbers resulted in unreliable ﬁndings. With more participants per level it would be

possible that the results would be different.

RQ4: Which group has a higher retention rate in the progression of the English courses,
bilingual or multilingual learners?
H4 : Multilingual learners have a higher retention rate in advanced English
language courses.
Hypothesis 4 was partially proved correct. Although there were no signiﬁcant
differences between the numbers of L2 versus L3 language learners in Group 1 versus
Group 3 for either of the L3 categorizations, Categorization 1 had a p-value of 0.0992
and Categorization 2 had a p-value of 0.05 82. In this case, the most interesting value

was that of Categorization 2, Perceived Positive Language Interaction. Although this

135

value was not signiﬁcant, it approached signiﬁcance. The idea that there is a higher
retention rate for those students who understand the interaction of all of the languages
that they have studied would not be surprising. Understanding language relationships is
one strategy for successful language acquisition; thus, those students who understand this
fact might be more successful in their language acquisition, encouraging them to
continue learning the language. Although these results are not statistically signiﬁcant,
they represent a salient pattern.

The qualitative analysis provides an insight into the types of language interactions
that the participants found, including lexical, grammatical, and language learning
strategies. Many of the participants also found interactions between Portuguese and the
foreign languages studied. In fact, many of the participants indicated that studying a
foreign language helped them to appreciate the difﬁculty of their mother tongue. As can
be seen the retention rates of bilingual and multilingual learners, those who perceive
these interactions are more likely to stay with the language class, making these perceived
interactions a crucial part of the language learning process.

At this point it is also interesting to note the relationship with research question
one and research question four. Students’ motivation is one of the strongest factors in
dealing with the idea of retention in the language classes. Those students who lack
motivation are less likely to continue taking language courses. It is also true that
motivation can change over time (e. g. Domyei, 2005), and that even those student who
start out being motivated to study a certain language might not have the sustained
motivation needed to reach advanced levels of a language course. What is especially

interesting about the results from research questions one and four is that despite the utter

136

similarity of the bilingual and multilingual groups for research question one, a pattern
emerges in research question four. In research question one, none of the results are even
remotely approaching signiﬁcance, whereas in research question four, there is a clear
pattern of higher retention in language classes according to both categorizations of
multilingualism. These results indicate that though retention in language classes is

linked to student motivation, there are other factors interacting with this retention as well.

137

CHAPTER SIX: CONCLUSION

This dissertation is an exploration of bilingual and multilingual learners in a monolingual
setting. There are several important ﬁndings in this dissertation. First, no difference was
found in the motivational proﬁles of bilingual and multilingual learners, perhaps because
of a shared attitude towards English as a global language for communication and the
monolingual nature of the research setting. Also, unlike other studies involving
multilingualism and motivation, these multilingual participants did not grow up in a
bilingual setting, learning a subsequent language later in life. These participants were
raised monolingual and learned all languages in a classroom setting, making them much
different than the participants in other studies who were raised bilingually. The results of
this study suggest that caution must be used when transferring the results of prior studies
on multilingualism in bilingual settings to other contexts. '

In addition, no signiﬁcant differences were found in the independent measure of
English language proﬁciency, the MSUELT, between the bilingual and multilingual
learners. As discussed above, this could be partly the result of the very small numbers
that resulted from the need to divide the participants into comparable proﬁciency levels.

An interesting ﬁnding was the retention rate in the language classes for the
multilingual learners compared to the bilingual learners. The results for both L3
categorizations approached signiﬁcance, indicating a tendency for the multilingual
learners to have less attrition than the bilingual learners in the advanced levels of the
language classes. Perhaps the most interesting ﬁnding was that of the language aptitude

scores. For both categorizations of multilingualism, the difference in the language

138

aptitude scores was signiﬁcant. This result indicates that multilinguals have a higher
language aptitude than bilinguals.

Taking into consideration the lack of signiﬁcant ﬁndings with the motivational
proﬁles and the signiﬁcant ﬁndings of the language aptitude scores, the concept of the
constructs involved in deﬁning language aptitude are elucidated. It is clear that Carroll’s
(1981) claim that language aptitude and motivation differ is illustrated by these results.
Also true is Carroll’s idea that language aptitude differs from performance, as evidenced
by the results of the third research question. Although the multilingual participants
outperformed the bilingual participants on the language aptitude test, they did not do so
on the independent measure of English proﬁciency. The fact that the multilingual
learners outperformed the bilingual learners in language aptitude also indicates the
dynamic nature of language aptitude, as proposed by Stemberg (2002). The experience
of learning multiple languages could have improved the language aptitude of the
multilingual participants. These results show evidence for both Carroll’s (1981)
construct of deﬁning language aptitude as well as Stemberg’s (2002) argument for
language aptitude dynamic nature.

Finally, the idea of how to deﬁne or categorize multilingual learners is a key
concept in this dissertation. These results indicate that there is indeed an effect for even
very limited experience in a previous language. There were also signiﬁcant differences
when differentiating between those multilinguals as those with Perceived Positive
Language Interaction and those without it. Thus, the effect of a small amount of language
learning and the perceived interaction between languages are two important ideas to

remember regarding studies involving potentially multilingual participants.

139

There are some aspects to language learning that simply cannot be expressed in
numbers or statistics. For those participants who completed the optional interview, there
was a chance at the end for the participants to comment on any aspect of the interview or
of language learning in general. Most participants declined, saying that the interview had
been very complete. However, those who did answer wanted to have their voice heard
about the importance of language learning. Below are some of the most striking

responses:

Example 1:

Eu ainda sou um iniciante no estudo de linguas estrangeiras, enta'o assim
coma a estudo da F isica, matemdtica e outras coisas que eu aprendi na
vidafoi s6 uma caisa mais infarmativa, mais teérica, enﬁm abriu muitas
harizontes para a minha mente...

I’m only a beginner in the study of foreign languages, so the same as the
study of Physics, math, or other things I have learned in life, it was the
most informative and most theoretical; it opened up new horizons for my

mind (Participant 31).

Example 2:
Eu diria que a aprendizado da lingua é ﬁmdamental para vocé abrir a
mente, para a mundo e ter acessa as vdrias expectativas, que nc‘io sa'o

iguais as suas.

140

I would say that learning a language is fundamental for you to open your
mind to the world and to have access to various points of view that aren’t

the same as yours (Participant 54).

Example 3:

Quanta mais as pessoas se entenderem melhor e menos guerra. Parque
todas se entendendo, claro que na'o hd mais guerra. Eu acha seria bem
melhor.

The more people who understand each other the less war there will be
because when everyone understands each other, clearly there will be no

more war. I think that will be a lot better (Participant 55).

Example 4:

Acho que se vocé quer seguir aprendendo como a mundoﬁmciona, coma
existem novas culturas e novas lugares, tem que aprender mesmo uma
lingua.

I think that if you want to learn how the world works, that there are new
cultures and new places, you have to learn another language (Participant

62).

Example 5:
Vocé tem que derrubarﬁonteiras para poder ter essa aproximacc‘io cam

as idiomas e com as pessoas. Eu descobri por acaso umaﬁase escrita nos

141

pés da estatua da liberdade: “Give me your tired, your poor, Your
huddled masses yearning to breathe ﬁee ” isso me ajudou de certaforma
a agir e pensar diferente cam relaca'o a minha profissa'o que é de
policial.... Eu tenha essa noca'a, entt‘iofoi par isso que aprendi inglés... ja
ajudou na minhaforma de agir e pensar em relacdo c‘r diminuicdo do
preconceito que tenha com a pobreza e me ajuda também no trabalha...
ela me ajudou também em relacdo a procura de novas culturas, através
dos idiomas.

You have to break down barriers to be able to have this closeness with
languages and with cultures. I discovered at random the phrase written at
the feet of the Statue of Liberty: “Give me your tired, your poor, your
huddled masses yearning to breathe free” and this helped me to behave
and to think differently in relation to my job, which is a policeman...I have
this idea, and because of this I decided to learn English...It already helped
my behavior and thoughts in relation to decreasing the preconceived
notions that I have about the poor, and it also helped me at work...I was

also able to ﬁnd new cultures through language (Participant 45).

These examples illustrate that learning a new language is not only about learning

grammar structures or memorizing new vocabulary words. Learning a language means

learning a culture, and with this new knowledge, barriers can be broken and insights to a

new world are created. The participants strongly indicated that foreign languages greatly

affected their lives in a variety of positive ways, including the notion learning languages

142

comes with increased opportunities; that language is the portal to other worlds, cultures,
and people, and is something that should be essential in everyone’s life. In order to fully
understand the complexity of language learning in the multilingual brain, researchers
need to take the time to talk to participants for further insights. Only with this
combination of qualitative and quantitative methods will the inner workings of language

acquisition be further elucidated.

143

APPENDIX A

Background Questionnaire

 

Identidade do estudante _
(56 para a investigadora)

 

 

Questiona'rio
Um modelo de aquisicao de idioma: Proﬁciencia adicional do idioma, aptidao, e motivacao.

 

Data:

 

Por favor, preencha a informaca'o:

l. Nome: a. Nome:

 

c. Sobrenome:

 

2.1dade: Data de nascimento:

 

3. Sexo U Masculino D Femenino

4.N1'rmero de telefone: ( ) '

 

5. E-mail:

 

6a. Escolaridade :

 

b. Curso que vocé esta estudando:

 

7. Idioma matemo (lingua nativa):

 

3. Se seu idioma matemo nao é portugués, como vocé aprendeu portugués?

 

 

b. Quantos anos vocé tinha quando vocé comecou a aprender portugués?

8. Idioma que vocé fala em casa:

 

9. Vocé esta aprendendo inglés. Ha outros idiomas que vocé esta aprendendo agora? Quais?:

 

10. For que vocé esta estudando inglés? En outras palavras, quais 850 as suas motivacOes para estudar
inglés? Por favor, seja especiﬁco.

 

 

l 1. Vocé tern planos para continuar seus estudos em inglés? Por que sirn ou por que nao?

 

 

12. Estudou outros idiomas antes?

 

13. Se vocé estudou outros idiomas antes, vocé pensa que isso ajudou ou retardou a aprendizagem do
inglés? Dé exemplos espectﬁcos, por favor.

 

 

144

14. Vocé tem amigos ou alguém na sua fanrilia que fala inglés?

DSim

D Nao

Se sim, explique.

 

 

 

15. Vocé viajou para um pais onde se fala inglés ou qualquer outra lingua que vocé

estuda/estudou?

ElSim

D Nao

Se sim, preencha a informacao abaixo:

 

 

 

 

 

Llngua: Onde vocé Por quanto Quantos Qual era 0 proposito da viagem?
estudou ou tempo? anos tinha?

a. moron?

Llngua: Onde vocé Por quanto Quantos Qual era 0 proposito da viagem?
estudou ou tempo? anos tinha?

b. morou?

Lingua: Onde vocé Por quanta Quantos Qual era 0 propésito da viagem?
estudou ou tempo? anos tinha?

c. morou?

Lingua: Onde vocé Por quanto Quantos Qual era 0 prop681to da viagem?
estudou ou tempo? anos tinha?

d. morou?

 

 

 

 

 

145

 

Por favor, preencha a informacao abaixo sobre as linguas que vocé estudou ou que vocé

 

 

 

 

 

 

 

 

 

 

 

 

 

 

esta estudando.
Lingua Como aprendeu a Desde qual idade Como vocé acha que fala a lingua? (Escolha
A lingua? (Escolha ate qual idade vocé so uma altemativa)
todos que sao estudou a lingua?
pertinentes) 1. mal
1. Familia a 2. mais ou menos
(Quem?
) 3. bem
2. Escola de idiomos 4. muito bem
3. Colegio 5. ﬂuente
4. Universidade 6. como nativo
5. Outro:
Lingua Como aprendeu a Desde qual idade Como vocé acha que fala a lingua? (Escolha
B lingua? (Escolha ate qual idade vocé so uma altemativa)
todos que sao estudou a lingua?
pertinentes) 1. mal
1. F amilia a 2. mais ou menos
(Quem?
) 3. bem
2. Escola de idiomos 4. muito bem
3. Colegio 5. ﬂuente
4. Universidade 6. como nativo
5. Outro:
Lingua Como aprendeu a Desde qual idade Como vocé acha que fala a lingua? (Escolha
C lingua? (Escolha ate qual idade vocé so uma altemativa)
todos que sao estudou a lingua?
pertinentes) 1. mal
1. F amilia a 2. mais ou menos
(Quem?
) 3. bem
2. Escola de idiomos 4. muito bem
3. Colegio 5. ﬂuente
4. Universidade 6. como nativo
5. Outro:
Obrigada!

146

 

 

BIBLIOGRAPHY

Al-Fallay, I. (2006). The role of self-assessment in modifying some psychological and
personality traits of foreign language learners. Arab Journal for the Humanities,
24(93), 239-256

Alderson, J. C., & McIntyre, D. (2006). Implementing and evaluating a self-assessment

mechanism for the web-based Language and Style Course. Language and
Literature, 15(3), 291-306.

Bachman, L. F. (1990). Fundamental considerations in language testing. Oxford: Oxford
University Press.

Bandura, A. (1986). Social Foundations of thought and action: A social cognitive theory.
Englewood Cliffs, NJ: Prentice Hall.

Barke, EM. (1933). A study of the comparative intelligence of children in certain
bilingual and monoglot schools in South Wales. British Journal of Education
Psycology 3, 237-50.

Ben-Zeev, S. (1977). Mechanisms by which childhood bilingualism affects
understanding of language and cognitive structures. In P.A. Homby (Ed.),
Bilingualism: Psychological, Social and Educational Implications. (pp. 29-55)
New York: Academic.

Bialystok, E. (1986). Factors in the growth of linguistic awareness. Child Development,
57, 498 — 510.

Bialystok, E. (1987). Words as things: Development of word concept by bilingual
children. Studies in Second Language Learning, 9, 133 — 140.

Bialystok, E. (1988). Levels of bilingualism and levels of linguistic awareness.
Developmental Psychology, 24, 4, 560-567.

Bialystok, E. (1991). Language processing in bilingual children. Cambridge: Cambridge
University Press.

Bialystok, E. (2001a). Bilingualism in Development, Language, Literacy and Cognition.
Cambridge: Cambridge University Press.

Bialystok, E. (2001b). Metalinguistic aspects of bilingual processing. Annual Review of
Applied Linguistics, 21, 169-181.

147

Bialystok, E. (2006). The impact of bilingualism on language literacy development. In
T.K. Bhatia & W.C. Ritchie (Eds), The Handbook of Bilingualism (pp. 577-
601). Malden, MA: Blackwell.

Bialystok, E. & Majumder, S. (1998). The relationship between bilingualism and the
development of cognitive processes in problem solving. Applied
Psycholinguistics, 19, 69-85.

Birckbichler, D., Car], K., & Deville, C., (1993). The dynamics of language program
testing: Implications for articulation and program revision. The Dynamics of
Language Program Direction. Boston, MA: Heinle & Heinle.

Blanche, P. & Merino, B. (1989). Self-assessment of foreign language skills:
Implications for teachers and researchers. Language Learning, 39, 313-40.

Bonk, W. J ., & Ockey, G. J. (2003). A many-facet Rasch analysis of the second language
group oral discussion task. Language Testing, 20(1), 89-110.

Brantmeier, C. (2006). Advanced L2 learners and reading placement: Self-assessment,
CBT, and subsequent performance. System, 34(1), 15-35.

Brantmeier, C. (2005). Non-linguistic variables in advanced L2 reading: learner’s self-
assessment and enjoyment. Foregin Language Annals, 38(4), 493-503.

Brown, J .D., Cunha, M., I. A., Frota, S. de F. N., & Ferreira, A. B. F. (2001). The
development and validation of a Portuguese version of the motivated strategies f
or learning questionnaire. In Z. Damyei & R. Schmidt (Eds), Motivation and
Second Language Learning (pp. 1-20). Honolulu: University of Hawaii Press.

Carroll, J .B. (1962). The prediction of success in intensive foreign language training.
In R. Glaser (Ed.), Training research in education (pp. 87 - 136). Pittsburgh,
PA: University of Pittsburgh Press.

Carroll, J.B. (1981). Twenty-ﬁve years of research on foreign language aptitude. In
K.C. Diller (Ed.), Individual differences and urriversals in language learning
aptitude, (pp.83 — 118). Rowley, MA: Newbury House.

Carroll, J .B. (1990). Cognitive abilities in foreign language aptitude: Then and now.
In T. Parry & C. Stansﬁeld (Eds), Language aptitude reconsidered (pp. 11 —
19). Englewood Cliffs, NJ: Prentice Hall.

Carroll, J .B., & Sapon, S. (1959). Modern Language Aptitude Test (MLAT) manual.
San Antonio, TX: Psychological Corp.

Cenoz, J. (2001). The effect of linguistic distance, L2 status and age on cross-
linguistic inﬂuence in third language acquisition. In J. Cenoz, B. Hufeisen, &

148

U. Jessner, (Eds), Cross-linguistic inﬂuence in third language acquisition:
Psycholinguistic perspectives. (pp. 8-20) Clevedon: Multilingual Matters.

Cenoz, J. (2003). The additive effect of bilingualism in third langauge acquisition: a
review. International Journal of Bilingualism, 7(1), 71-87.

Cenoz, J ., & Valencia J.F. (1994). Additive trilingualism: Evidence from the Basque
Country. Applied Psycholinguistics, 15, 195 — 207.

Clyne, M. (1997). Some of the things trilinguals do. The International Journal of
Bilingualism, 1(2), 95-116.

Clyne, M., & Cassia, P. (1999). Trilingualism, immigration and relatedness of
languages. I.T.L. Review of Applied Linguistics, 123-124, 57-74.

Cook, G. (2001). Investigating growth trajectories on English as a second language
listening and reading comprehension test. Unpublished doctorial dissertation.
Michigan State University, East Lansing.

Cortina, J .M. (1993). What is coefficient alpha? An examination of theory and
applications. Journal of Applied Psycology, 78, 98-104.

Crookes, G., & Schmidt, R. (1991). Motivation: Reopening the research agenda.
Language Learning, 41, 469-512.

Cummins, J. (1976). The inﬂuence of bilingualism con cognitive growth: A synthesis of
research ﬁndings and explanatory hypotheses. Working Papers on Bilingualism,
9, 1-43.

Cummins, J. (1979). Linguistic interdependence and the educational development of
bilingual children. Review of Educational Research, 49(2), 222-251.

Cummins, J. (2001). Instructional conditions for trilingual development. International
Journal of Bilingual Education and Bilingualism, 4(1), 61-75.

Da Fontoura, H. & Siegel, LS. (1995). Reading, syntactic and working memory skills of
bilingual Portuguese-English Canadian children. Reading and Writing, 7(1), 139
— 153.

De Angelis, G. (1999). Interlanguage transfer and Multiple Language Acquisition: a case
study. Paper presented at TESOL 1999, New York City.

De Angelis, G. (2005a). Interlanguage transfer of frmction words. Language Learning,
55(3), 379-414.

149

De Angelis, G. (2005b). Multilingualism and non-native lexical transfer: an identiﬁcation
problem. International Journal of Multilingualism, 2(1), 1-25.

De Angelis, G. (2005c). The acquisition of languages beyond the L2: psycholinguistic
perspectives. Rassegna Italiana de Linguistica Applicata, 2-3, 397-409.

De Angelis, G. (2007). Third or Additional Language Acquisition. Clevedon:
Multilingual Matters.

Deci, EL. (1980). The psychology of self-determination. Lexington, MA: D.C. Heath.

Deci, E.L. & Moller, AC. (2005). The concept of competence: A starting place for
understanding intrinsic motivation and self-determined extrinsic motivation. In
A.J. Elliot & C.S. Dweck (Eds), Handbook of competence and motivation
(pp.579-597). New York: Guilford Press.

Deci, E.L. & Ryan, RM. (1985). Intrinsic motivation and self-deterrrrination in human
behavior. New York: Plenum.

Deci, E.L. & Ryan, RM. (1991). A motivational approach to self: Integration in
personality. In R.A. Dienstbier (Ed.), Nebraska symposium on motivation 1990
(Vol. 38, pp. 23 7-288). Lincoln, NE: University of Nebraska Press.

DeKeyser, R. (1995). Learning second language grammar rules: An experiment with
a miniature linguistic system. Studies in Second Language Acquisition, 17,
379 — 410.

DeKeyser, R. (2000). The robustness of critical period effects in second language
acquisition. Studies in Second Language Acquisition, 22(4), 499 — 533.

Dewaele, J -M. (2002). The effect of multilingualism and socio-situational factors on
communicative anxiety of mature language learners. In J. Ytsma & M.
Hooghiemstra (Eds), Proceedings of the Second International Conference on
Trilingualism, Leeuwaarden: Fryske Akadenrie (CD Rom).

Diaz, RM. (1985). Bilingual cognitive development: addressing three gaps in current
research, Child Development, 56, 1356-1378.

Donitsa-Schmidt, S., Inbar, O. & Shohamy, E. (2004). The effects of teaching spoken
Arabic on students’ attitudes and motivation in Israel. The Modern Language
Journal, 88, 217-228.

Domyei, Z. (2005). The psychology of the language learner Mahwah, NJ: Lawrence
Erlbaum. ‘

150

Dbmyei, Z., & Csizér, K. (2001). Some dynamics of language attitudes and motivation:
Results of a longitudinal nationwide study. Applied Linguistics, 23(4), 421-462.

Domyei, Z., Csizér, K., & Németh, N. (2006). Motivation, Language Attitudes and
Globalization: A Hungarian perspective. Buffalo: Multilingual Matters.

Ecke, P. (2001). Lexical retrieval in a third language: evidence from errors and tip-of-
tlre-tongue states. In J. Cenoz, B. Hufeisen, and U. Jessner (Eds), Cross-linguistic
Inﬂuence in Third Language Acquisition: Psycholinguistic Perspectives (pp.90-
114). Clevedon: Multilingual Matters.

Edwards, J. (2003). The Importance of Being Bilingual. In J .M. Dewaele, A. Housen, &
L. Wei (Eds), Bilingualism: Beyond Basic Principles. (pp. 28-42). Buffalo:
Multilingual Matters.

Ehrman, ME. (1996). Understanding second language learning difficulties. Thousand
Oaks, CA: Sage.

Ehrman, M. (1998). The Modern Language Aptitude Test for predicting learning
success and advising students. Applied Language Learning, 9, 31 — 70.

Ehrman, M.E., & Oxford, R.L. (1995). Cognition plus: Correlates of language learning
success. Modern Language Journal, 79, 67-89.

Ekbatani, G. (2000). Moving toward leamer-directed assessment. In G. Ekbatani & H.
Pierson (Eds), Learner-directed assessment in ESL. (pp. 1-11). Mahwah, NJ:
Erlbaum.

Ellis, N. (1994). Consciousness in Second Language Learning: Psychological
Perspectives on the Role of Conscious Processes in Vocabulary Acquisition.
AILA Review, 11, 37-56.

Evitar, Z., & Ibrahim, R. (2000). Bilingual is as bilingual does: metalinguistic abilities of
Arabic-speaking children. Applied Psycholonguistics, 21, 451-471.

Field, A. (2005). Discovering Statistics Using SPSS. London: Sage Publications.

Gajar, A. (1987). Foreign language learning disabilities: The identiﬁcation of
predictive and diagnostic variables. Journal of Learning Disabilities, 20, 327
- 330.

Galarnbos, J .S., & Goldin-Meadow, S. (1990). The effects of learning two languages on
levels of metalinguistic awareness. Cognition, 34, 1-56.

Gardner, R.C., (1985). Social psychology and the role of language learning: The role of
attitude and motivation. London: Edward Arnold.

151

 

Gardner, RC. (2001). Integrative motivation and second language acquisition. In Z.
Domyei & R. Schmidt (Eds), Motivation and Second Language Learning (pp. 1-
20). Honolulu: University of Hawaii Press.

Gardner, R.C., & Lambert, W.E. (1965). Language aptitude, intelligence, and second
language achievement. Journal of Educational Psychology, 56, 191-199.

Geeslin, K. L. (2003). Student self-assessment in the foreign language classroom: The
place of authentic assessment instruments in the Spanish language classroom.
Hispania, 86(4), 857-868.

Geva, E. Wade-Woolley, L., & Shany, M. (1997). Development of reading efficiency
in ﬁrst and second language. Scientiﬁc Studies of Reading, 1, 119-144.

Goldberg, E, & Noels, K. (2006). Motivation, Ethnic Identity, and Post-Secondary
Education Language Choices of Graduates of Intensive French Language
Program. The Canadian Modern Language Review, 62(3), 423-447.

Grigorenko, E. Stemberg, R. & Ehrman, M. (2000). A theory-based approach to the
measurement of foreign language learning ability: The CANAL-F theory and test.
Modern Language Journal, 84, 390-405.

Group, T. C. (2000). TOEIC can-do guide: Linking TOEIC scores to activities performed
using English [Electronic Version]. Retrieved January 4, 2007 from
https://www.ets.org/Media/Research/pdf/TOEIC_CAN_DO.pdf.

Hargan, N. (1994). Learner autonomy by remote control. System 22(4), 45 5-503.

Harris, M. (1997). Self-assessment of language learning in formal settings. ELT Journal,
51(1), 12-20.

Hasselgreen, A. (2005). Assessing the language of young learners. Language Testing,
22, 337-354.

Heilenman, K. (1991). Self-assessment and placement: A review of the issues. In R.V.
Teschner (Ed.), Assessing Foreign Language Proﬁciency of Undergraduates,
AAUSC Issues in Language Program Direction (pp. 93-114). Heinle & Heinle,
Boston.

Home, K.M. (1971). Differential prediction of foreign language testing. Paper presented
before the Bureau of International Language Coordination, London.

Hudelson, S. (1987). The role of native language literacy in the education of language
minority children. Language Arts 64, 826 — 41.

152

Ianco-Worral, A. (1972). Bilingualism and cognitive development. Child Development,
43, 1390-1400.

Inbar, 0., Donitsa-Schmidt, S., & Shohamy, E.. (2001). Student’s motivation as a
function of language learning: The teaching of Arabic in Israel. In Z. Domyei &
R. Schmidt (Eds), Motivation and Second Language Acquisition (pp. 149-169).
Honolulu: University of Hawai’i Press.

J essner (2005). Multilingual metalanguage, or the way multilinguals talk about their
languages. Language Awareness, 14(1), 56-68.

Kanfer, F.H., & Gaelick, L. (1986). Self-management methods. In F.H. Kanfer & A.P.
Goldstein (Eds), Helping people change: A textbook of methods (3rd ed., pp.
283- 345). New York: Pergamon.

Kellerman, E. (1977). Towards a characterization of the strategy of transfer in second
language learning. Interlanguage Studies Bulletin, 2, 58-145.

Kellerman, E. (1978). Giving Learners a break: native language intuitions as a source of
prediction about transferability. Working Papers on Bilingualism, 15, 59-92.

Kellerman, E. (1979). Transfer and non-transfer: where are we now? Studies in Second
Language Acquisition, 2, 37-57.

Kellerman, E. (1983). Now you see it, now you don’t. In S. Gass & L. Selinker (Eds),
Language Transfer in Language Learning (pp. 112-134). Rowly, MA: Newbury
House.

Kemp, C. (2001). Metalinguistic Awareness in Multilinguals: Implicit and Explicit
Grammatical Awareness and its Relationship with Language Experience and
Language Attainment. Unpublished Ph.D. Thesis, University of Edinburgh.

Keshavarz, M.H., & Astaneh, H. (2004). The impact of bilinguality on the learning of
English vocabulary as a foreign language (L3). Bilingual Education and
Bilingualism, 7(4), 295-303.

Kline, P. (1999). The Handbook of Psychological Testing (2"4 ed) London: Routledge.

Klein, EC. (1995). Second versus third language acquisition: Is there a difference?
Language Learning, 45, 419 —465.

Kollberg, P. (1998). S-notation - a computer-based method for studying and representing

text composition. TRITA-NA-P9808. Licentiate thesis, Royal Institute of
Technology, Stockholm.

153

‘~'s‘..§"

 

Lim, H. Y. (2002). The interaction of motivation, perception, and environment: One EFL
learner’s experience. Hong Kong Journal of Applied Linguistics, 7(2), 91-106.

Little, D. (2005). The Common European Framework and the European Language
Portfolio: Involving learners and their judgments in the assessment process.
Language Testing, 22(3), 321-336.

Malabonga, V., Kenyon, D. M., & Carpenter, H. (2005). Self-Assessment, preparation

and response time on a computerized oral proﬁciency test. Language Testing,
22(1), 59-92.

Malakoff, M., & Hakuta, K. (1991). Translation skill and metalanguage in bilinguals. In
E. Bialystok (Ed.), Language Processing in Bilinguals (pp. 141-166). Cambridge:
Cambridge University Press.

McKeachie, W. J ., Pintrich, P.R., & Lin, Y. (1985). Teaching learning strategies,
Educational Psychologist, 20, 153 — 160.

McLaughlin, B., & Nayak, N. (1989). Processing a new language: Does knowing other
languages make a difference? In H.W. Dechert & M. Raupach (Eds),
Interlingual Processes. (pp. 5 — 16). Tllbingen: Gunter Narr.

Nation, R., & McLaughlin, B. (1986). Novices and experts: An information
processing approach to the good language learner problem. Applied
Psycholinguistics, 7, 41 — 56.

Nayak, H., Hansen, N., Krueger, N., & McLaughlin, B. (1990). Language learning
strategies in monolingual and multilingual adults. Language Learning, 40, 221 —
244.

Nikolov, M. (2001). A study of unsuccessful language learners. In Z. Domyei & R.
Schmidt (Eds), Motivation and Second Language Acquisition (pp. 149-169).
Honolulu: University of Hawai’ 1 Press.

Noels, K. (2005). Orientations to Learning German: Heritage Language Learning and
Motivational Substrates. The Canadian Modern Language Review, 62 (2), 285-
3 12.

Noels, K. (2003). Learning Spanish as a second language: Learners’ orientations and
perceptions of their teachers’ communication style. Language Learning, 53, 97-
136.

Noels, K., Pelletier, L., Clement, R., & Vallerand, R. (2000). Why are you learning a

second language? Motivational orientations and self-determination theory.
Language Learning, 50(1), 57-85.

154

Nunan, D. (1989). The leamer-centered curriculum. Cambridge: Cambridge University
Press.

Otto, I. (1996). Hungarian language aptitude test: Words in sentences. Budapest:
Department of English Applied Linguistics, Eotvos University.

Pandey, P. (1991). A psycholinguistic study of democratic values in relation to mono-,
bi- and trilingualism. Psycholingua 21, 111-1 13.

Parry, T.S., & Child, J .R. (1990). Preliminary investigation of the relationship between
VORD, MLAT, and language proﬁciency. In T. Parry & C.W. Stansﬁeld (Eds),
Language aptitude reconsidered. (pp. 33-60). Engkewood Cliffs, NJ: Prentice
Hall.

Peal, E., & Lambert, W. (1962). The relation of bilingualism to intelligence.
Psychological Monographs 76 (Whole No. 546).

Peirce, B. N., Swain, M., & Hart, D. (1993). Self-assessment, French immersion, and
locus of control. Applied Linguistics, 14(1), 25-42

Pennington, M. (1999). Equivilance classiﬁcation in language transfer. Paper presented
at TESOL 1999, New York.

Petersen, C.R., & Al-Haik, AR. (1976). The development of the Defense Language
Aptitude Battery (DLAB). Educational and Psychological Measurement, 6, 369-
380.

Pimsleur, P. (1966). The Pimsleur Language Aptitude Battery. New York: Harcourt
Brace Jovanovich.

Pintrich, P. R. (1986a). Anxiety, motivated strategies and student learning. Paper

presented at the International Congress of Applied Psychology, Jerusalem, Israel,
July, 1986.

Pintrich, P.R. (1986b). Motivation and learning strategies interactions with achievement.
Paper presented at the American Educational Research Association convention,
San Francisco, April, 1986.

Pintrich, PR. (1989). The dynamic interplay of student motivation and cognition in the
college classroom. In M. Mathr & C. Ames (Eds), Advances in motivation and
achievement: Motivation enhancing environments, Volume 6 (pp. 117-160).
New York: JAI Press.

Rao, N., & Sachs, J. (1999). Conﬁrmatory factor analysis of the Chinese version of the

motivated strategies for learning questionnaire. Educational and Psychological
Measurement, 59, 1016-1029.

155

Reeve, J ., Deci, E.L., & Ryan, RM. (2004). Self-determination theory: A dialectal
framework for understanding sociocultural inﬂuences on student motivation. In
D.M. McInemey & S. Van Etten (Eds), Big theories revisited (pp. 31-60).
Greenwich, CT: Information Age Publishing.

Ricciardelli, L.A. (1992a). Bilingualism and cognitive development in relation to
threshold theory. Journal of Psycholinguistics Research, 21, 301-16.

Ricciardelli, L.A. (1992b). Creativity and bilingualism. Journal of Creative Behavior, 26,
242-254.

Richards, K. (2003). Qualitative Inquiry in TESOL. New York: Macmillan.

Ringbom, H. (1967). The Role of the First Language in Foreign Language Learning.
Clevedon: Multilingual Matters.

Ringbom, H. (2003). If you know Finnish as L2, there will be no major problem learning
Swahili. Paper presented at the Third International Conference on Trilingualism
and Third Language Acquisition, Tralee, Ireland, 4-6 September.

Roever, C. (2000). Web-based language testing. Language Learning Technology, 5(2),
84-94.

Rosier, P., & F arella, M. (1976). Bilingual Education at Rock Point — some early results.
TESOL Quarterly 10, 379 — 88.

Ross, S. (1998). Self-assessment in second language testing: a meta-analysis and analysis
of experimental factors. Language Testing, 15 (1), 1-20.

Ryan, R. M. & Deci, EL. (2000). Intrinsic and extrinsic motivations: Classic deﬁnitions
and new directions. Contemporary Educational Psychology, 25, 54-67.

Sanders, M., & Meijers, G. (1995). English as L3 in the elementary school. ITL Review
of Applied Linguistics 107-8, 59-78.

Sanz, C. (2000). Bilingual education enhances third language acquisition: Evidence from
Catalonia. Applied Psycholinguistics, 21, 23 — 44.

Sanz, C. (2008). Predicting Enhanced L3 learning in bilingual contexts: The role of
biliteracy. In C. Perez-Vidal, M. Juan-Garau & A. Bel (Eds), A Portrait of the
Young in the New Multilingual Spain (pp. 220-240). Buffalo: Multilingual
Matters.

Schlesinger, A.J. (1991). The Disuniting of America. New York: W.W. Norton.

156

Schneider, G., & North, B. (2000). Fremdsprachen konner — was heisst das? Chur and
Zurich: Riiegger.

Schunk, DH. (1994). Self-regulation of self-efficacy and attributions in academic
settings. In D.H. Schunk & B.J. Zimmerman (Eds), Self-regulation of learning
and performance: Issues and educational applications (pp. 75-79). Hillsdale, NJ:
Erlbaum.

Schunk, DH. (1995). Self-efﬁcacy and education and instruction. In J .E. Maddux (Ed.),
Self-efﬁcacy, adaptation, and adjustment: Theory, research, and application (pp.
281-303). New York: Plenum Press.

Schunk, D.H., Pintrich, P.R., & Meece, J. L. (2008). Motivation in Education: Theory,
Research, and Applications. Upper Saddle River, NJ: Pearson.

Schunk, D.H. & Sirnrnerman, B]. (2003). Self-regulation and learning. In W.M.
Reynolds &. G.E. Miller (Eds), Handbook of psychology Vol. 7: Educational
psychology (pp. 59-78). Hoboken, NJ: Wiley.

Selinker, L., & Baumgartner-Cohen, B. (1995). Multiple Language Acquisition: ‘Damn
it, why can’t I keep these two languages apart?’ Language, Culture, and
Curriculum, 8(2), 115-121.

Severinson Eklundh, K., & Kollberg, P. (1996). A computer tool and framework for
analyzing online revisions. In C.M. Levy and S. Randsdell (Eds), The Science of
Writing (pp. 163-188). New Jersey: Lawrence Erlbaum.

Sikogukira, M. (1993). Inﬂuence of languages other than the L1 on a foreign language: a
case of transfer from L2 to L3. Edinburgh Working Papers in Applied Linguistics
4, 110-132.

Skehan, P. (1986). The role of foreign language aptitude in a model of school learning.
Language Testing, 3, 188-221.

' Skehan, P. (1990). The relationship between native and foreign language ability:
Educational and linguistic factors. In H. Dechert (Ed.), Current trends in
European second language acquisition research (pp. 83 — 106). Clevedon:
Multilingual Matters.

Skehan, P. (2002). Theorising and updating aptitude. In P. Robinson (Ed.), Individual
differences and instructed language learning (pp. 69-93). Philadelphia, PA:
John Benjamins.

Stansﬁeld, C. W. and Reed, D. J. (2004) “An Interview with John B. Carroll: The Story

behind the Modern Language Aptitude Test.” Language Assessment Quarterly
1,1.

157

Stemberg, R.J. (2002). The theory of successful intelligence and its implications for
language aptitude testing. In P. Robinson (Ed.), Individual Differences and
Instructed Language Learning (pp. 13-43). Amsterdam: John Benjamins.

Sullivan, K., & Lindgren, E. (2002). Self-assessment in autonomous computer-aided
second language writing. ELT Journal, 56(3), 258-266.

Swain, M., Lapkin, S., Rowen, N., & Hart, D. (1990). The role of mother tongue literacy
in third language learning. Language, Culture and Curriculum, 3(1), 65 — 81.

Thomas, J. (1988). The role played by metalinguistic awareness in second and third
language learning. Journal of Multilingual and Multicultural Development, 9,
235 — 246.

Thomas, J. (1992). Metalinguistic awareness in second- and third-language learning.
In R.J. Harris (Ed.), Cognitive processing in bilinguals (pp. 531 — 545).
Amsterdam: North-Holland.

Thompson, A. (2008). Prominent Factors in the Acquisition of Portuguese: Language
Aptitude versus Previous Language Experience. In J .B. Garavito & E. Valenzuela
(Eds), Selected Proceedings of the 10th Hispanic Linguistic Symposium (pp. 134-
145). Somerville, MA: Cascadilla Press.

Valencia, J .F. & Cenoz, J. (1992). The role of bilingualism in foreign language
acquisition. Journal of Multilingual and Multicultural Development, 13(5), 433-
449.

VanPatten, B. (1996). Input processing and grammar instruction. Norwood, NJ: Ablex.

van Gelderen, A., Schoonen, R., de Glopper, K., Hulstijn., Snellings, Simis., &

Stevenson, M. (2003). Roles of linguistic knowledge, metacognitive knowledge and
processing speed in L3, L2 and L1 reading comprehension: A structural equation
modeling approach. The International Journal of Bilingualism, 7(1), 7-25.

von Elek, T. (1987). A test of Swedish as a second language: An experiment in self-
assessment. In Y. Lee, A. Fok, R. Lord, & G. Low (Eds), New Directions in
Language Testing. (pp. 47-57) Oxford: Oxford University Press.

Ullrnan, M. T. (2001). The neural basis of lexicon and grammar in ﬁrst and second
language: The declarative/procedural model. Bilingualism: Language and
Cognition, 4(1), 105-122.

Wagner, D.A., Spratt, J .E., & Ezzaki, A. (1989). Does learning to read in a second

158

language always put the child at a disadvantage? Some counterevidence from
Morocco. Applied Psycholinguistics, 10, 31-48.

Wei, L. (2001). The Bilingualism Reader. New York: Routledge

Weiner, B. (2000). Intrapersonal and interpersonal theories of motivation from an
attributional perspective. Educational Psychology Review, 12, 1-14.

Weiner, B. (1992). Human Motivation: Metaphors, theories, and research. Newbury
Park, CA: Sage.

White, R.W. (1959). Motivation reconsidered: The concept of competence.
Psychological Review, 66, 297-333.

Williams, S., & Harnmarberg, B. (1998). Language switches in L3 production:
implications for a polyglot speaking model. Applied Linguistics, 19(3), 295-333.

Wilson, D., & Sperber, D. (2006). Relevance theory. In G. Ward & L. Horn (Eds),
Handbook of Pragmatics (pp. 607-632). Oxford, Blackwell.

Yelland, G., Pollard, J ., & Mercury, A. (1993). The metalinguistic beneﬁts if limited
contact with a second language, Applied Psycholinguistics, 14, 423 — 444.

Zimmerman, B.J. (1990). Self-regulating academic learning and achievement: The

emergence of a social cognitive perspective. Educational Psychology Review, 2,
173 -201.

159

1

062 6182

000

0000

1293 03