25 years ago, Peter Skehan [1] affirmed that, unlike in other disciplines like psychology,
“a robust ID (individual differences) tradition” was somewhat lacking in the field of second
language learning. Recent volumes such as the one edited in 2012 by Pawlak for Springer
[2], however, show that this line of enquiry is in very good health.
The role played by learners’ individual differences in language learning is paramount
to understand academic success. This concern is especially present and even urgent
in our country where it seems to be generally agreed that, despite the fact that English
is a part of the curriculum since Infant to Secondary education, students do not achieve
a proficiency level in this foreign language high enough to be able to communicate with
sufficient correction and appropriateness. No wonder, then, that authors like Romero [3]
have claimed that:
La situación actual es alarmante: según estudios recientes, de los 27 países de la Unión,
España ocupa el lugar 23 en aprendizaje de idiomas, superado por países tales como Hungría,
Bulgaria o Rumania. Incluso Portugal, nuestro eterno vecino, nos supera en el apartado del
aprendizaje del inglés como principal lengua extranjera.
Something must be wrong in our approach to the teaching/learning of English as a
foreign language in Spanish instructional settings and this failure is possibly related to
methodological aspects.2
We fully agree with Portolés [4] when she states that “language is not a predictable
process since there is not any magic potion with the clue for success. [...] All in all, we also
believe that exploring individual factors might help to define the traits of the “Good
Language Learner”.
Theoretical framework
Skehan [1] states that aptitude is still “the most successful predictor of language
learning success and failure. To include some attempt to measure aptitude will render the
interpretation of any research study very difficult”. What is more, authors such as Kiss &
Nikolov [5] warn that although aptitude and motivation have been widely analysed in the
field of second language acquisition, “few studies have been devoted to the role of cognitive
variables in foreign language learning contexts in the case of young learners”, pointing to a
clear research gap.
As far as motivation is concerned, nobody doubts of its relevance
among the factors related to achievement in a second or a foreign
language [6]. In fact, motivation is the most abstract variable we have
measured and motivational factors have a special importance in our
research. Among the seventeen independent variables considered in
this study, eleven belong to motivational aspects (Interest in Foreign
Languages, Parental Encouragement, Motivational Intensity, Anxiety,
English Teacher Evaluation, Attitudes toward Learning English,
Attitudes toward English-speaking people, Integrative Orientation,
Desire to Learn English, English Course Evaluation and Instrumental
Orientation). According to Madrid & Pérez Cañado [7], motivation
is normally defined by psychologists as “the set of processes which
involve the arousal, direction, and sustaining of behaviour (conduct).
It is employed to indicate, for instance, a subject’s persistence and
his/her pervasive work on certain tasks and not on other activities”.
Among those eleven aspects of motivation, we distinguish
between integrative and instrumental motivation. Following Skehan
[1], we consider that these two kinds of motivational factors are
different in nature; integrative motivation is rooted in the personality
and it is not susceptible to external changes of learning conditions
such as a new textbook or a new teacher, although instrumental
motivation could be affected by external changes.
Over the years, a major assumption has underlined research
about motivation and it is based on Gardner & Lambert’s investigation
[8,9] where it was hypothesised that an instrumental motive is less
effective to guarantee language achievement because it is not rooted
in the personality of the learner, but more dependent on fallible
external pressures. Bearing this caveat in mind, in our study, we have
included integrative motivation into the group of learners’ internal
affective characteristics and instrumental motivation into the group
of external influences on learners.
Current research focused on EFL instructional settings [10,11],
however, proposes that, although integrativeness is still very
influential, “more attention should be directed towards what goes
on in the language classroom and in changes over time” [12] This
opens the possibility of identifying ways for teachers to enhance
their students’ motivation. In fact, when paying attention to those
studies focusing on younger learners’ motivation, we can see how
instrumental motivation gains importance over integrative one.
Nikolov [13] carried out a longitudinal study about the motivational
patterns of three groups of children (aged 6-14) in Pécs, Hungary,
between 1977 and 1995. The results showed that the participants
in her sample were mostly motivated by factors related to the class
situation such as positive attitudes towards the teacher, the materials
used or the activities and tasks done in the English class. Thus, no
traces of integrative motivation were detected. In contrast, Clément
et al.’s [4] pioneer research with secondary school students (aged
14-18), which was also conducted in Hungary, found, more in line
with Gardner’s Canadian studies [15-17], that integrativeness
was still predominant being the elements with a strong impact on
achievement those of attitudes towards the British/the Americans,
identification orientation or motivational intensity. The comparison
of the findings in these two studies suggest that age in the case of
young learners might also play a role in the kind of motivational
orientation with a higher impact on language achievement. The fact
that the participants in our study are from 12 to 15 years old makes
our sample a very suitable one to explore this hypothesis.
In spite of having defined aptitude and motivation as the most
influential individual variables, language learning strategies have
also received lots of attention. In 1990, Oxford proposed a test (The
Strategies Inventory for Language Learning, also known as SILL)
consisting of 80 questions for English-speaking students learning
a new language, and 50 for speakers of other languages in which
respondents are requested to rate how frequently they make use
of each technique. Both versions provide information about six
different behaviours, coined as memory, cognitive, compensation,
metacognitive, affective and social strategies. According to Gardner
[12], as resorting to such strategies is “a conscious process”, this individual factor cannot be separated from other variables such as (i)
motivation, attitudes or beliefs, (ii) personality; or (iii) gender.
Precisely, in both Sunderland’s [18,19] and Ehrlich’s [20] views,
three of the areas clearly connected with individual variables in
which gender is relevant in the ESL/EFL classroom are those of
(i) learning styles and strategies, (ii) attitudes toward the target
language and culture; and (iii) perceived career opportunities
created by the acquisition of the target language. This link raises the
following two questions: are there two ways of learning a second or
foreign language according to the learner’s gender? Is there a higher
motivation among girls to study foreign languages? Are girls more
strategically oriented than boys when learning a foreign language?
This stronger motivation has been documented in contexts so
different as high-school and university students learning Spanish as a
foreign language in the United States of America [21,22], EFL Spanish
secondary students [23] non-immersion Canadian girls learning
French [24], or high-school students learning French as a foreign
language in the United Kingdom [25]
Among the reasons put forward to explain female students’
higher motivation and better grades as far as foreign language
learning is concerned, we find firstly, girls and women’s propensity
for strategic use when learning a foreign language [22,23]; secondly,
the higher social-orientation in the case of non-immersion Canadian
female students of French in terms of travel, knowledge and personalachievement goals [24]; and, finally, stronger peer, teacher, guidance
counsellor, and parental support, especially on the part of middleclass families, who favour girls studying languages and boys pursuing
science subjects [26,27].
Since research conducted by Oxford and her associates [28-35]
established that gender does have a significant impact on students’
strategies aimed at learning a language, a number of studies have
explored the effect of this variable on second or foreign language
learning strategies, but without achieving conclusive results. For
example, the advantage of EFL female students over male ones is
suggested, with respect to EFL vocabulary learning, by Jiménez
Catalán’s [23] on Spanish-speaking students learning Basque and
English. Female students in Jiménez Catalán’s survey [36] show a
preference for formal rule, input elicitation, rehearsal and planning
strategies; while males mostly opt for image vocabulary learning
strategies. The superiority of female students in the use of all strategy
categories is also confirmed in the case Korean junior high school
students by Ok [37].
However, a non-significant difference regarding the role of gender,
is reported by (i) Kaylani [38], who do not find differences between
successful female and male learners in Jordanian high schools, (ii)
Phakiti [39], according to whom, if we take into account learners’
proficiency level, no gender differences in reading performance
and use of cognitive and metacognitive strategies are detected; and,
(iii) by Aliakbari and Hayatzadeh’s [40] analysis of Iranian English
students’ performance, who also deny statistically-significant
disparities between men and women, even when the former have
reported higher frequency of strategic use.
Be that as it may, whenever it is female students the ones who are
reported as showing higher strategic learning competence, what both
Western and Eastern pieces of research shares is the factors which
explain this advantage. Drawing on psychological and socialization
research, along with findings on the conversational behaviour of
learners in the ESL/EFL classroom, Oxford and her colleagues tend
to highlight the fact that women present a higher use of general,
social and affective/emotional learning strategies because they are
more empathic, more polite-oriented and display more cooperative
speech styles in the classroom. These psychological features, besides,
are supposed to be due to female students’ stronger motivation and
greater conformity to academic and linguistic norms.
Furthermore, Oxford [32] has noted that as field dependent
subjects, women tend to be more sensitive to the social context. In order to support this idea, she [33] states that females, due to their
higher interpersonal and global orientation, might be superior “in
less analytic aspects of overall L2 communicative competence, such
as sociolinguistic competence, discourse competence, and strategic
competence”. Additionally, in the same article, Oxford [33] also
held that women have a feeling style which favours greater concern
for their interlocutors’ feelings and values (that is, other people’s
faces), and more reflective and analytic learners’ features which
result in both a higher grammatical accuracy and a higher tendency
“to carefully analyze sociolinguistic factors in order to produce the
appropriate response”.
Therefore, the likely interplay existing between gendered strategic
use and affective factors such as degree of motivation [32,36] has led
authors like López Rúa [41] to advocate for an integrative approach
to the effect of gender in foreign language learning, which is the one
we adopt in this paper.
If there is an individual factor that has been neglected in the
language acquisition area is that of learners’ linguistic background,
an individual factor that, according to Aronin & Bawardi [2] cannot
be overlooked any more in light of what Aronin & Singleton [42]
coined as “the new linguistic dispensation”. Research about the
so-called multilingual factor, from Cummins 1976 Threshold
Hypothesis to Herdina & Jessner [43] and their 2002 Dynamic
Model of Multilingualism, has shown that bilingual and multilingual
users when learning additional languages have access to some
cognitive and linguistic benefits and possess a number of abilities
which facilitate this language learning process in terms of attitudes,
learning strategies, metalinguistic awareness [44] or communicative
sensitivity [45].
In this sense, it is a fact that, in the Valencian Country, English is
taught from a monolingual perspective instead of from a multilingual
one, ignoring that it might be introduced as a third language
considering that the vast majority of students are receptive or
productive bilinguals or even trilinguals. This monolingual view in
the EFL classroom implies the loss of methodological opportunities
to make the most of bi- or multilingual students’ skills.
All in all, we also agree with Portolés [4] when she warns that
there are also other studies which suggest that bilingualism per
se is not a determinant condition for additional language success
and claims that many other variables may influence the process of
language acquisition, as our analysis will show.
Last but not least, another element that we have analysed is that of
the students’ Out-of-School Contact with English. Failure to achieve a
good command of English is an issue that worries most parents to the
extent of enrolling their children in extracurricular English lessons
or paying for private tuition in commercial language schools. It is
our contention, however, that the perceived need of extracurricular
English lessons and the im(possibility) of affording them might
increase social differences and it works against the universal right of
citizens to receive quality education for free.
The Study
Research questions
The first aim of this research is to explore the correlation between
learners’ individual factors and school grades as a quantitative index
of academic success in learning EFL. This aim leads us to formulate
the following research question: Which are the most important
individual variables in terms of academic achievement in the EFL
classroom? In other words, which individual factors are the most
successful predictors of language learning success? Internal and/or
psychological dimensions versus social factors and/or conditions of
learning? Besides, concerning the former and following Robinson
[46], are cognitive variables more influential than affective ones?
The second aim of the present research is to explore the
correlation among learners’ individual factors. This aim implies the following research question: Which correlations exist among
individual variables that condition academic success in the EFL
classroom?
All in all, by analysing the relation between each individual
variable and the target variables and also by analysing the correlation
among individual variables, we will try to define the profile of a
prototypically good learner. In light of the literature review presented
in this section, however, we might hypothesize that our good language
learner would be a productive bilingual or multilingual [43-45]
female student with some aptitude to language learning [1], a strong
motivation [18,19,23] with a special emphasis on integrativeness [14]
and positive attitudes towards the target language and culture [20],
who resorts to a wide range of learning strategies, especially social
ones, who shows simultaneously a feeling learning style along with
reflective and analytic learners’ features which are very extremely
useful in grammar-oriented and teacher-led classrooms [33], who
actively searches for teacher guidance and finds parental support,
especially on the part of middle-class families, who still favour girls
studying languages over sciences [27,47].
Participants
We have studied a number of individual factors conditioning the
academic success in the case of forty-one 12 to 15-year-old students
in a state high-school based on a bilingual territory where there is also
a great amount of trilingual students coming from migrant families.
In our sample, there are students with the following mother tongues:
Catalan, Spanish, Romanian, Arabic and Portuguese. All the students
had the same English teacher and belong to groups with Catalan as
the main language of teaching and learning (Programa Plurilingüe
d’Ensenyament en Valencià). The research was implemented at the
IES Jaume I in Borriana (province of Castelló, Valencian Community,
Spain) during the second and third trimesters of the academic year
2014-2015. This is a high-school with around 670 students located in
a neighbourhood with both working and middle-class families. The
percentage of immigration in this high-school is about 12%. Borriana
has almost 35.000 inhabitants. This middle-size city is situated in
the northern part of the Valencian Country, which is a Spanish region
with two cooficial languages: Catalan (Valencian) and Spanish.
The Sample consists on 41 Students from three 2nd of ESO
groups. All of them have ages between 12 and 15 years old. We have
measured academic success in EFL, in terms of score achievement,
and we have tried to measure and analyse the relationship with
academic success and seventeen individual variables and also the
correlations among individual variables. Our sample is balanced in
terms of sex. We have 22 girls and 19 boys.
Instruments of data collection
According to Gardner [12], we cannot only focus on the variables
individually. We also need to analyse the interdependency among the
individual variables that condition the target variables:
We often speak of the individual different variables as independent
variables and interpret the correlation as indicating that the variable
is responsible for that achievement [marks, academic success], but
because individuals are not randomly assigned to the independent
variable, it is only an interpretation. […] Correlation [between IDs and
academic success] does not mean causation. As a consequence any
‘casual statement’ is arbitrary, and arguments about what causes what
are meaningless.
According to this, we have also taken into consideration the
correlations among independent variables, because what matters is
that such correlations indicate that there is an association between
the two variables in the population, and such connection has a
meaning.
According to the statistical analysis conducted by means of the
SPSS program version 22.0, our variables are normally distributed,
according to the Kolmogorov Smirnov tests. After having checkedthat all the variables were normally distributed, we studied the
correlations among independent variables and also in relation to the
target variable. To study these correlations and their significance, we
have used different statistical elements: box and whisker plot, scatter
plot, ANOVA test and t-test.3
Now we are going to measure the instruments used to compile the
information about all the variables. We have used six questionnaires
to collect the information about the individual variables.
To be sure about the validity of the tests implemented, we have
used tests already employed in similar statistical investigations in
foreign language learning. Some of them are well-known and are
accessible through the Internet.
Test 1: Attitude/Motivation Test [48].4
This is a validated test
that has been used in many EFL studies. It consists on 104 questions
and measures twelve variables in a rating scale with six options.
The variables measured are: Interest in Foreign Languages, Parental
Encouragement, Motivational Intensity, English Class Anxiety, English
Teacher Evaluation, Attitudes toward Learning English, Attitudes
toward English-speaking people, Integrative Orientation, Desire to
Learn English, English Course Evaluation, English Use Anxiety and
Instrumental Orientation. Although there are many data that we have
omitted.
Test 2: Test of bilingualism. From this test, we have only
considered the question about the student’s mother tongue.
Test 3: My history as English learner. From this test, the question
in which we have focused our analysis has been the one about
whether the student has any Out-of-school contact with the English
language.
We decided not to implement the aforementioned questionnaires
in English in order to facilitate our respondents’ comprehension.
Moreover, we have considered important to alternate between
tests written and administered in Spanish and tests written and
administered in Catalan because we did not want to show any implicit
preference for one of the two co-official languages.
Test 4: Aptitude. This test is an adaptation from the one included
in the doctoral thesis Language Aptitude in Young Learners: The
Elementary Modern Language Aptitude Test in Spanish and Catalan
elaborated by Suárez et al. [49] in 2010. The test is divided into four
parts. Each part counts 25 points up to 100. In the first part, called
“Hidden Words”, the students have to answer five questions. They
have some letters that altogether form a word. Next, they have four
options of possible words and they have to choose the correct one.
In the second part, called “Correspondent Words”, the students have
five groups of two sentences each. In the first sentence, they have
one word in capital letters. In the second sentence, they have to
circle the word that they intuitively consider that has an equivalent
grammatical function.
In the third part, called “Words that rhyme”, the students have
five options of one isolated word followed by four options of words.
They have to circle the word that they consider that rhymes with the
first one. The last part is called “Numbers in another language”. Here,
the students are expected to recognise the numbers in an invented
language. Firstly, they listen to some examples of this system of
numbers in an unknown language. Then, they have to write down
the numbers in this new language that they are able to identify by
listening them.
Test 5: Personality. The test of personality used in this research
is an adaptation into Catalan from the MBTI test (Myers–Briggs Type
Indicator), about the way people make decisions and perceive the
world. This test codifies its results into sixteen possible combinations
of the next four features: Extraverted/Introverted, Thinking/Feeling,
Sensing/Intuitive and Judging/Perceiving.5
Test 6: Learning Strategies. The test about learning strategies
used in our research has been an adaptation into Catalan from
Oxford’s “Strategy Inventory for Language Learning” (SILL).6
This test
has a total amount of 50 questions. Each question has to be measured
in a scale from 1 to 5 depending on the degree of identification to
the student with each statement (1 is “never or almost never true in
my case” and 5 corresponds to “always or almost always true in my
case”).
From all the learning strategies including in the SILL test, we have
measured the following ones: remembering more effectively, using all
mental processes, compensating for missing knowledge, organising
and evaluating the learning, managing emotions and learning with
others. We have marked each learning strategies’ test according to
the next orientations7
.
The final average is between 1 and 5. For the statistical analysis,
we have codified these results in three levels:
• High level of use of learning strategies when learning English
(codified as 3): 4.4 - 5.0.
• Medium level of use of learning strategies when learning English
(codified as 2): 2.5 - 4.4.
• Low level of use of learning strategies when learning English
(codified as 1): 1.0 - 2.4.
Next, we shortly describe the two parts of the statistical analysis,
each of one respectively focused on the answer of one of the two
questions.
To answer the question 1, we carry out a thorough analysis of each
individual variable with the target variable. Each analysis will depend
on the characteristics of the independent variable because we have
a wide range of individual variables with different features and the
descriptive analysis must take these differences into consideration.
The target variable is the mean of the score obtained for each
student in the last 5 trimesters. The independent variables are the
next ones:
