The Role of Motivation and Willingness to Communicate in Second Language Learning

(1)

The Role of Motivation

and

Willingness to Communicate

in Second Language Learning

Marina Smirnova

S2069334

Supervisor: Dr. M.H. Verspoor

Second reader: Dr. S. L. Thorne

MA thesis

Department of Applied Linguistics

Faculty of Arts

Rijksuniversiteit Groningen

(2)

Acknowledgements

First of all, I would like to express my deep and sincere gratitude to my supervisor, Professor Marjolijn Verspoor, who gave me the useful guidance and advice. I am also deeply grateful to my second reader, Professor Steve Thorne, for his detailed and constructive comments. I wish to thank Xiaoyan Xu who provided me advice with the statistical analyses in SPSS and the bilingual secondary school “Stedelijk College Eindhoven” for making it possible to collect the data. I would also like to thank the participants of the study for their time and effort. Finally, I am grateful to my family and friends for their encouragement and support.

(3)

Abstract

The present study investigated the impact of individual differences such as motivation and willingness to communicate on L2 proficiency. An empirical research was carried out on 540 HAVO and VWO schoolchildren enrolled in bilingual programmes in a secondary school in Eindhoven, the Netherlands. Two types of materials were used in the study: a questionnaire on leaner characteristics and the writing and the vocabulary tasks for measuring the students’ L2 proficiency. The results showed that VWO students had higher levels of English proficiency than HAVO students; this difference was not only due to the differences in school type, grades and scholastic aptitude, which were controlled for, but also due to motivation and WTC, which contributed significantly to the students’ L2 proficiency. We can say that motivation did influence the students’ L2 proficiency – the participants with high levels of motivation outscored low motivated ones. We found that a great majority of our participants were instrumentally motivated; however, the difference in L2 proficiency between integratively and instrumentally motivated students was not confirmed. Moreover, we could not prove that two school types, HAVO and VWO, differed in their levels of motivation and WTC. Furthermore, we found no difference between female and male subjects in motivation levels, while there was a significant difference in WTC, with boys being more willing to communicate than girls. And finally, the age did not show any effect either on the students’ motivation or WTC.

(5)

1. Introduction

Language learning is a complex dynamic process where a lot of factors such as learner characteristics play a role. Learner characteristics in applied linguistics have traditionally been investigated within the context of individual differences (IDs). As Dornyei (2009) writes, four ID factors - motivation, language aptitude, learning styles, and learning strategies, have received special attention in L2 research. Language aptitude and language learning motivation are both seen as the primary ID variables with the latter perceived as the most important affective variable, and the former – as the main cognitive factor that shapes L2 developmental process (Dornyei 2010).

Recently ID factors have started to be looked at from a dynamic system theory perspective, according to which language is a complex system consisting of different interacting subsystems, which can be easily affected by different factors that influence language development (De Bot, 2008). We see aptitude and motivation as such subsystems, which are interconnected with each other and influence each other. In a greater study these factors will be investigated together. We will also look in the greater study at the context in which Dutch schoolchildren learn a L2. It is well known that besides from school, they receive a huge amount of English input from out-of-school context, for example, television, Internet, etc. However, in the present study, I am going to focus on motivation. Oxford and Shearin (1994) state that “motivation is believed to be one of the main determining factors in success in developing a second or foreign language” (Oxford & Shearin 1994; p.12). Another learner characteristic, closely linked to motivation and at which I will also look in the present study, is willingness to communicate (WTC), which is defined as “a readiness to enter into discourse at a particular time with a specific person or persons, using a L2” (MacIntyre et al., 1998; p.547).

(6)

VWO schoolchildren enrolled in bilingual programmes in a secondary school in Eindhoven, the Netherlands. Bilingual education has become very popular since it offers courses taught not only in Dutch but also in English; such programmes are more challenging and more prestigious. Verspoor et al (2009) show that students of bilingual education have a higher level of English proficiency than students in regular programmes. However, in the present study we are interested in the difference between bilingual students from two types of programmes: HAVO and VWO, with the latter being a more advanced level of education. Both HAVO and VWO bilingual programmes are characterized as selective types of secondary education. After elementary school students continue either with HAVO or VWO depending on the results of a scholastic aptitude test (Cito test), parents’ opinion and pupil’s own choice. To get into VWO students should obtain a higher Cito score than for HAVO. In both types of programmes students are exposed to the same amount of English input at school: they have the same number of English classes and 50 % of all their classes are taught in English.

The main question then is if HAVO and VWO students show different levels of language proficiency and if so, what the reasons are for that. Does only scholastic ability between two types of students contribute to their differences or is it the case that motivation and WTC do play a role here too? Which students are more motivated and more willing to communicate? Should learner characteristics be taken into account?

(7)

(8)

2. Background

In this literature review I will first look at the concept of motivation and review what Gardner wrote about it in his socio-educational theory; then I will focus on the distinction between two principal types of motivation, integrative and instrumental. Later on I will discuss current theories of motivation research. Then the concept of WTC and its connection with motivation will be explored, followed by the subsection on dynamic nature of both learner characteristics. Finally, I will formulate my research questions and give some predictions for my hypotheses.

2.1. Motivation

How is it that some people can learn a second or foreign language so easily and so well while others, given what seem to be the same opportunities to learn, find it almost impossible?

Gardner, 1972

2.1.1 Foundations of the research on motivation: R. Gardner and his socio-educational theory

(9)

Gardner (1985) established a model of motivation in L2 learning called the socio-educational model. He writes that motivation involves four aspects: a goal, effortful behavior, a desire to attain the goal and favourable attitudes toward the activity in question. The goal helps to define the motivational orientation of a learner, which is either instrumental or integrative. Instrumental is related to the potential pragmatic gains from learning a language like a better job or a higher salary, while integrative is associated with positive attitudes toward the L2 group and a willingness to integrate in that community. According to Dornyei (2001), a key issue in Gardner’s motivation theory is the relationship between motivation and orientation. Motivation itself includes three components: motivational intensity, desire to learn the language and attitudes towards learning the language. The role of orientations then is to help to raise motivation and direct it towards either integrative or instrumental sets of goals.

The most elaborate aspect of Gardner’s theory has been the integrative motive which contains three main parts:

1. Integrativeness (integrative orientation, attitudes towards the L2 community, etc) 2. Attitudes towards the learning situation (attitudes towards the language teacher and

the L2 course)

3. Motivation (effort, desire and attitude towards learning)

The components of integrative motive are operationalized in Gardner’s “The Attitude/ Motivation Test Battery” (1985a). The AMTB consists of over 130 items which make up five categories: integrativeness, attitudes toward the learning situation, motivation, instrumental orientation and language anxiety.

(10)

individual is studying the language. The reasons can be various - to be able to speak with members of a particular L2 community, to get a job, to be able to travel, to satisfy a language requirement and many, many others.

2.1.2 Integrative and instrumental motivations

Integrative and instrumental motivations have become the most widely known concepts in L2 motivation research. According to Gardner (1972), an integratively oriented learner is directed primarily towards representatives of an ethnolinguistic community in which he is interested, whereas as Gardner writes later, an instrumentally motivated individual is looking for “a goal that does not seem to involve any identification or feeling of closeness within the other language group, but instead focus on a more practical purpose” (Gardner 2001, p.10). The purpose for language learning in the latter case can be meeting requirements for school or university graduation, applying for a job, etc. Gardner and Tremblay (1994) remark that integrativeness and instrumentality are not antagonistic counterparts but are often positively related. The same point of view is shared by Brown (2000) who suggests that learners seldom select one form of motivation when learning a L2, but rather a combination of both.

(11)

The main purpose of Gardner and MacIntyre’s (1995) study was to investigate the effects of integrative and instrumental motivation on the learning of French/English vocabulary. The results proved that both integrative motivation and instrumental motivation facilitated learning. Interestingly, instrumental motivation was determined in terms of immediate financial reward for doing well (10$ for completing a learning task) and the results indicated that instrumentally motivated students studied longer than non-instrumentally motivated ones when they had an opportunity to profit from learning, but this distinction vanished when the incentive was removed.

Sawhney’s (1998) study which focused on the effects of instrumental motivation on university students learning German in India showed that such reasons as getting a good job, doing international business, etc demonstrated that the students were interested in learning German for prestige, social recognition and economic gain. A large-scale study on 5000 Hungarian teenagers was undertaken by Dornyei and Clement (2001) in which they looked at motives that encouraged the students to learn English, German, Russian, French or Italian. In case of American English, the main three integrative factors were the attraction by Americans, the desire to be like Americans and the wish to travel to the USA, while the Russian language was placed the first for instrumental reasons.

(12)

learner has no opportunity to use the target language or to interact with members of this target community.

2.1.3 Current theories of motivation

In the mid 1990s Gardner’s socio-educational model was criticized by such respected researchers as Dornyei (1994), Oxford and Shearin (1994), Crookes and Schmidt (1991) who proposed that motivation should be studied from different perspectives. The main argument was that Gardner’s theory put too much stress on the integrative and instrumental distinction and thus ignored a list of variables from the broad psychological literature in motivation.

As a result, other possible kinds of L2 learning motivation have come to the attention of researchers. The study by Oxford , Shearin and Barnard (1994) on 218 American high school students learning Japanese showed that besides the instrumental motivation to learn the language for future career and integrative motivation for making Japanese friends, more than two-thirds of the students had additional reasons for studying Japanese, including a search for personal challenge, enjoyment of the elitism of studying a difficult and unusual language, possibility of having a private code that parents would not know, etc. Dornyei (1990) in his study of young Hungarian adult learners found three loosely related dimensions of a broadly conceived integrative motivational subsystem: interest in foreign languages, cultures and people; desire to broaden one’s view and desire for new challenges. Noels (2001) offered that besides integrative and instrumental motivation people can learn a L2 for various reasons such as intellectual stimulation, showing off to friends, a need for achievement, interest, friendship, prestige, school, etc.

(13)

in achievement behaviours in future while past failure will provoke fear and suppress achievement behavior. The instrumentality theories center on the individual’s expectation of receiving a reward, sometimes they are also called expectancy-value theories. The equity theory is characterized by a ratio of inputs to outcomes; inputs consist of intellectual ability, personality traits and experience, those things that individuals contribute to the work while outcomes include grades, promotion, praise, etc. A language learner must believe that the results are worth the effort he spends. In reinforcement theories individual behaviour is attributed to stimuli and reward. Oxford and Shearin argue that the concept of motivation can be expanded to comprise various kinds of motivations that are present in learners and they believe that contributions from many aspects of psychology (general, educational, industrial and cognitive developmental) are needed for a new vision of L2 motivation.

Dornyei (1994) writes that many studies tried to broaden Gardner’s motivation construct by adding new components such as intrinsic/ extrinsic motivation, need for achievement, attiribution about past successes/ failures, self-confidence, etc. The most general distinction in motivation theories is between intrinsic and extrinsic motivation. “Extrinsically motivated behaviours are the ones that the individual performs to receive some extrinsic reward or to avoid punishment” (Dornyei 1994; p.275). Intrinsic motivation, as defined by Noels (2001) refers to reasons for L2 learning that come from one’s inherent pleasure and interest in the activity.

(14)

Cognitive theories of motivation see it as a function of a person’s thoughts rather than of some instinct, need or state. Weiner (1992) presents three main cognitive conceptual systems: attribution theory, learned helplessness and self-efficacy theory. They all deal with a person’s self-appraisal of what he or she can or cannot do, which influences how a person will strive for achievement in the future. Another important construct close to self-efficacy is self-confidence, which reflects the belief that one has the ability to produce results or accomplish goals; it includes two parts - language use anxiety and self-evaluation of L2 proficiency.

As Dornyei points out, although researchers acknowledged Gardner’s social psychological model, they also called for “a more pragmatic, education-centered approach to motivation research” (Dornyei 1994; p.273). In his paper Dornyei (1994) tries to further understand L2 motivation from an educational perspective by examining three groups of motivational components that are specific to learning situations: 1. course-specific (e.g. syllabus, teaching method); 2. teacher-specific (e.g. teacher’s personality, teaching style, etc); 3. group-specific (dynamics of the learning group). Then he presents a general framework of L2 motivation consisting of three levels: the Language Level, the Learner Level, and the Learning Situation Level. The most general level is the Language Level which can be described by two motivational subsystems, an integrative and an instrumental ones. The Learner Level includes need for achievement and self-confidence while the Learning Situation Level is made up of course-specific, teacher-specific and group-specific motivational components. Thus, Dornyei tried to outline a comprehensive motivational construct relevant to L2 classroom motivation.

(15)

in an empirically grounded, explicit, and testable manner, the research on motivation benefited from consideration of motivational constructs from other research areas. What we cannot question in relation to motivation is that the studies that used the socio-educational model to look at the role of motivation in language learning showed that motivation influences achievement.

Clement and Kruidenier (1985) tested Clement’s model of second language proficiency with the help of the causal modeling technique. The results of the analysis of the data on attitudes, aptitude, motivation and second language proficiency from 1,180 francophone students enrolled in grades 7, 9 and 11 in schools of eastern Québec support the hypothetical constructs and causal sequence proposed by Clément. Tremblay and Gardner (1995) in their study investigated the relation of a number of new measures of motivation such as persistence, attention, goal specificity, and causal attributions to each other, to existing measures of attitudes and motivation and to indices of achievement in French courses. The study was conducted in a bilingual school environment on 75 students registered in a French school but the language environment outside the school was mainly English. The structural equation modeling analysis confirmed the hypotheses about the relationships among the motivational constructs and achievement. Gardner’s (2006) research, in which he focused on six individual difference variables, demonstrated that among the correlations of variables with English grades for Spanish students the highest correlate was found for motivation (.40 and .49) which indicates that the more highly motivated students have higher grades than the less motivated ones.

2.2. Willingness to communicate

(16)

probability of engaging in communication when free to choose to do so. According to McCroskey (1985), WTC reflects the stable predisposition to talk in various situations and is seen basically as a personality trait. They related WTC to such attributes as communication apprehension, perceived communication competence, introversion-extraversion, self-esteem, etc. While McCroskey’s concept of WTC was seen as personality trait, MacIntyre et al. (1998) see it as a situation-based variable. MacIntyre and his colleagues (1998) provide an account of the linguistic, communicative, and social psychological variables that might affect one’s WTC and they outline a comprehensive conceptual model that can be useful in describing, explaining, and predicting L2 communication.

The authors present a pyramid-shaped model which shows the range of potential influences on WTC. They believe the broadest factors to be the basis on which the other influences operate, as they move from these basic influences, they focus more on L2 communication. The model is made up of six categories which are referred to as “layers”. The first three layers represent situation-specific influences on WTC at a given moment in time:

- Layer I Communication behaviour (1.L2 use) - Layer II Behavioral intention (2.WTC)

- Layer III Situated Antecedents (3.Desire to communicate with a specific person and 4.State Communicative Self-confidence)

The latter three layers represent stable, enduring influences on the process:

- Layer IV Motivational propensities (5.Interpersonal motivation; 6.Intergroup motivation; 7.L2 Self-confidence)

- Layer V Affective-Cognitive Context (8.Intergroup attitudes; 9.Social situation; 10.Communicative Competence)

(17)

WTC is an act of willing to engage in a conversation that shows self-confidence and communicative competence and it consistently relates positively with perceived communicative competence and negatively with language anxiety. Peng (2007) wrote that L2 WTC can also be concerned with social support, personality traits, and gender. Cultural traits such as shyness, introversion or group-orientedness may influence WTC. For instance, Peng’s study (2007) demonstrated that Chinese culture has an influence on Chinese learners’ L2 WTC. Since Chinese culture concerns such interpersonal relations as face protected orientation and other-directed self, it turns out to be submissive and to strongly influence WTC in L2. Situational trait can influence individual’s WTC as well. L2 WTC can be determined by such situational variables as topics, interlocutors, and conversational context of the communication (Kang 2005).

MacIntyre (2007) notes that initiating communication is a matter of choice, thus deciding to communicate in L2 at a particular moment of time is an act of volition. A language learner makes a decision to start communication when the opportunity appears. In other words, WTC, defined as the probability of initiating communication, given choice and opportunity, integrates motivational processes with communication competencies and perceived self-confidence (MacIntyre 2007). There are a lot of variables that can change or influence a person’s WTC, including the number of people present, the formality of the situation, the degree of acquaintance between communicators, the topic of discussion, etc.

(18)

investigated age and gender differences in WTC, communication apprehension, and self-perceived communication competence by examining three age groups of participants (junior high, high school, and university). The results indicate that patterns of WTC across different age groups are different in men and women, for instance, junior high females demonstrated higher WTC than their male counterparts, while females at the university level were higher in communication apprehension and lower in self-perceived competence than male university students. Interestingly, in all three age groups self-perceived competence turned out to be the most consistent predictor of WTC among males, while among females it was communication apprehension.

MacIntyre (2007) points out that language anxiety and motivation are vital factors in WTC in L2 as well as there are two principal antecedents of WTC - communication apprehension and self-perceived competence (MacIntyre 1994). The study on Japanese ESL students by Hashimoto (2002) showed that motivation and WTC affected L2 communication frequency in classrooms which means that students who had greater motivation for language learning and who were more willing to communicate reported using the language more frequently. The variables underlying WTC were also examined: perceived competence and L2 anxiety were found to be causes of WTC, and L2 anxiety was found to negatively influence perceived competence. A path from L2 WTC to motivation appeared to be significant in this study, indicating that willingness to communicate has motivational properties.

(19)

gender with respect to WTC, which uncovers different patterns among male and female L2 learners. In general WTC may be seen as both an ID factor facilitating L2 acquisition and as a nonlinguistic outcome of the language learning process. Individuals can feel uncomfortable to speak in L2 with a stranger or on the opposite among friends, some people are afraid that others will judge their language ability; others get anxious when teachers are present or are afraid to speak up in a public and prefer to speak in a small group. Numerous factors like group size, task content, topic, etc. can influence WTC.

2.3. Research questions of the present study

In this study I will try to answer the following research questions that were formulated on the basis of the theoretical insights:

1. Having the same amount of English input at school, do HAVO TTO and VWO TTO students differ in their levels of L2 proficiency,

a. based on the results of writing task

b. based on the results of vocabulary meaning task

c. based on an aggregated score of writing and vocabulary ? 2. Do motivation and WTC affect the students’ L2 proficiency and how?

a. Which students, integratively or instrumentally motivated, score higher?

b. Which students, HAVO TTO or VWO TTO, are more motivated and more willing to communicate?

c. Are there differences in the students’ motivation and WTC in relation to age and sex?

(20)

(21)

3. Method

In this section I will give the information, firstly, on who my subjects are, then what kinds of materials are used and, finally, how the data are arranged and analyzed.

3.1. Subjects

The subjects of our project were students from the bilingual secondary school “Stedelijk College Eindhoven” (SCE), the Netherlands, which offers both regular and bilingual programmes. For our greater research we collected the data from both regular and bilingual students, but for this study I will focus on the data only from bilingual HAVO and VWO classes.

Unlike in regular programmes, in bilingual programmes there are not only more hours of English classes but about 50 percent of all the classes are taught in English. According to the standard set by the European Platform, an organization that coordinates the network of schools, the bilingual education should meet a number of certain requirements, for instance, only authentic English materials are used, at least two subject are taught by native speakers of English, form aspects of English are most important in those classes where English is taught as a subject, etc. The bilingual education is based on the principles of Content and Language Integrated Learning (CLIL), in which the L2 is used as a medium of both instruction and transferring of content, so that students acquire the language by actively using it and having meaningful L2 interactions. It means that bilingual education differs from regular one not only in the amount of English input but also in relation to English output since students have to actively use the L2 for communication.

(22)

students of bilingual education were more proficient in English than students that did not receive bilingual education. In the present study, I am not going to compare bilingual versus regular education but instead I will look only at bilingual students but from two different school types – HAVO and VWO.

These are two programmes of general education that lead to higher education in the Netherlands. HAVO (hoger algemeen voortgezet onderwijs; literally, "higher general continued education") lasts five years (1-5 grades) and is attended from age twelve to seventeen. VWO (voorbereidend wetenschappelijk onderwijs; literally, "preparatory scientific education"), that has six grades and is attended from age twelve to eighteen, is the highest level of secondary education. Compared to HAVO the difficulty level of the profiles at VWO is higher and to enter this type of education students should get a higher score (545-550) for an aptitude test called the Cito Eindtoets Basisonderwijs (literally, "Cito final test primary education"). VWO students are said to be more gifted, more motivated and more willing to learn. In our study we investigate if bilingual HAVO students and bilingual VWO students differ in L2 proficiency and if learner characteristics, such as motivation and WTC, help account for this difference. This question has been a matter of interest both for educators and school authorities; that is why we believe the present research to be novel and original.

(23)

eliminated from the analysis. All in all we had 540 students from bilingual stream. Table 1 shows the overall number of participants and specifically the number for each school type and grade. School type grade Total 1 2 3 4 5 6 VWO 57 62 87 47 47 30 330 HAVO 78 36 34 40 22 - 210 Total 135 98 121 87 69 31 540

Table 1 Subjects of the study

Among 540 students, 53 of them had either no vocabulary or writing score which led to a missing proficiency score and 25 of them had no cito score. For all later analyses only those students (487 in total) with a proficiency score were included. Proficiency score in this study is a standardized score of the sum-up of the two standardized scores of vocabulary and writing.

3.2. Materials and procedures

(24)

The questionnaire included several parts (see Appendix A): motivation, WTC, in class engagement and out of school contact. The latter two are not included in the design of the present study. The motivation section consisted of fifteen statements about attitudes towards English and about integrative and instrumental motivation. The statements were measured on a four-point Likert scale: 1.strongly agree; 2.agree; 3. disagree; 4. strongly disagree. We excluded a middle point “neither agree nor disagree” so that the respondents did not avoid making a real choice. Each response option was assigned a number for scoring purposes (e.g. strongly agree=1, strongly disagree=4). The items were borrowed from Gardner’s AMTB (2004) and some were partially changed to suit our study, for example: I would like to have

many English speaking friends; Studying English is important because it will be useful in getting a job, etc. A reliability analysis for motivation showed that the Cronbach’s alpha is

0.823, which is very satisfactory and means that the statements measured the same underlying construct.

WTC was constructed by eight statements (four statements about WTC at school and four statements about WTC out of school), e.g. I’m afraid that my classmates will laugh at

me when I have to speak English; I feel worried when I have to speak Englis, etc. The items

were adopted from Gardner’s AMTB (2004) and McCroskey’s (1992) WTC questionnaire with some changes to make the statements more appropriate for our participants. The answers were measured on the same four point Likert scale used for motivation. The Cronbach’s alpha for WTC was also very good (0.863).

(25)

We believe that a writing task is a good way to measure overall language proficiency. All in all, we chose the combination of two types of tests because it may give the best insight into the development of L2 proficiency.

For the vocabulary task we used the “English as a Foreign Language Vocabulary test” (EFL Vocabulary test), devised by linguist Paul Meara. The EFL-test is a yes/no test that measures the receptive vocabulary of learners of English as a foreign language. It consists of two types of words: real English words and non-existent words that are made according to English phonological rules. A test-taker has to indicate if he is familiar with the meaning of a word or not. There are four different response possibilities for the test: two correct responses (“yes” to a real word or hit and “no” to a non-existent word or correct rejection), and two incorrect responses (“no” to a real word or miss and “yes” to a pseudo-word or false alarm).

The scores on the EFL-test were calculated with the help of Isdt scoring method based on the Signal Detection theory (SDT) and suggested by Huibregste et al. (2001). The following formula was used:

4h(1-f)-2(h-f)(1+h-f) Isdt=1 – —————————

4h(1-f)-(h-f)(1+h-f)

In the formula h stands for hit, f for false alarm and Isdt is the EFL-score. The final EFL- scores range from 0 to 1, with 0 indicating a very low vocabulary score and 1 indicating a very high vocabulary score. Huibregste’s scoring method reflects not only the ratio of hits to false alarms, but also the response style of the test-taker, thus providing a score that is more valid (Huibregtse et al. 2001).

(26)

For the writing task the students from grades 1-3 wrote a short essay about the vacation they had. They could write stories that were not necessarily true and as many words as they wanted. Students of grades 4-6 wrote a short newspaper article (200-500 words) on a topic “All Dutch high schools should teach half of their lessons in English”, they had to agree or disagree with the statement, providing some facts and examples. As it can be seen the topics for essays were chosen in a way that they would be appropriate for the age of the subjects and their expected level of L2 proficiency.

The essays were graded by a team of evaluators that consisted of several native speakers of English, of Dutch and other languages. Grades 1-3 were given scores between 1 and 5, with some possibilities of 6 and 7 for extremely good writings. The scale for grades 4-6 ranged between 4 and 8, with 9 in rare cases. Attention was paid to the following aspects of written texts: complexity (sentences, clauses, tenses, vocabulary), fluency (length, cohesion, chunks) and accuracy (errors).

Since participation in the study was not obligatory and although most students filled in the questionnaire, not all of them approached the vocabulary and writing tasks seriously. In case of the vocabulary test some participants randomly ticked words and as of essays some, obviously not motivated, students either did not write anything at all or copied and pasted texts from the Internet. Such subjects were excluded from the study.

3.3. Design and Analyses

(27)

and tests, but also some qualitative data collected through one-to-one interviews conducted with the HAVO and VWO students and some teachers (see Discussion). The quantitative data were analyzed with the help of statistical software PASW Statistics 18.

(28)

4. Results

This section is built according to research questions and presents eleven tables, eleven figures and the results of thirteen statistical analyses.

Research question 1

Having the same amount of English input at school, do bilingual HAVO and VWO students differ in their levels of L2 proficiency,

a. based on the results of the writing task

b. based on the results of the vocabulary recognition test c. based on an aggregated score of writing and vocabulary?

In order to examine the difference in language proficiency between the two types of school, I looked separately at scores for the writing task, for the vocabulary test and at the accumulated score of both, which is called general proficiency here.

Firstly, I present graphically the means of the writing scores for five HAVO grades and for six VWO grades in Figure 1.

(29)

We see that there is almost no difference in the writing scores between the 1st and 3rd grades and there is a little difference between the 4th grades. The difference between the 2nd grades and between the 5th grades is more visible. Since the 6th grade is present only in the VWO stream, we cannot make any comparisons.

The means and standard deviations (SDs) of the writing scores for each school type are in Table 2. The mean of the writing scores for VWO is higher than the mean for HAVO. It is also higher than the total mean for both school types.

VWO HAVO Total

mean SD mean SD mean SD 4.38 1.503 3.86 1.556 4.16 1.546

Table 2 Mean and SD of writing score

For all further statistical analyses, the 6th grade of VWO is excluded in order to make an equal number of classes. To check if the difference in writing scores is significant, the independent samples t-test was run. The analysis reveals that there is a significant effect of school type on L2 writing score (t(460)=3.598, p<0.01); VWO students score higher on the writing test than HAVO students.

(30)

Figure 2 Average vocabulary score for VWO and HAVO per grade

As it can be seen from the bar graph, the 3rd grades have the same scores while all other grades of VWO score a bit higher than those of HAVO.

The means and SDs are presented in Table 3:

VWO HAVO Total

mean SD mean SD mean SD

.8023 .09821 .7388 .10692 .7749 .10668

Table 3 Mean and SD of vocabulary score

(31)

Figure 3 Average vocabulary score for VWO and HAVO per grade

If we look at the Figure 3 it is obvious that all VWO grades have higher levels of general L2 proficiency than HAVO grades. We see the same pattern from the overall mean scores of two school types (Table 4):

VWO HAVO Total

mean SD mean SD mean SD

.5710 1.23872 -.1384 1.34938 .2654 1.33333

Table 4 Mean and SD of general proficiency score

(32)

Main question: Do motivation and WTC affect the students’ L2 proficiency and how?

First of all, reliability analyses were done in order to see if the statements in the questionnaire measure the same underlying constructs of motivation and WTC respectively. Cronbach’s alpha for motivation indicates a very satisfactory reliability level of the 15 motivation items (0.823). Reliability analysis for WTC shows also a very good result (0.863). Since all items measure the same underlying constructs, we will be able to use the means of motivation and WTC in our later analyses.

Then, two separate factor analyses were done on motivation and WTC in order to find possible smaller clusters of variables (See Appendix C). Loadings higher than .3 can be found in Table 5 and Table 7. If one item loaded on more than one factor, I looked only at the largest factor loading.

(33)

Factor 1 Factor 2 Factor 3 Factor 4 I enjoy speaking English. .792

I like the English language. .693 I would like to have many English

speaking friends.

.655

I like to talk to native speakers of English.

.622

I am good at English compared to my fellow students.

.567

It is important for me to learn English.

.749

Studying English is important because it will be useful in getting a job.

.669

English is an important language of communication in the world.

.626

I will need English after high school for my studies.

.607

English enables me to communicate with people from different language groups.

.603

I want to be fluent in English. .418

I like my English class. .766

I am interested in the cultures of English speaking countries.

.561

I understand everything during my English lessons.

.634

English helps me to find new friends on holiday.

-.410

Table 5 Motivation Factors

The first factor seems to cluster the items positive attitude towards English, integrative

(34)

of the items attitude to English class and integrative motivation. The last factor includes only two items and can be described as a measure for self-evaluation in English class and

instrumental motivation.

Since our motivation questionnaire included items not only on integrative and instrumental motivation but also on self-evaluation, attitudes to the English language and to the English class, we see that the factors, clearly clustering either integrative or instrumental motivation, could not be defined. To distinguish those factors, we ran another factor analysis (Cronbach’s alpha=0.753) only on the items that characterize these two types of motivation and we got two distinct factors that describe instrumental (factor 1) and integrative (factor 2)

motivation (Table 6).

Factor1 Factor 2 It is important for me to learn English. .781

I will need English after high school for my studies.

.726

Studying English is important because it will be useful in getting a job.

.696

English enables me to communicate with people from different language groups.

.586

I would like to have many English speaking friends.

.779

I am interested in the cultures of English speaking countries.

.731

I like to talk to native speakers of English. .668 English helps me to find new friends on

holiday.

.602

(35)

Later on, these two factors were recalculated as separate variables of integrative and instrumental motivation and used for statistical analysis in order to answer one of the research questions.

For the 8 WTC items, the factor analysis results in a very clear pattern with two factors. Four statements load on factor 1 that was defined as negative feeling about communicating in

English and other four statements load on factor 2 and describe confidence about speaking English.

Factor 1 Factor 2 I’m afraid that my classmates will laugh at me

when I have to speak English.

.803

I feel worried when I have to speak English .796

I get nervous speaking English .761

I do not feel comfortable speaking English .760 I feel comfortable talking English in a small

group of people my age

.787

I feel comfortable talking English in a large group of people my age

.746

I often give answers in English in class .714 I feel comfortable speaking English in a group

discussion

.634

Table 7 WTC factors

(36)

relationship between motivation and L2 proficiency (r=.169; p<0.01) and between WTC and L2 proficiency (r=.185; p<0.01).

However, a correlation analysis doesn’t tell us anything about the causality, about the amount of influence that affective variables have on English proficiency. We also should keep in mind another important cognitive variable – scholastic aptitude (defined by cito scores) that definitely influences L2 proficiency to a great extent. In order to see which of the learner characteristics is a better predictor of general proficiency, we ran a multiple regression analysis, having added also grades and school type as other factors that can contribute to language proficiency. We got two models (Table 8):

Model R R2 Adjusted R2 R2 Change Sig.

Model 1* .685 .469 .466 .469 .000

Model 2** .707 .499 .494 .030 .000

Table 8 Regression *Predictors: cito score, grades, school type

**Predictors: cito score, grades, school type, motivation, WTC

In the first model, cito score, grades and school type predicted L2 proficiency with R2=0.469 (3, 478); when motivation and WTC were added to the second model, R2 changed by 0.030 and this result was also significant (R2=0.499 (5, 478)). It means that, even after the effects of ‘aptitude’, ‘school type’ and ‘grade’ were controlled for, motivation and WTC still contributed to L2 proficiency.

(37)

Predictors Standardized Coefficients Beta t Sig. School type -.085 -1.985 .048 Grade .605 17.981 .000 Cito score .118 2.713 .007 Motivation .112 2.945 .003 WTC .089 2.332 .020

Table 9 Standardized Coefficients Beta

Although out of all three learner characteristics the cognitive variable ‘scholastic aptitude’ has a higher coefficient (0.118), our affective variables ‘motivation’ and ‘WTC’ still contribute significantly to the students’ English proficiency.

Subquestions:

a. Do the students with high motivation show higher L2 proficiency than the students with low motivation?

To answer this question we took a sample of all 487 students and looked at their motivation means in relation to general proficiency score. The students who scored >=3 on motivation questionnaire were defined as the high motivation group, while those who scored <3 were labeled as the low motivation group. Table 10 shows the number of students in each group, their means and SDs.

Motivation Number Mean SD

General proficiency

score

>=3.00 359 .4372 1.36173

<3.00 128 .1538 1.39887

(38)

An independent Samples t-test (2-tailed) proved that there was a significant difference in general proficiency between high and low motivated students with the former group scoring higher (t(485)=2.007; p<0.05).

b. Which students, integratively or instrumentally motivated, score higher?

To see what kind of students, integratively or instrumentally motivated, show higher L2 proficiency, we calculated a new variable with two levels - integrative and instrumental motivations. Thus, we have 25 students with integrative motivation and 402 with instrumental motivation; 60 students who had the same score on both types of motivation were excluded. Table 11 presents the means and SDs of two groups.

Motivation Number Mean SD SE mean

General proficiency

score

integrative 25 .8885 1.24433 .24887

instrumental 402 .3670 1.38741 .06920

Table 11 Integrative - instrumental motivation

(39)

c. Which students, HAVO or VWO, are more motivated and more willing to communicate?

The bar graph 4 presents the means of the motivation scores for five grades of VWO and HAVO. The 2nd grade of VWO scores slightly higher on motivation while in grades 1, 3 and 4 we see the opposite; HAVO students seem to be a bit more motivated. And only the 5th grade of HAVO seems to score significantly higher than the VWO grade.

Figure 4 Means of motivation per grade

The graphical representation of motivation means for each school type can be found in Figure 5.

(40)

In Figure 6, which demonstrates the means of WTC per each grade, we see that the 1st, 2nd, 3rd and 4th VWO grades seem to be more willing to communicate than HAVO grades and only in case of the 5th grade do HAVO students outscore VWO students.

Figure 6 Means of WTC per grade

The means of WTC for VWO and HAVO students are presented in Figure 7.

Figure 7 Means of WTC

(41)

MANOVA test. The results of multivariate tests showed a significant model: Wilk’s Lambda=0.985, F(2, 459)=3.459, p=0.032, which means that on the combination of motivation and WTC, two school types are different. But separately neither motivation nor WTC reached a significant level with F(1, 460) =1.408; p=0.236 for the first variable and F(1,460) = 2.116; p=0.146 for the second one. Thus, we cannot reject our null-hypothesis and say that HAVO and VWO students differ in their levels of both learner characteristics.

d. Are there differences in the students’ motivation and WTC - in relation to sex

- in relation to age?

To answer this research question all 5 grades of HAVO and 6 grades of VWO were taken into account since the focus is on gender and age of participants and not on the school type. Figure 8 shows that the difference in motivation between male and female subjects is very subtle with girls being only .03 more motivated than boys. Concerning our second learner characteristic, it seems that males score significantly higher on WTC than females (Figure 9).

(42)

Figure 9 The means of WTC score for boys and girls

In order to determine if there were significant differences in the students’ motivation and WTC in relation to sex, a MANOVA test was conducted. The multivariate tests showed that there was a significant difference but based on the combination of two variables (Wilks’Lambda=0.928, F(2, 484)= 18.657, p<0.01). However, the results of between-subjects effects analysis revealed no significant effect of gender on motivation (F(1, 485) = .727; p=0.394) while in case of WTC, there was a significant difference between boys and girls, with the former being more willing to communicate (F(1, 485) = 22.644; p<0.01).

(43)

Figure 10 The means of motivation score for each grade

The results for WTC can be found in Figure 11, where we see that the 2nd grade students score higher than the 1st year students, but the 3rd graders show lower results than the 2nd ones as well as the 5th graders score lower than the 4th graders.

Figure 11 The means of WTC score for each grade

(44)

5. Discussion

In this section I will interpret the results from chapter 4 and discuss them in relation to my predictions and the theoretical insights from chapter 2. I address my hypotheses in the same order as presented in the Results chapter.

Do bilingual HAVO and VWO students differ in their levels of L2 proficiency?

(45)

Main question: Do motivation and WTC affect the students’ L2 proficiency and how?

Since it is known that motivation is an important factor playing a role in development of L2, we assumed that both affective variables influence the students’ L2 proficiency. We found that motivation, WTC and L2 proficiency correlate positively, although showing weak correlation coefficients of 0.169 for motivation and 0.185 for WTC. However, this finding is consistent with other studies which have found positive but moderate correlations between the above mentioned variables and L2 achievement. For example, Sparks et al. (2009) in their longitudinal study found weak to moderate correlations of 0.23-0.47 between the L2 motivation survey and the L2 proficiency measures; in an earlier analysis of the contribution of motivation to L2 achievement by Masgoret and Gardner (2003) the L2 motivation survey showed overall correlations ranging from 0.29 to 0.39. Despite the fact that in our study correlations were low, they were positive which means that if students’ motivation and WTC grow, their proficiency also increases.

(46)

found the modern language aptitude test (MLAT) to be the best predictor of overall L2 proficiency while motivation and anxiety added a small amount of variance to the prediction models. The MLAT alone accounted for 56% of the variance in the model, the anxiety and motivation for an additional 7% and 3% of the variance, respectively.

Although we expected that L2 aptitude would be the most robust predictor of L2 proficiency, our variables under investigation, motivation and WTC, still contributed significantly to the students’ L2 proficiency. Such results prove that students’ L2 proficiency may depend on both cognitive variables - language aptitude, and on affective learner characteristics such as motivation and WTC. This point of view is in line with dynamic systems theory according to which we view each of these variables as dynamic subsystems that have continuous and complex interactions with each other and which cannot exist in isolation from one another.

Subquestions:

a. Do the students with high motivation show higher L2 proficiency than the students with low motivation?

(47)

b. Which students, integratively or instrumentally motivated, score higher?

Two major types of motivation – integrative and instrumental, have been in the centre of L2 motivation research for a long time. In our study we wanted to see how many students could be defined as integratively or instrumentally motivated and we expected that the latter group could show higher scores on L2 proficiency since these students are more driven by grades and learning success in school rather than by natural interest in English community. Out of 487 subjects, 60 could not be assigned to any of two groups because they scored the same on both types of motivation. The majority of the students, 402 people, turned out to be instrumentally motivated while there were only 25 students with integrative motivation. It means that mostly Dutch schoolchildren are motivated by more practical goals such as, for example, good grades, watching English movies and shows, using English as language of communication on vacation with other foreigners and not necessarily English native speakers. Although English is a widely used language in the Netherlands, Dutch schoolchildren do not have a lot of contact with English native speakers; neither are they interested in the cultures of the English speaking world, that is why there are very few integratively motivated students. They see English as a language of international communication, as something that is indispensable for their studies, future careers and overall future success and not solely as the means of connection with native speakers per se.

(48)

c. Which students, HAVO or VWO, are more motivated and more willing to communicate?

In order to get a better insight into the differences between HAVO and VWO students we also conducted some interviews with teachers of the participants. Firstly, the teachers were asked which students – VWO or HAVO, are more eager to learn and if there is a big difference between them. Secondly, I asked what they think about the reasons why VWO students do better in their English classes than HAVO students.

Most teachers confirmed that VWO and HAVO students differ considerably, usually VWO students are more attentive, diligent and motivated. They want to get good grades because they realize that they need them for future studies, for example, for university. If HAVO students work for a teacher, then VWO students work for themselves because mostly they have already set up their goals, they know what they want to become and which career to build. One teacher mentioned that in general HAVO students have lower capacity for studying and, moreover, since they are adolescents, it is difficult for them to concentrate and they are easily distracted, while VWO students can control themselves. In case of HAVO students a teacher needs to make a connection with a class; in other words, to find an approach to those students. Interestingly, the opinion of one teacher of English was that VWO students do not always do better than HAVO students; VWO classes are often self-assured, they think that they are very smart and sometimes do not take tasks seriously, they think more about grades and if the task does not involve grading they do not care and can even be not willing to complete it. In contrast, her HAVO students if they are interested in a task will be eager to do it and they do not care that much about grades.

(49)

we assumed that due to individual differences HAVO students would be less motivated and willing to communicate than VWO students.

Unexpectedly, when we looked at the means for motivation for each class, we saw that only the 2nd VWO grade showed higher motivation than the 2nd HAVO grade, while in the case of all other cohorts HAVO students scored higher than VWO students. The total mean for VWO was 3.17 and for HAVO – 3.2. The means for WTC demonstrate that 1st, 2nd, 3rd and 4th VWO grades are more willing to communicate than the respective HAVO grades and only the 5th HAVO class outscores VWO class. The total mean for WTC for VWO was 3.14 and 3.09 for HAVO. However, the statistical analysis did not show that the differences in motivation and WTC between two school types were significant, which means that our assumption was wrong and actually HAVO and VWO students do not differ in their levels of motivation and WTC.

d. Are there differences in the students’ motivation and WTC - in relation to sex

- in relation to age?

(50)

means that our assumption that girls were more motivated than boys was incorrect; however, in the case of WTC we got the unexpected result that actually boys showed higher WTC than girls, which differs from the previous studies.

(51)

6. Conclusion and recommendations for further research

In this section I will briefly summarize the main findings of the study, discuss limitations of the present study and give some suggestions and recommendations for further research. The purpose of this study was to investigate the impact of affective learner characteristics, motivation and WTC, on L2 proficiency of bilingual HAVO and VWO students. The main finding was that, after we controlled for scholastic aptitude, which is believed to be the most robust predictor of L2 proficiency, motivation and WTC still accounted for part of the variance in L2 proficiency. It means that differences in language proficiency were not only due to aptitude but also due to motivation and WTC. However, it is obvious that motivation alone is unlikely to be sufficient for being proficient in L2.

We saw that motivation, WTC and L2 proficiency correlated positively in our study; that is why the students with higher motivation showed higher levels of English proficiency than the students with low motivation. Interestingly, we found that the majority of the participants were instrumentally motivated, that is interested in learning English for personal practical goals, and only 25 students were guided by integrative motivation. Such unequal division can be explained by the fact that Dutch students are primarily interested in English language as the language of international communication, as a formula for success in their later studies and careers but not because they want to be part of the English speaking community. Instrumentally and integratively motivated students did not show significant difference in L2 proficiency, none of them scored higher than another group.

(52)

motivated students and we cannot say that HAVO students were less motivated than VWO. The reason for that can be that both school types are from bilingual stream which already means that these students are motivated individuals oriented towards learning.

Concerning gender differences, our assumption that female students were more motivated than male students was not confirmed, while in case of WTC the result was unexpected - males were more willing to communicate than females. This finding is different from the previous studies and can be partly explained by the idea that every learning situation is unique and in case of our participants this hypothesis just did not work. As for the second learner characteristic, we could not distinguish any age differences in the levels of motivation or WTC. An expected increase in both variables with each grade was not found, which means that in our case age did not affect the student’s motivation and WTC.

As we can see motivation is not an easy concept to research. The study of motivation can have some limitations; first of all, motivation is something abstract that cannot be observed directly. There are no special and correct measures of motivation. For example, a questionnaire can be a subjective measure since it depends crucially on kind of questions you include in it. Furthermore, motivation is a multifaceted concept which comprises many aspects, such as extrinsic/intrinsic motivation, attitudes to teacher or class, etc. It is almost impossible to include all the dimensions in one study; that’s why we chose to focus on integrative and instrumental distinction in motivation. Last but not least, motivation is not a stable trait, it changes over time and depends on many other factors.

(53)

personality traits but rather dynamic and fluid learner factors, which is in line with the description of motivation by Ellis and Larsen-Freeman: “Motivation is less a trait than fluid play, an ever-changing one that emerges from the processes of interaction of many agents, internal and external, in the ever-changing complex world of the learner” (Ellis & Larsen-Freeman 2006; p.563). Both motivation and WTC constantly interact with other subsystems and they can change and develop over time. As Dornyei notes individual learner characteristics are not monolithic but “constitute complex constellations that are made up of different parts that interact with each other and the environment synchronically and diachronically”. (Dornyei 2010; p.260) These parts keep changing depending on the overall state of the system and in response to external influences, thus making motivation and WTC variables of the dynamic system. Keeping this in mind, there is a recommendation for further research. I would suggest carrying out a longitudinal study from a DST perspective which can help to see the development of motivation, how it changes over a period of time continuously showing some degree of fluctuation. In the present study we used motivation questionnaire once so that motivation was measured only at one moment in time.

Another recommendation would be to look not only at bilingual VWO and HAVO students but to compare bilingual VWO and HAVO with their peers in regular programmes. We would expect significant differences in L2 proficiency and in the levels of motivation and WTC between bilingual and regular streams. A bigger sample size would also probably increase a number of integratively motivated students and maybe then it would be more reliable to compare instrumentally and integratively motivated students in relation to language proficiency.

(54)

there are other motivation-related factors that influence L2 proficiency and that would be interesting to look at, for example, teacher’s personality, teaching styles, etc.

(55)

References

Admiraal, W., Westhoff, G., & De Bot, K. (2006). Evaluation of bilingual secondary education in the Netherlands: Students’ language proficiency in English. Educational Research and Evaluation, 12 (1), 75 – 93.

Baker, C. A. & MacIntyre, P. (2000). The Role of Gender and Immersion in Communication and Second Language Orientations, Language Learning, 50, 311- 341.

Berwick, R. and Ross, S. (1989) Motivation after matriculation: Are Japanese learners of English still alive after exam hell? JALT Journal, 11(2), 193-210.

Brown, H.D. (2000). Principles of language learning and teaching (4th ed.). Englewood Cliffs NJ: Prentice-Hall

Clement, R. And Kruidenier, B.C. (1985). Aptitude, attitude and motivation in second language proficiency: A test of Clement’s model. Journal of Language and Social

Psychology, 4, 21-37.

Crookes, G. & Schmidt, R.W. (1991), ‘Motivation: reopening the research agenda’,

Language Learning, 41, 4, 469-512.

De Bot, K. (2008). Introduction: Second Language Development as a Dynamic Process. The

Modern Language Journal, 92, 166-178.

Deci, E.L. & Ryan, R.M. (1985). Intrinsic motivation and self-determination in human behavior. New York: Plenum Press.

Donovan, L.A., & MacIntyre, P.D. (2005). Age and sex differences in willingness to communicate, communication apprehension and self-perceived competence.

Communication Research Reports, 21, 420-427.

Dornyei, Z. (1990), ‘Conceptualizing motivation in foreign-language learning’, Language

Learning, 40, 1, 45-78.

Dörnyei, Z. (1994). Understanding L2 Motivation: On with the Challenge! The Modern

Language Journal, 78, 515-523.

Dörnyei, Z. (1994). Motivation and motivating in the foreign language classroom. The

Modern Language Journal, 78/3, 273-284.

(56)

Dornyei, Z., (2009). Individual Differences: Interplay of Learner Characteristic and Learning Environment. Language Learning, 59 (Supp/1), 230-248.

Dornyei, Z., (2010). The relationship between language aptitude and language learning motivation: Individual differences from a dynamic systems perspective. In: MACARO, E., ed. Continuum companion to second language acquisition. London: Continium, 247-267.

Dornyei, Z., & Clement, R. (2001). Motivational characteristics of learning different target languages: Results of a nationwide survey. In Z. Dornyei and R. Schmidt (Eds.), Motivation and second language acquisition Honolulu: University of Hawaii, Second Language Teaching and Curriculum Center, Technical Report 23, 399-432.

Dornyei, Z., & Otto, I. (1998), Motivation in action: A process model of L2 motivation.

Working Papers in Applied Linguistics, 4, Thames Valley University, 43-69.

Ellis, N.C., & Larsen-Freeman, D. (2006). Language emergence: Implications for applied linguistics: Introduction to the special issue. Applied Linguistics, 27 (4), 558-589. Gardner, R.C. (1985). The Attitude/ Motivation Test Battery: Technical Report. London, ON:

University of Western Ontario

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

Gardner, R. C. (2001). Integrative Motivation: Past, Present & Future. A paper presented At the Distinguished Lecturer Series, Temple University Japan, Tokyo, February 17, 2001; Retrieved October 30, 2003, from http://publish.uwo.ca/~gardner/ GardnerPublic Lecture1.pdf

Gardner, R.C. (2004) Attitude/Motivation Test Battery: International AMTB Research project (English pdf version)

Gardner, R.C. (2006) Motivation and second language acquisition (paper, pdf)

Gardner, R.C. and Lambert, W.E. (1959). Motivational variables in second language acquisition. Canadian Journal of Psychology, 13, 266-272.

Gardner, R.C. & Lambert, W. E. (1972), Attitudes and Motivation in Second Language

Learning, Newbury House, Rowley, Mass.

The Role of Motivation and Willingness to Communicate in Second Language Learning