Volition is Key Self-Perceived Willingness to Communicate and Actual Willingness to Communicate among Swedish EFL Learners

Degree project

Volition is Key

Self-Perceived Willingness to Communicate and Actual Willingness to Communicate among

Swedish EFL Learners

Author: Josef Rudberg Supervisor: Ibolya Maricic Examiner: Charlotte Hommerberg Date: June 7^th, 2019

Subject: English

Abstract

It is generally assumed that in order to learn a language, learners need to master reading, writing, listening and speaking. However, merely possessing the skills and abilities to communicate is not sufficient; one must have the will to use the language. In order to formulate a model that can accurately describe the willingness to communicate, Mac- Intyre et al. (1998) formulated a model that attempts to do just this. The goal of this study is to verify the veracity of the Willingness to Communicate (WTC) model in terms of to what extent students’ self-reported WTC correlate with their actual WTC in the class- room. To accomplish this, this study utilized interviews with Swedish EFL students in southern Sweden, based on the WTC framework, and classroom observation in order to collect data on said connection. The results of this study showed that self-reported WTC and actual WTC do not necessarily correlate with one another; for some students, they spoke English more than they reported, some spoke less, and others’ WTC reflected their self-assessed WTC. Although the WTC model could account for the students’ actual WTC, it could not account for this discrepancy. However, this discrepancy may not be due to the model itself, but rather to the situational factors that, to a certain degree, influenced the outcome of this study. Based on this study, teachers are recommended to consistently speak English, as this raises the students’ own WTC, and to encourage English usage among students themselves. For future research, it is therefore recom- mended that non-linguistic factors be taken into account, possibly through factor analysis, in order to produce more nuanced data, as well as conducting a longitudinal study.

Keywords

Interviews; classroom observations; self-reported Willingness to Communicate; Swedish EFL students; upper secondary education; Willingness to Communicate; WTC.

Acknowledgement

First of all, I would like to extend my gratitude to my supervisor, Ibolya Maricic, whose

advice proved invaluable. Also, I would like to thank my examiner, Charlotte Hommer-

berg, as well as my opponent, Kristoffer Keinvall, for their valuable input. Finally, I wish

to thank the teacher and the students that participated in this study. Without them, this

study would not have been possible.

Table of Contents

1 Introduction _________________________________________________________ 1

1.1 Aim and Research Questions ________________________________________ 2

2.1 Willingness to Communicate _______________________________________ 2 Layer VI: Social and Individual Context ________________________________ 3 Layer V: Affective-Cognitive Context ___________________________________ 4 Layer IV: Motivational Propensities ___________________________________ 5 Layer III: Situated Antecedents _______________________________________ 6 Layer II: Behavioral Intention ________________________________________ 6 Layer I: Communication Behavior _____________________________________ 7 2.2 Summary and Discussion __________________________________________ 7 2.3 Research on WTC ________________________________________________ 8

3.1 Participants ____________________________________________________ 10 3.2 Method ________________________________________________________ 11 3.2.1 Interviews __________________________________________________ 11 3.2.2 Observations ________________________________________________ 12 3.2.3 Validity and reliability ________________________________________ 13 3.3 Data analysis procedure ___________________________________________ 14 3.4 Ethical Considerations ____________________________________________ 15

4.1 Interviews _____________________________________________________ 15 Layer VI: Social and Individual Context _______________________________ 16 Layer V: Affective-Cognitive Context __________________________________ 18 Layer IV: Motivational Propensities __________________________________ 20 Layer III: Situated Antecedents ______________________________________ 21 Layer II: Behavioral Intention _______________________________________ 23 Layer I: Communication Behavior ____________________________________ 24 4.2 Observations ___________________________________________________ 24 4.2.1 Observation 1 – Student 1, 2, and 3 ______________________________ 25 4.2.2 Observation 2 – Students 4, 5, and 6 _____________________________ 26

5.1 Self-perceived WTC and actual WTC ________________________________ 26 5.2 Pedagogical Implications __________________________________________ 30

Appendices ___________________________________________________________ I

Appendix A: Information and Consent Form (Swedish) _______________________ I

Appendix B: Information and Consent Form (English) _______________________ II

Appendix C: Interview Questions (Swedish) ______________________________ III

Appendix D: Interview Questions (English) _______________________________ V

Appendix E: Transcripts of Interviews (Swedish) __________________________ VII

Student 1 (S1) ____________________________________________________ VII

Student 2 (S2) _____________________________________________________ XI

Student 3 (S3) ___________________________________________________ XVI

Student 4 (S4) ___________________________________________________ XXI

Student 5 (S5) __________________________________________________ XXVI

Student 6 (S6) __________________________________________________ XXXI

Appendix F: Transcripts of Interviews (English) _______________________ XXXVI

Student 1 (S1) _________________________________________________ XXXVI

Student 2 (S2) ____________________________________________________ XL

Student 3 (S3) __________________________________________________ XLIV

Student 4 (S4) _________________________________________________ XLVIII

Student 5 (S5) ___________________________________________________ LIII

Student 6 (S6) __________________________________________________ LVIII

Appendix G: Observation Schedule ___________________________________ LXIII

1 Introduction

It is generally accepted that foreign language learning consists of four skills: reading, listening, speaking, and writing (Harmer 2007:265). It is therefore not far-fetched to assume that in order to attain foreign language competence, one should strive to achieve mastery of all of these skills. One implication of this assumption for the teaching profession is that in order for students to improve their skills, the students should be given opportunities to activate and practice these skills. However, this leaves one essential factor out of the equation: volition. One integral part in initiating any kind of action is intent. As such, it is not enough for the teacher to simply present opportunities to practice, the students themselves must also have the intent or willingness to activate and practice their skills. Therefore, the first and most important question ought not to be “How does one improve one’s language skills?”, but rather “How does one increase one’s willingness to practice?”. This is a question that has been at the core of the language teacher profession since its founding, which is why it is not surprising that extensive research has been conducted in order to find an answer to this question from multiple angles, e.g.

whether or not there is a correlation between WTC and grade level (Bergström 2017), students’ perceived classroom participation (Bernales 2016), and the effects of group size and topic on WTC (Zarrinabadi et al. 2014).

One notable attempt to formulate a theory that incorporates volition within a foreign language learning context, at least with regards to foreign language learners’ oral production and interaction skills, is the well-known theory called Willingness to Communicate (henceforth WTC). Formulated by MacIntyre et al. (1998), this theory posits that the learner’s willingness, or lack thereof, to communicate is dependent on a number of factors, such as their perceived competence, their attitude(s) towards the foreign language, and their relationship with their interlocutors. Although a promising theory, “[w]ork on WTC is in its infancy” (Ellis 2008:698). Thus, this research project attempts to further research on the phenomenon of Willingness to Communicate. More specifically, the aim of this study is to explore whether or not there is a connection between L2 learners’ self-reported Willingness to Communicate and their actual foreign language use. This is accomplished through a mixed-methods approach involving inter- views with Swedish upper secondary school students and subsequent classroom observations of said students.

Understanding the underlying connection between learners’ Willingness to Communicate and their actual foreign language use is undoubtedly valuable for foreign language teachers. Skolverket

(2011), a legal body of the Swedish government whose jurisdiction involves education, writes in the curriculum for English that students should, for example, be given the opportunity to practice their ability to express themselves with variation and complexity (2011:53). Since students are to be presented with the oppor- tunity to practice, then it follows that it is crucial for students to also have the willingness to seize the opportunities when they arise.

1 The English title is Swedish National Agency for Education.

1.1 Aim and Research Questions

The aim of this study is to explore the correlation between Swedish EFL learners’ WTC and their actual, observed WTC. In order to delimit the aim of this study, the following research questions are addressed:

1) To what extent does self-perceived WTC correlate with the actual WTC among upper secondary school students in a school in the south of Sweden?

2) What are the pedagogical implications of the findings for the language teaching profession with regards to increasing the learners’ WTC?

This paper utilizes the following structure: first, the theoretical framework is presented.

Second, the research method, along with specifics on participants as well as practical and ethical considerations, are presented. Then, the data is summarized, followed by an analysis and a discussion. Finally, the conclusion and suggestions for future research are offered.

2 Theoretical Framework

In this section, one major theory with regards to foreign language learning will be presented, namely, Willingness to Communicate. This theory will lay the foundation for the formulation of the interview questions, which will be described later in Section 3, as well as assist in the analysis and discussion of the gathered data.

2.1 Willingness to Communicate

Willingness to Communicate (WTC), a theory used in e.g. language acquisition research, is based on the idea that when a learner is ready to communicate in the foreign language, (s)he actively seeks opportunities that facilitate such interaction. In fact, even if no opportunity presents itself to the learner, (s)he will create an opportunity. With this in mind, the goal of language teaching is not only to offer the opportunity, but also to promote the willingness to engage in communicative interactions (Maftoon and Amiri 2011; Wood 2016; Zarrinabadi et al. 2014).

MacIntyre et al. (1998) propose a model that describes the variables that influence

an individual at all times (see Figure 1). This model is similar to Maslow’s Hierarchy of

Needs: the lower levels must be fulfilled before the person can ascend to the next level

(Maslow 1943). It is just the same with the model of WTC, i.e., when the learner ascends

the model, the learner attains more control over the communicative tasks at hand:

Figure 1: Heuristic Model of Variables Influencing WTC (MacIntyre et al. 1998:547)

In their model, MacIntyre et al. (1998) divide the concept of WTC into six distinct layers.

Layers I, II, and III represent situation-specific influences on the learner’s WTC, whereas layers IV, V, and VI represent stable, near-unchanging influences.

Layer VI: Social and Individual Context

Roughly speaking, communication involves the interaction between two participants:

society and the individual. MacIntyre et al. (1998) characterize society in two ways:

through its structural characteristics and perceptual and affective correlates. The first term refers to the vitality of a language group derived from its demographic representation in society. The consequence of this is twofold: if a language group is a majority, that group gains more respect, and members of other groups will want to communicate in the majority language, but if a group is a minority, that group eventually becomes outnumbered and subsequently loses its adherents. The second part does not concern itself with the actual demographic group, but rather with how it is perceived by members of other groups. When a group is perceived in a positive light, interactions will thrive. When outsiders have a negative perception, discrimination may occur. A possible solution to prevent discrimination, according to MacIntyre et al. (1998:556), is that the minority group assimilate and acquire the foreign language. This may, however, be hindered in a number of different ways, for example if parents, through their influence, impede the acquisition of foreign languages or if society in general hinders assimilation.

As far as the individual is concerned, personality, i.e. the way in which one interacts with members of another group, influences the WTC of the individual. For this reason, research on personality traits has spawned a taxonomy called “The Big Five” (Rothmann

& Coetzer 2003) which has five common character traits, all of which play a part in

determining a learner’s WTC to some degree: Openness to new experiences,

Conscientiousness, Extraversion/Introversion, Agreeableness, and Neuroticism, the

acronym of which is OCEAN. For instance, a learner who is more open to new

experiences may be more willing to seek out opportunities in which (s)he confronts a new situation in which the L2 is employed. However, one ought to be careful not to come to the conclusion that these qualities make up the good language learner, but rather that these qualities may be present in a good language learner; these features are in no way crucial.

It should be taken into consideration that all these layers, illustrated in Figure 1, occur simultaneously and have different meanings depending on the individual. Nevertheless, the lower layers must be fulfilled in order to ascend to the next level (MacIntyre et al.

1998:558).

Layer V: Affective-Cognitive Context

Macintyre et al. (1998) conceptualized that within the individual, there are two opposing forces in learning a foreign language: integrativity, i.e. the desire to belong to the foreign language community, and fear of assimilation, i.e. the fear of losing the connection with the native language community. Depending on which one is stronger, the individual may either avoid foreign language communication or immerse themselves completely in the foreign language, potentially threatening their linguistic and cultural heritage. In addition to these opposing forces, there is the motivation to learn the foreign language, which is more concerned with the language itself, either for intrinsic or extrinsic reasons. This motivation can be enhanced by positive experiences associated with the foreign language, for example through an exchange program in a country where the foreign language is a majority language, thus ensuring foreign language exposure. However, this may lead to fear of assimilation, or even to the individual swinging like a pendulum between integrativity and fear of assimilation (MacIntyre et al., 1998:552f.).

Within this layer, the social context is also taken into account. Firstly, the interlocutor’s social status, gender, age, and their relationship to the learner plays a role in how they engage with one another; one would communicate in a certain way with a professor in linguistics and then proceed to speak in another way with close friends.

Second, the situation in terms of time and place plays a role; libraries tend to be as quiet as possible, while birthday parties and other celebrations allow jubilant exclamations.

Third, the purpose of the exchange is significant. The presentation of a dissertation sets a different standard than a eulogy during a funeral. Finally, the medium through which communication takes place influences the act of communication. For example, when one speaks face-to-face, one has access to body language, a communication layer that is missing during a phone call (MacIntyre et al., 1998:553f.).

Also, communicative competence, i.e. the ability to use language in a functional manner in different circumstances (Canale & Swain 1980), has a significant impact on one’s own WTC; if a learner trusts their capacity for successfully formulating a message that is well-received by their interlocutor, the fear of embarrassment is lowered, thus making the learner more willing to communicate.. This type of competence consists of at least five interconnected abilities (Lundahl 2012:139ff.):

1. Linguistic competence: Knowledge of the grammatical rules of a language, e.g.

syntax, morphology, phonology and vocabulary, which is necessary to

communicate successfully and accurately.

2. Discourse competence: The arrangement of words, structures and sentences in a cohesive and coherent order.

3. Pragmalinguistic competence: The ability to express oneself in accordance with a given situation, such as ordering fast food.

4. Sociocultural competence: The ability to express oneself in harmony within the general social and cultural context.

5. Strategic competence: Tools and strategies that enable compensating for deficits in certain areas of the foreign language, such as describing a word that one has forgotten.

Layer IV: Motivational Propensities

When discussing motivation, MacIntyre et al. (1998:550f.) suggest three categorizations:

1) interpersonal motivation, that is, motivation derived from the individual, 2) intergroup motivation, that is, the collective expectations and social norms of the group, and 3) self- confidence in foreign language use, i.e. confidence in one’s skills and abilities.

Interpersonal motivation seems to be connected to the individual’s sense of control and affiliation. Control means that the learner acts in accordance with his or her will in a social environment on the basis of his or her perceived competence. Usually, control acts as a force that limits the cognitive, affective, and behavioral freedom of the participants if someone else possesses said control. For example, teachers exercise control over the language learning situation. Students, in turn, communicate their opinions and/or answer the teacher’s questions within the parameter of what the teachers control. This kind of communication is, in essence, hierarchical, in which it is clear to the participants who is in control. As such, communication may be encouraged or discouraged, not just by the teacher, insofar as (s)he allows the students to participate, but also by the students, insofar as they freely accept and participate in the hierarchically stratified situation. In short, although all participants have control, they may have it to varying degrees (MacIntyre et al. 1998:550). Affiliation deals with the amount of interest that the individual has in establishing a relationship. This desire is affected by multiple factors, such as the similarity between the participants, physical proximity to one another, and how often the interlocutors meet. On top of that, personality also plays a role; some personalities may have a stronger need to feel connected to others, while other personalities prefer solitude (MacIntyre et al. 1998:550).

In contrast to interpersonal motivation that relates to the individual, intergroup motivation directly relates to all of the participants in a certain group. MacIntyre (1998:551) argues that collective motivation affects the individual’s motivation towards foreign language learning; if the group’s attitude is positive, the individual’s motivation might be the same, and vice versa. Similar to interpersonal motivation, control and affiliation also seem to play a role, although not within the group, but rather between groups; when an interlocutor from a superordinate language group

interacts with a member of a subordinate language group

, it is incidentally seen as a way of maintaining and reinforcing social positions, i.e. control. The same can be said of affiliation;

2 A language that is a majority group and/or is officially recognized as a native language.

3 A language that is not officially recognized as a native nor majority language.

interactions between groups are seen as a way of connecting and establishing a relationship (MacIntyre 1998:551).

Finally, L2 self-confidence, as it relates to motivational propensities, deals with the relationship between the individual and the L2. More specifically, it corresponds with the individual’s belief in their ability to communicate in the L2 in a manner that is well- adapted to the situation. There are two components to this: the foreign language learner’s cognitive self-evaluation of their own L2 competence, and the anxiety that (s)he may feel when using the L2. Research has found that employing these two components in a productive manner can heighten WTC (MacIntyre et al. 1998:551).

Layer III: Situated Antecedents

Much like the previous layer, the individual’s sense of affiliation and control are additional factors that increase the desire to communicate. A sense of belonging usually occurs in an individual when the one with whom they are communicating is physically nearby, meets them regularly, and is similar to them in some respects. Moreover, the participant who has the greatest control, real or perceived, may be the one who decides which language is used (MacIntyre et al. 1998:548f.).

Depending on the learner’s perceived state of affiliation and control, (s)he may experience varying levels of trust and anxiety towards the interlocutor and the situation in its entirety. These two concepts must be separated in two ways: trust and anxiety as permanent personality traits and as transitory, situational states of being. An individual may by nature be more trusting and/or more anxious. This does not mean, however, that a more trusting individual will never be anxious. Rather, permanent characteristics and situational perceptions are connected to one another in a complex interplay. Thus, an individual may have a perceived sense of trust and lack of anxiety in one situation but not another. The most optimal state for the language learner is a state of maximum (perceived) trust and minimal anxiety (MacIntyre et al. 1998:549).

One weakness of this part of the model is the difficulty in determining where trait- like anxiety ends and situational anxiety begins, i.e. if a person can by nature be more anxious, how does one distinguish this from anxiety caused by the situation? As is generally the case with other theories that pertain to human psychology, this model deals with concepts and constructs that are difficult to verify empirically, as there is no way of detecting, measuring or weighing anxiety within a person’s mind. Even if one could quantify anxiety, the next problem is how one compares the anxiety of one individual with that of another. Given the fact that human psychology is a highly complex phenomenon, then it is reasonable to assume that how individuals perceive their own anxiety may vary widely, which further problematizes any kind of comparative research on this phenomenon.

Layer II: Behavioral Intention

This level postulates that students not only have to have the opportunity to use the foreign

language, but also have the motivation to communicate. It must be noted, however, that

these two factors, opportunity and willingness, are not what actually coaxes the student

to communicate; the intention is crucial. For example, if a teacher asks the class to explain

the English tenses, a certain number of students will raise their hands. In other words, a

raised hand indicates the intention to communicate, even though only one student will have a chance to utter their response. Therefore, a non-actualized communication request must also be considered WTC. Finally, the intention to communicate must also be paired with a sense of control over the action and the certainty that the action can be successfully completed (MacIntyre et al. 1998:547f.).

It should be mentioned, however, that intent itself does not appear out of thin air.

When a student raises their hand, (s)he exhibits the intent to communicate, but in order to communicate, (s)he must have understood the question, formulated an answer in their mind, and then, if given the opportunity, vocalized it. In short, students who are competent have sufficient self-confidence to have WTC (MacIntyre et al. 1998:548), meaning that willingness is in and of itself not the only driving force behind WTC.

Layer I: Communication Behavior

Although it is important for the teacher to offer many communication possibilities, it is equally important to arouse the students’ desire to actively seek communication possibilities, i.e. to promote their willingness to communicate (MacIntyre et al.

1998:547). In this regard, the presence of genuine L2 communication, or lack thereof, plays a role for this layer. An arguably important factor in the language acquisition process is to what extent L2 communication is authentic, that is, if it occurs naturally or not, e.g. when reading a news article in the foreign language, when using the foreign language at work, or when interacting with friends and family. Such genuine communication options may not be available in the average classroom, especially if said classroom is not situated in the country in which the foreign language is the majority language.

2.2 Summary and Discussion

To summarize, WTC is a highly complex model that postulates the existence of a multitude of factors that influence learners when it comes to language and their subsequent willingness to speak it. One potential strength of this theory is that it does not attempt to over-simplify the concept of willingness to communicate, rather, it acknowledges that the process of language acquisition is a highly complex one and that in every interaction, a significant number of factors must be taken into consideration.

Nevertheless, when it comes to accepting this specific theory and its subsequent model, caution is advised.

It is not clear why MacIntyre et al. (1998) have chosen the specific order and

placement of all of the factors, for instance, why they chose to place “Personality” as a

foundational element within Layer IV, rather than a secondary or tertiary element in the

upper layers (see Figure 1). One might even question if the order of these factors matters

at all because MacIntyre et al. (1998) themselves claim that all of these factors are present

in every interaction. This raises two problems: (i) if the order of the factors does not

matter, then one would be hard-pressed to justify the use of a pyramidic model; a

pyramidic model carries with it the underlying assumption that the lower layers are

foundational, and the following layers come into play at a later point of time. If one wishes

to convey the idea that a multitude of factors influences the individual simultaneously, it

would be preferable to use a form of circular figure, in which items such as “Personality”

are closer to the center, and more external factors such as “Intergroup Attitudes” are further from the center. (ii) Merely stating that a multitude of factors influences the learner simultaneously is in itself insufficient, both out of a theoretical perspective as well as out of a practical one. Merely detecting a factor that influenced a phenomenon does not explain why or how they are causally linked. In order to promote effective promotion of WTC, teachers need to be able to quantify each factor in order to successfully allocate resources; if the learners lack in “State Communicative Self-Confidence”, for instance, the teacher needs to be able to detect this and then determine how much of the available resources can be spent on mitigating this particular factor.

2.3 Research on WTC

WTC can be seen as a relatively new theory, given its recent formulation. As such, research into the theory has been primarily focused on validating it (Ellis 2008:697). One notable result of such studies is that WTC seems to be highly context dependent.

Cao and Philp (2006) performed a study that focused on the dual nature of WTC, i.e. trait-like WTC and situational WTC. As their method, they employed questionnaires, classroom observations, and interviews. The results indicated that WTC as a trait could predict communication tendencies (2006:485). However, situational WTC proved difficult to correlate with the participants’ self-reported WTC. Instead, the findings seem to indicate that situational WTC correlates to some degree with the context, rather than how the participants perceived their WTC, and as such, the researchers asked the participants during the interviews to name a few factors that might influence their WTC.

The three highest were group size, self-confidence, and familiarity with interlocutor respectively. One major weakness of this study is the admittedly generic nature of the self-report questionnaire; the participants were asked to rate their WTC in a non-specific sense, rather than within an EFL classroom context. One could argue that if the researchers had devised a classroom-specific WTC questionnaire, the self-reported WTC might have shown a stronger correlation with their actual classroom WTC. The reason for this is that if participants are asked to estimate their WTC in a general sense, it might be the case that they conflate their WTC in one context with another; a student may have a high WTC while playing a computer game, but when (s)he enters the classroom, the situational WTC might be lower, as was indeed the case for some of the participants in their study (Cao and Philp 2006:485ff.). It is therefore not surprising if their general WTC does not correlate with their actual, classroom-specific WTC.

In contrast to the above study, Dörnyei and Kormos (2000) found a strong

correlation between learners’ WTC and their attitude towards the English course. In their

study, they wanted to determine to what degree students’ verbal behavior in oral task

situations was determined by non-linguistic and non-cognitive factors, such as social

status. To accomplish this, they administered an oral task in the students’ L2 and, later

on, in their L1, which measured the number of words as well as turns used, in order to

objectively determine to what degree language has an effect on their WTC. The students

were also given a C-test, a language proficiency test validated for Hungarian EFL

learners, to determine their global language proficiency, and two self-report

questionnaires, through which the researchers gained information about group

cohesiveness, learners’ interrelationship, and WTC in the L1. Based on these data, they concluded that only the situation-specific factors – such as attitudes towards the English course, attitudes towards the task, and linguistic self-confidence – correlated significantly with the criterion measures (Dörnyei & Kormos 2000:286). These results were contrary to past research that suggests, among other things, that social variables should have affected their WTC (Gardner & MacIntyre 1993). To examine this conundrum, the researchers decided to separate students into two groups based on their attitude towards the task: high task-attitude (HighS) and low task-attitude (LowS). Then, they repeated the procedure. The results indicate that for the HighS group, WTC has a strong correlation with both criteria, i.e. how many words they used and how many turns they took. In contrast, the LowS group showed no such correlation. In conclusion, WTC seems to be affected to a high degree by the learners’ attitude towards the task itself. To their credit, this is a valuable piece of information; it stresses the importance of actively engaging the students to raise their motivation for the task, as well as for the teacher to not give haphazard tasks, but rather to actively construct tasks that the students find interesting.

However, the same can be said of any subject; if a math teacher constructs an interesting task, students will be more engaged and thus perform better. Therefore, the question is:

have Dörnyei and Kormos (2000) measured Willingness to Communicate, or have they rather measured “Willingness to Perform a Task”? If so, then one could argue that they have stepped outside of language didactics and into general didactics. This is not to say that their research has no value. The point is that if one wishes to unravel an aspect of language didactics that is as of yet undiscovered, one ought to draw conclusions that are not so general that any teacher, through their own experience, could discover them on their own.

Much akin to the previously discussed research articles, Kang’s (2005) study

investigates the duality of WTC, i.e. as a trait and situation dependent. The research

utilized interviews, videotaped conversations, and stimulated recalls, through which the

author reached the conclusion that situational WTC was primarily affected by three

psychological conditions that interact with each other, i.e. excitement, responsibility, and

security. These are not fixed traits that exist prior to social interaction, but are rather

constructed within the interaction, according to Kang’s (2005:288) model, which in turn

affect the learner’s WTC. Since excitement, responsibility, and security are socially

constructed, Kang (2005) proposes that WTC itself is a purely situational concept, rather

than a fixed trait. There is however an inherent flaw within Kang’s proposed model: it

does not take personality into account which is an integral part of the lower levels of the

WTC pyramid model. It is not far-fetched to assume that whether or not an individual

feels excitement is dependent on the situation. However, one cannot deny that any social

interaction occurs between two or more individuals, each of whom brings their own

personality into the interaction. Therefore, if an individual is inherently more neurotic

than others, for instance, they might be less trusting of others and may, therefore, have

trouble connecting with their interlocutors. In short, limiting WTC only to the situation

itself creates just as incomplete a picture as if one would limit WTC only to traits; neither

trait-like WTC nor situational WTC can be excluded if one wishes to create a

comprehensive model for WTC.

In short, the validity and nature of WTC are as of yet a disputed matter.

Nevertheless, further research is still of interest, as WTC constitutes a connection between already established constructs and language proficiency (Ellis 2008:697).

3 Material and Method

In this section, the material and method of this study are presented and discussed. First, information with regards to the participants is presented. Next, the choice of method, i.e.

a mixed-method approach, will be discussed. Finally, the data analysis procedures, i.e.

how the collected data will be analyzed, and ethical considerations will be presented.

3.1 Participants

One English teacher at an upper secondary school in the south of Sweden was randomly chosen

, and this teacher, in turn, was asked to find students at English 5 who are willing to participate in this study. Consequently, the researcher had no influence on the selection process, lessening, to some degree, the selection bias. One might argue however that if the purpose of this study is to find students who are willing to speak, then the ones who truly are more willing will naturally volunteer, thereby skewing the data. Although a valid point, one potential counterargument is that the purpose is not to find the most willing students, but rather to see to what degree their words align with their actions. This kind of data would thus have appeared, regardless of how the informants were selected.

A total of six students at the upper secondary school level participated in this study.

Three of them are studying in the Social Science program, and the other three in the Humanist program. Both of these programs are oriented towards the social sciences, however, the Humanist program focuses more on language and culture, whereas the Social Science program aligns itself with societal studies in general, such as behavioral science and communication (Skolverket 2011:45 and 49). The students are currently studying their first year at upper secondary school, and as such, they are studying English 5

together. Four students were female and two were male. Although this particular sample is not evenly distributed in terms of gender, one might argue that it is representative of the distribution of boys and girls in Sweden in general; in 2018, 19% of all applicants for the Humanist program were boys, and for the Social Science programs, the number of male applicants was 36% (Skolverket 2019). Their ages range between 16 and 17 years old. Four of the students have Swedish as their mother tongue, one student has German, and one student has Albanian. One potential objection to this fact is that if these two students do not have Swedish as their L1, but rather learned it at a later point in time, then EFL would not be classified as their L2, but rather their L3. Two potential counter-arguments can be raised: (1) Although these two students learned Swedish at a later point in time, it nevertheless became apparent during the interviews that they had

4 A homepage called “random.org” was used (https://www.random.org/lists/). On this page, the names of every English teacher was entered into the box, after which the “Randomize” button was pressed. The name on top of the list was chosen. The other teachers could be contacted in case the first teacher could not or did not wish to participate.

5 In upper secondary school, English is divided into three levels: English 5, English 6, and English 7, which according to CEFR places them on the level of B1 during English 5.

attained a mastery of Swedish that made them virtually indistinguishable from the rest of their classmates (see Appendix F). Though one could argue that they should not be classified as native Swedish speakers, they nevertheless could be classified as such based on their skills. (2) Regardless of which language is their mother tongue, all of these students are studying English as a foreign language, meaning that the circumstances are the same for everyone. One could say that if some of these students had English as their L1, then one could make a strong case for how this may influence their WTC, which was not the case for any of the students, however. Finally, the number of years of studying English ranges from 7 to 10 years. A summary of the gathered background information can be seen in Table 1 below (informants are referred to as Student 1, Student 2, etc.).

Table 1: Background Information on Study Participants

Student 1 Student 2 Student 3 Student 4 Student 5 Student 6

Gender Female Female Female Male Female Male

Age 16 17 16 16 16 16

L1 Swedish Swedish Swedish German Swedish Albanian

Years studying English

7 10 7 7 9-10 6-7

Program Social Humanist Humanist Humanist Social Social

3.2 Method

As mentioned above, this study utilizes a mixed-methods approach to the WTC of Swedish EFL learners. One significant advantage of a mixed-method approach is that the disadvantages of one method can be compensated by the advantages of another (Denscombe 2009:153f.). For example, observations enable gaining information about an event or phenomenon as it occurs, such as when a student raises their hand to answer the teacher’s question. However, merely observing an event does not reveal the reasons behind an action. On the other hand, interviews offer an opportunity to discover the reason or reasons why someone acts in a certain way, yet, it does not confirm if the informant’s word accurately reflects reality. Subsequently, interviews combined with observations offer a multifaceted set of data that enable a deeper analysis of the concept of WTC.

3.2.1 Interviews

As mentioned in the introduction of this study, the goal is to determine to what extent

self-reported WTC correlates with actual foreign language use in the classroom. The word

self-reported necessitates some kind of data collection method in which the EFL learners

themselves provide information. There are two methods to achieve this: questionnaires

and interviews. Fundamentally, these methods are the same; a question is posed to which

an informant responds. The distinct difference between them is the medium through

which this occurs: one is spoken, and the other is written. Interviews, however, have some

advantages in comparison to questionnaires. First of all, an interview is often perceived

as more personal, which might raise the informant’s willingness to respond (Denscombe

2009:268). Second, if an informant gives an answer that is interesting or unexpected, the

interviewer has the opportunity to ask a follow-up question, which potentially leads to

information that otherwise would not have appeared (Denscombe 2009:267f.). Finally, if the interviewer successfully creates a bond with the interviewee, albeit temporarily, the response frequency may be higher than otherwise, i.e. they may be more willing to answer the questions (Denscombe 2009:246).

The ultimate goal of this study is to verify the validity of WTC. As such, the interview questions themselves were formulated so as to discover the informants’ WTC.

First, the theory itself was studied extensively in order to understand it as much as possible. Then, a number of questions were formulated for each layer based in part on the number of subcomponents, e.g. Layer III: Situated Antecedents gave rise to a question that asked for the students’ sense of affiliation and control (see 2.1 Willingness to Communicate). For a translation of the interview questions from Swedish into English, see Appendix D.

Although it could be argued that Swedish students have a high proficiency in English (Stockholm University 2016 [www]), the interviews were nevertheless conducted in Swedish. When conducting an interview, it may be preferable if it is done in the language in which the informant is most comfortable, as (s)he is able to formulate and express their thoughts and opinions more precisely. To further increase the chances of obtaining rich data, this study utilized semi-structured interviews, which means that the interview has a set of topics and questions that need to be addressed, but the informant has every opportunity to elaborate and develop their answer (Denscombe 2009:234f.).

Finally, to ensure the reliability of this study, the interviews were recorded via a cellphone app called “Voice Recorder” and later transcribed (see Appendix E). For simplicity’s sake, the transcripts were ordered in a way that clearly delineates what the interviewer and the interviewee said; when two interlocutors communicate, their turns overlap at times, however, such information is not necessary for this study; the purpose of this study is not to determine to what degree the interviewee’s responses overlap with the interviewer’s, but rather to gather information on their self-perceived WTC.

3.2.2 Observations

When conducting a research project that incorporates observations, one ought to keep in mind that observations have certain limitations, mainly due to the limitations of human beings themselves, physiologically as well as mentally. First of all, human memory is selective; the brain chooses what to keep and what to discard automatically, which may be an obstacle in recording data accurately. Second, the human senses, such as sight and hearing, have their own innate limitations. Third, perception itself is selective, i.e. what a researcher perceives may be affected by a number of factors, such as levels of stress, energy or even what (s)he intentionally chooses to focus on (Denscombe 2009:273ff.).

To compensate for the shortcomings of perception, this study utilized what is

known as an observation schedule (Denscombe 2009:274ff.). It is an aid in which the

researcher creates a protocol or table that clearly and systematically delimits what is to

be observed. In this case, the objective is to gather data on the informants’ WTC, and as

such, the schedule was written for and adapted to the purpose of this study and included

things such as initiating a conversation with the teacher in English, which is to be seen as

the informants’ desire to use English being actualized. For the complete observation

schedule, see Appendix G.

Once the students had been interviewed and the transcriptions were finished, the observations took place on two separate occasions. Observing multiple students at once might have been disadvantageous, as it would, to some degree, be more difficult to record a greater number of events that occur simultaneously, especially for only one researcher.

In order to ensure that enough attention could be given to each student, the first three students were observed during the first occasion, and the latter three were observed during the second occasion. The first one occurred during a class that was 65 minutes long. The students were asked by the teacher to find and correct four specific grammar mistakes in a newspaper article, either individually or in pairs. The second observation occurred the following day, in which students continued with the same task, which lasted for 80 minutes

. Performing the observations during two lessons with similar content has one advantage: if the factors that shape the circumstances are the same, or nearly the same, one could argue that the two lessons would produce the same data, while at the same time taking into consideration the argument presented above that enough attention needs to be given to all students.

During these observations, the author of this study sat down at a location where a non-intrusive overview of the previously interviewed students was achieved; it is advisable to achieve a balance between being as invisible as possible, so as to lower the effect of the observer’s presence, and being able to detect anything of interest (Denscombe 2009:279).

3.2.3 Validity and reliability

Two concepts can be argued to be the cornerstones of scientific research. The first one is reliability, understood in a research context as to what extent a result can be replicated and still be the same (McNeill & Chapman 2005:9). The second is validity, which means to what extent a tool accurately measures what it intends to measure (Denscombe 2009:425; McNeill & Chapman 2005:9).

In terms of reliability, the advantages of this study are at least threefold. First, all of the informants were asked the same questions, meaning that any differences in the collected data could be attributed to the fact that not all answers are exactly the same.

Second, the same observation schedule was utilized in the observation of all participants.

Third and finally, the data were analyzed within the same theoretical framework, i.e.

WTC. Perhaps the most significant disadvantage of this study, however, is the small sample size. If one wishes to collect statistically reliable data of a phenomenon, a sample size of six participants is not sufficient. This also has consequences for the generalizability of the conclusions drawn from the small sample size; it would be far- fetched to assume that one can apply data from six upper secondary school students to the rest of Sweden. Although such a small sample size would be detrimental for quantitative research, it may prove adequate for more qualitative research; the small sample size is compensated by a deeper analysis of the data. Another potential weakness

6 The first lesson had to be ended 10 minutes earlier due to the teacher having to leave earlier for a conference. Otherwise, the lesson would have been 75 minutes long.

of this study is the selection process. Although the selection was randomized

, there is no guarantee that the results of the study will be the same, in terms of which students volunteer to participate; if a researcher were to repeat this study, those who volunteer could potentially provide information that is entirely unlike the data of this study. Finally, when it comes to self-reporting, one should always take into consideration that human beings are, to some degree, prone to inaccuracy and subjectivity. This is not to say that the informants did not tell the truth, but rather that what they believe to be true might not necessarily be the case. For example, one student could have the impression that they always use English in the classroom, yet, the observation contradicts this belief.

As for validity, this study could be argued to be situated on the stronger side of the spectrum. First of all, this study does not solely rely on the informants’ self-reported attitudes and behavior patterns; it also attempts to verify whether or not their words reflect reality through observation. Second, the interview questions, as well as the observation schedules, were grounded in a theory relevant to the theme of this study. The data was then used in order to verify the veracity of said theory. Third, the participants were not chosen by the researcher, meaning that researcher bias was minimized. It should be noted, however, that the validity of this study is dependent on its reliability; in order to come to the conclusion that the data are valid, one has to assume that the method through which they were gathered is reliable. Finally, one thing to take into consideration is the Observer’s Paradox (Denscombe 2009:81), i.e. if one aims to observe an event, ensuring that one’s presence does not affect this phenomenon may be nigh impossible. It could very well be the case that since the students are aware of the researcher’s presence, they might alter their behavior. The same can be said of the Interviewer Effect: there is no way of verifying to what extent the informants altered their behavior or their responses when being interviewed (Denscombe 2009:244). One potential solution could be to film the students with a hidden camera. This is not advisable, however, mainly for ethical reasons:

the participants must be informed of their participation and give their explicit consent, which is not possible if they do not know that they are being observed in the first place.

Also, parental consent would have to have been given, which would have required a greater amount of time. The best, and simplest, solution is to be as inconspicuous as possible.

3.3 Data analysis procedure

As explained above (see Section 3.2.1), the interview questions and the observation schedule were formulated so as to gather data in accordance with the theory of WTC. For the data analysis, the same theoretical framework was used in order to verify if the theory can explain the data, with the ultimate goal of verifying the theory as a whole. Thus, the results will be shortly presented in Section 4, followed by an analysis and discussion of the data with regards to the research questions in Section 5.

7 The teacher was randomly chosen, but the teacher asked the students if they wanted to participate. If only one part of the selection process was randomized, perhaps it would be more accurate to claim that it was semi-randomized.

3.4 Ethical Considerations

When it comes to scientific research, there are certain guidelines that need to be in place in order to ensure that researchers do not cross ethical and moral boundaries, especially so when the research pertains to human beings. As such, the Swedish Research Council

has put forth four requirements that need to be fulfilled in order for the research to be ethically valid: the information requirement, the consent requirement, the confidentiality requirement, and the use requirement (Vetenskapsrådet 2002).

The first requirement means that the informants must be informed of the purpose of the study, their role in the study as well as the fact that their participation is voluntary, and that they thus have the right to withdraw their participation (Vetenskapsrådet 2002:7).

Thus, the informants received both oral and written information of this before the interviews proceeded, as well as contact information (see Appendix A). The second requirement means that the informants must give their explicit consent to the researcher to utilize the collected data (Vetenskapsrådet 2002:9). In certain cases, such as if the participants are younger than 15, parents must also give their consent. For this study, the informants were given the opportunity to give their consent in written form (see Appendix A), and, as these students were older than 15, the consent of their parents was not needed.

In accordance with the third requirement, any information that may reveal the identity of the informants must be stored with the utmost integrity (Vetenskapsrådet 2002:12). As such, all information that may reveal the identity of the participants was anonymized.

Finally, the fourth requirement states that data may not be distributed or otherwise given to any person or organization outside of the research, i.e. the data may only be used for the purpose of the study (Vetenskapsrådet 2002:14). In order to ensure the security of the gathered data, the recordings, which may reveal the identity of the informants, were deleted once the transcripts were complete.

4 Results

This section presents the results of the interviews and observations. First, the participants’

answers to the interview questions will be presented in a summarized form as well as within each layer of the theory in greater detail, beginning with Layer VI and ending with Layer I. After that, the results of the observations will be summarized, both in text and in a table.

4.1 Interviews

Overall, the replies of the students indicate a relatively high amount of WTC, most notably with their classmates and their teacher. For a summarized version of the interviews, see Table 2 below:

Table 2: Summary of Students’ Answers

Layer Question Student 1 Student 2 Student 3 Student 4 Student 5 Student 6

IV 1) English in Sweden

A little Yes Yes Not really Yes Yes

8 The Swedish title is Vetenskapsrådet.

2) Talkative personality

Yes More open in English

Often Yes Depends If I have

to

V 3) Residence Yes Yes Yes No Yes Yes

4) Days Daily Thursdays

and Fridays

Thursdays and Fridays

Daily Mostly in school

Mostly in school

5) Interlocutors Brothers Friends Student 2 Close friends

Friends Close friends

6) Meaningfulness Yes Yes Yes Yes Yes Yes

VI 7) Classmates’

influence on WTC

Yes Yes,

mutually

Yes A little bit Yes Yes

8) Linguistic confidence

Depends Yes, in school

Yes Yes Yes Yes

III 9) Comfort Yes Yes Yes Yes Yes Nothing

special

10) Hindrances No No No No No No

11) Trust Yes Yes Yes Yes Yes Yes

12) Anxiety No No No No No No

II 13) Seizing the opportunity

No Yes, if

comfortable

Yes Yes No If I can

I 14) Opportunities presented

Yes Yes Yes Yes Yes -

For the sake of analysis, the excerpts below have been translated into English (see Appendix F). The original Swedish responses can be viewed in Appendix E below. A white background indicates the interviewer’s line, and a grey background is from the interviewee. The chosen excerpts are not to be seen as exhaustive, but rather relevant examples illustrating the topic at hand.

Layer VI: Social and Individual Context

The students’ responses to the first question, i.e. to what extent the presence of English in Sweden has affected their use of English in the classroom, displayed at least two patterns: either it had little or no effect on their English use, or if it did, it had an impact in terms of vocabulary and idiomatic expressions. This is most apparent when comparing the answers of Student 1 and 6:

Student 1, Question 1

You could say that English has a strong presence in Sweden, in commercials, movies, music and so on. Would you say that it has affected how much you use English during class?

During class, I usually talk in English when I get my tasks, but not so much, you know, normally when we just talk. Even if (name of teacher) wants it. So, I don’t think it has affected me that much. Maybe a little, I don’t know.

Contrast this response with Student 6:

Student 6, Question 1

You could say that English has a strong presence in Sweden, in movies, music, commercials, you know. Would you say that this has affected how often you use English during class, and if so, how?

During class?

Yes.

Well… do you mean English classes or classes in general?

English classes specifically.

Ok. Yeah, you might use expressions that are used in series and movies, that you don’t hear otherwise, that… you might walk up to someone and just “what’s up?” It’s maybe (unintelligible).

Of notable interest is that one student mentioned that she sometimes forgets a word in Swedish, and instead, she uses the English equivalent:

Student 5, Question 1

You could say that English has quite a strong presence in Sweden, movies, music, basically everywhere. Would you say that this has affected how often you use English in the classroom?

Yes, I think so.

In what ways?

Sometimes, maybe you don’t find the words in Swedish, so you might say it in English instead because you know it in English but not in Swedish.

In short, all students concede that the presence of English in society has an effect on their language use in school, for some more than others.

For the second question, students were asked if their personality, in their own opinion, is the open and talkative type that wants to use English. A majority of them responded that they see themselves as open and talkative, e.g. Student 1, 2, 3, and 4.

However, it should be noted that although most of them viewed themselves as open and talkative, for two of the students, it was conditional. At least two main reasons can be deduced: the relationship between the interlocutors (Students 4 and 5); and fear of negative evaluation (Student 5). For Student 4, for example, he feels shy when he meets strangers, but once he gets to know them better, he feels more comfortable in his English use:

Student 4, Question 2

Would you say that your personality is very talkative and that you’re very willing to use English, or other languages?

Yeah, I think so. I might be a little bit shy in the beginning if I get to know new people, but, after that, I begin to talk quite a lot, and I like using English. Also, if I were to point out something, my grammar in English is really bad, so it’s nice to be able to practice.

In other words, with whom he is speaking affects his WTC, specifically their closeness, or lack thereof. Student 5 mirrored the same views, and she also added fear of negative evaluation:

Student 5, Question 2

About your personality, would you yourself say that you’re an open and talkative person who gladly uses English?

Not exactly, but maybe a little… it depends on who you’re talking with. If you’re talking to an English person who comes from England, it becomes harder to speak English, but if you talk with your friends, it becomes a lot easier, cause you dare to talk more.

It’s like, you’re on the same level, so it’s not.

Yeah, exactly, you’re not afraid of making mistakes.

Layer V: Affective-Cognitive Context

When it comes to living in a country in which English is the main language, virtually all of the informants were positive to the idea. Great Britain, England or London were the main choices of places to live in (Students 2, 3, 4, 5, and 6). The desire to live in an English-speaking country can be seen in the response of Student 1, for example:

Student 1, Question 3

If you had the opportunity to live in an English-speaking country, would you take the chance?

Yes.

Definitely?

I would.

Would you like to explain why?

I think it would be fun to speak in English every day and to develop my language abilities in English.

The students gave a variety of reasons for why they would like to move to an English-

speaking country. Student 1, for example, had the view that living in countries like

England would improve her language skills. Student 6 also mentioned football as his

reason for going to Britain. An additional reason could be classified as an intrinsic desire

to learn more about the language for its own sake:

Student 2, Question 3

Maybe you’d like to explain exactly why you would like to go to an English-speaking country?

Well, to get closer to the language, closer to the culture that the language comes from.

And, I am quite interested in languages, so it’s always fun to learn about the roots of a language.

One student, however, expressed the view that leaving his family and friends behind would be less desirable:

Student 4, Question 3

So, if you had the opportunity to live in an English-speaking country, would you do it?

That’s a difficult question. I don’t know, it would have been fun to use English regularly, but I don’t know if could leave everyone behind, family and friends and everything.

As for the fourth question, i.e. if there are any days in particular during which they use English more often, all of them responded that it happens most often during the days in which they have English class, during which some of the students mentioned that their teacher regularly encourages them to use it as often as possible. As for extramural use of English, the students indicated that it did not happen as often, but if it did, it is usually connected to some kind of activity that is in English, such as watching movies or series, reading books, or playing computer and video games. In one case, there was even an opportunity to use English at home:

Student 3, Question 4

And, would you say that there are certain days sometimes where you more often or seldom use English, and why do think it’s so, in that case?

Well, I use it mostly during English class. Not so much outside of the classroom, so it’s actually on Thursdays and Fridays where I use it.

Would you like to use it more besides Thursdays and Fridays?

Yes, a little, maybe. I think it’s nice to be able to use Swedish as well. Sometimes we have conversations in English at home.

So, you have relatives or parents who come from England?

My mom lived in the US and Singapore when she was little. Back then, she spoke a lot in English, a lot in American schools. But, it’s not like I’ve spoken it since I was little.

But, it’s good to practice with her because she knows it very well.

When it comes to with whom they speak in English, it becomes apparent that it mostly

occurs with friends (Students 2, 3, 4, 5, and 6). Student 4, for instance, responded thus: