Polish and Finnish teenagers’ motivation to learn English: The role of context

: Language learning motivation has been investigated in a number of contexts (Dörnyei and Ushioda 2009). However, there has been only one attempt (Taguchi et al. 2009) at a systematic comparison of models of motivation to learn English by students from different countries. The aim of this study is to investigate the role of context by comparing motivational characteristics and SEM (Structural Equation Modelling) models of language learning motivation of 15 – 16 year-old students from Finland and Poland. The data from both countries was collected using the same motivational questionnaire with a total of 351 students completing it. A number of differences were identified between the two contexts. Finnish learners scored significantly higher on scales of the ideal L2, intrinsic motivation, knowledge orientation, motivated behaviour and peer pressure, whereas Polish learners reported higher levels of international orientation and parental encour-agement. The comparison of models of language learning motivation shows differences in the role of parents and peers, endorsement of goals between the two cohorts. Finally, we discuss the potential reasons behind the differences in motivation between the two contexts and conclude that contexts not only affect motivational characteristics but also the structure of language learning motivation.


Introduction
The paper deals with teenagers' motivation towards learning English in Poland and Finland. A number of previous studies has investigated the structure of language learning motivation using questionnaires in different countries. Typically such studies have attempted to model the structure of motivation in different contexts, for example Canadian (Gardner et al. 2004), Hungarian (Kormos and Csizér 2008) and Chilean (Kormos et al. 2011). However, the problem with such studies is the applicability of the results to different learning contexts. Furthermore, comparisons of the existing results is challenging, as the studies have used a variety of motivational constructs, and thus also, different questionnaires in different countries. Hence, it might be difficult to distinguish whether the difference can be truly ascribed to the specificity of the contexts or the research tool. In this research, the role of context in language learning motivation will be examined by comparing English as a foreign language (EFL) learners in two European countries by using the same questionnaire. Such an approach, that is the comparison of learners in different countries from the point of view of foreign language motivation, has previously been undertaken solely by Taguchi et al. (2009) who uncovered differences in the internal structure of motivation of Japanese, Chinese and Iranian learners of English. The countries under our investigation, Poland and Finland, are interesting contexts to compare for a number of reasons. First even though English is the most commonly studied foreign language in both countries, its role is different as Poland is still a largely monolingual country, whereas in Finland English is, at least for a sizable proportion of young people, "the third national language" (Leppänen et al. 2011). Second, compared to Poland, Finland has enjoyed much closer links to Western Europe and Finnish citizens have had more opportunities to travel freely, whereas in Poland such opportunities arose only after the admission of Poland into the European Union in 2004. Finally, Finland tops ranking of the Human Capital Index (World Economic Forum 2015) and has its GDP per capita is equal to other leading economies in the EU, whereas Poland's economic and human development is substantially lower, although the country is developing rapidly at the moment (World Economic Forum 2015).
Finnish and Polish contexts have not been thoroughly investigated in the light of more recent theories of motivation. Most studies on foreign language learning motivation in Finland were conducted over 20 years ago (Julkunen 1998;Julkunen and Borzova 1997;Laine 1977Laine , 1978Laine and Pihko 1991), and thus are, at least partly, out of date. A more recent study by Julkunen (1998) confirmed that motivation to study English was stronger than in the case of other modern foreign languages, whereas Julkunen and Borzova's (1997) comparison of Finnish and Russian students found higher levels of motivation to learn English among the latter group. Russian pupils had a stronger desire to be good language learners and achieve native-like proficiency but lower actual L2 self-concept than Finnish pupils (Julkunen and Borzova 1997). Motivational research in Poland often adopted a Gardnerian approach (Gardner 2012;Okuniewska et al. 2010;Okuniewski 2012Okuniewski , 2014). An exception to that was Pawlak (2012: 347), whose investigation into the dynamic nature of motivation revealed the presence of short and long-term fluctuations in interest and engagement, although no "meaningful changes in motivational intensity from one lesson to the next" were found.
As can be seen, there exists a need to research language learning motivation in Polish and Finnish contexts, as the role English plays in these countries is different. Thus, the aim of this paper is to investigate the role of context in language learning motivation by answering the following research questions: (1) Are there any significant differences between the scores on motivational scales between the two populations?
(2) Do the models of language learning motivation differ in the two contexts?

English in Polish and Finnish societies
The participants of the current study are Polish and Finnish pupils, aged 14-16, who study English as a foreign language as part of their compulsory education. Polish pupils are from gymnasium, a middle school compulsory for all the student population. Finnish pupils are from lower secondary school, which is the final three years of compulsory education.
English is by far the most frequently studied foreign language in Poland and Finland. In Poland in the school year 2011/12 91.4% of primary and secondary pupils studied English as a foreign language (Łysoń 2012), whereas the same was true for 99.4 % of the pupils in lower secondary school (grades 7-9) in Finland (Tilastokeskus 2012). There are two potential explanations for such high attendance rates in English courses. First is the popularity of English and the second is the wide availability of English language courses in Polish and Finnish schools compared to courses in other foreign languages. There are some differences concerning foreign language education and the role of English in Finland and Poland. Finnish pupils usually begin their first foreign language studies, most often English, in third grade at the age of nine. English has been a compulsory subject in the Finnish primary and secondary education since the 1970s. Instruction in other foreign languages (or English, if it is not the first foreign language) can begin in the fifth grade, but this is done only by 35 % of the student population (Taavitsainen and Pahta 2003). The next language to be studied is the other official language (Swedish for native Finnish speakers) in the 7 th grade, and in the 8 th grade and gymnasium students can choose to start new languages while continuing the studies of the ones they have already started.
Language education in Poland has undergone significant changes since the fall of communism in 1989. Until then, Russian had been the most widespread foreign language taught in state schools with few having access to English classes (Figarski 2008). After the fall of communism, English has rapidly replaced Russian as the dominant foreign language (Poszytek et al. 2005). Currently foreign language education begins as early as the first year of formal education (at the age of seven) (Ministerstwo Edukacji Narodowej [MEN], 2009). In gymnasium, all students are required to take courses in two foreign languages. At the end of this stage of education, they are also required to take an exam in one of the languages studied, the results of which affect their chances of high school admission. In the final compulsory stage of education, high school, the requirement to study two languages persists, although the students might choose to study languages other than the ones started at the primary level. Passing an exam in at least one foreign language after high school is a requirement to begin university education (MEN 2009).
The expected learning outcomes for learning English are similar in Finland and in Poland. According to the Finnish National Curriculum (Opetushallitus 2014), the pupils in question (at the age of 15-16) should be at the A2-B1 level of the Common European Framework of Reference for languages (CEFR). Speaking and writing in English are at A2 and listening comprehension and reading comprehension at B1. The Polish National Curriculum (MEN 2009) specifies that students graduating from gymnasium (age 15-16) should achieve A2 in the CEFR, if they started studying English in gymnasium, or A2+, if they started studying English in the primary school and continued it in gymnasium.
The role of English in Finnish and Polish societies is different today. English is more present in Finnish society than ever. In some areas of life English is used alongside, with or even instead of the two national languages (Finnish and Swedish, yet it needs to be noted that the latter is used by 5% of the overall population and its use is mostly limited to the coastal regions of the country . The increase in the use of English in the Finnish society is not due to the increased number of English-speaking immigrants but, rather, it is the result of several factors, for example, modernisation, urbanization, internationalization and new media (Leppänen and Nikula 2008). The presence of English in Finnish society was confirmed in a recent large-scale survey, where 80 percent of the respondents (n=1495) said that they hear or see English in their living environment (Leppänen et al. 2011). English is also present in the media as most foreign productions are subtitled in Finland, unlike in Poland, where such productions tend to be dubbed. Finland is not, however, an isolated case as proficiency in the use of English is prevalent in Scandinavian countries. In contrast, in Poland, speaking English is largely confined to the generations educated after 1989; thus, only 17% of the population can follow radio or TV news in English and 20% is able to use it in online communication (Europeans and their languages 2012). The status of English in Poland is similar to that of other Central European countries, where foreign language education had centred on Russian before 1989.

Previous research on language learning motivation
The beginnings of language learning motivation research can be traced back to 1950s, when Gardner and Lambert's (1959) first started their work in bilingual Canada. The result of this research was the concept of integrative motivation, which was characterised by "a willingness to be like valued members of the language community" (Gardner and Lambert, 1972: 271). Since then, the field has developed considerably. In this research, similarly to Kormos et al. (2011), we focus on constructs central to motivational research, such as language learning attitudes, goals and, self-constructs, as well as those that can potentially play a role in shaping the language learning environment of teenage learners, namely peer pressure and parental encouragement.
Recently, motivational research has witnessed a transformation due to strong criticism of the Gardnerian concept of integrative motivation (see Dörnyei and Ushioda 2009 for the critique). As a result, new theories attempting to explain language learning motivation have emerged. One of the most well-known is Dörnyei's (2005) L2 motivational self system, which subsumes three components: ideal L2 self, operationalized as a vision of oneself as a successful language learner, ought-to L2 self, or the qualities that one should possess to avoid negative consequences, and language learning experience, defined as the immediate language learning environment. The theory was put to the test (see Dörnyei and Ushioda 2009) and the ideal L2 self has been confirmed to have substantial motivational force. For example, Csizér and Kormos' (2009) structural equation models (SEM) 1 of language learning motivation of three groups of Hungarian language learners of English clearly illustrate a positive link between heightened levels of motivated learning behaviour and ideal L2 self. Similar results have been reported in the context of Chile, where the SEM model replicated the strong relationship between the two variables (Kormos et al. 2011).
Studies using SEM shed more light on the relationship between the ideal L2 self and language learning goals. For example, Csizér and Kormos' (2009) and Kormos et al.'s (2011) models confirm a link between Yashima's (2002: 57) concept of international posture defined as "interest in foreign or international affairs, willingness to go overseas to study or work, readiness to interact with intercultural partners ... and a nonethnocentric attitude toward different cultures" and the ideal L2 self. Knowledge orientation, which points to the important role of English as an instrument to learn more about the world, is directly linked to international posture, which further mediates its influence on the ideal L2 self (Csizér and Kormos 2009;Kormos et al. 2011). These findings suggest that a vision of oneself as a successful language learner might incorporate language learning goals. In fact, Dörnyei (2009) proposed that ideal L2 self includes a promotional aspect of 1 Structural equation modeling (SEM) is a family of statistical methods designed to test a conceptual or theoretical model (Bagozzi & Yi, 2012) instrumentality, or positive professional or personal outcomes connected with mastering English, a hypothesis which has been subsequently supported by Islam et al. (2013) and Taguchi et al. (2009). Therefore, the ideal vision of oneself as a language learner might reflect the extent to which language learners find English useful.
Whereas research suggests that the ideal L2 self is highly motivating, the motivational force of the second component of Dörnyei's (2005) L2 motivational self system, the ought-to L2 self, seems to be limited. This could be because the ought-to L2 self is the expression of extrinsic, rather than intrinsic, motives that tend to be prescribed by the learner's environment. Indeed, the construct was found to be problematic in the Central European context (Csizér and Kormos 2008a, Csizér and Kormos 2008b, and Csizér and Kormos 2009). In South East Asia, where there is more emphasis on the community, the ought-to L2 self emerged and was significantly related to the criterion measure. However, even there its motivational properties appear limited in comparison with the ideal L2 self (Taguchi et al. 2009). Therefore, this construct will not be used in the current study.
The third component of Dörnyei's (2005) L2 motivational self system is language learning experience. It takes into account the environment in which language learning takes place. The environment is, in turn, shaped by the people that it is shared with. Williams and Burden (1997) differentiate between three kinds of significant others in the language learning environment. These are parents, peers, and language teachers. As the latter group is not examined in the current study, it is not discussed in this literature review. The research into the influence of parents on language learning motivation has been present since the beginning of language motivation research (Gardner and Lambert 1972). Parents have been found to be highly supportive of language learning (Csizér et al. 2010;Lamb 2012); yet Bartram (2006) reported that the amount of parental encouragement varies from context to context. In China, the influence of parents on learning English has been found to be mild (Kyriacou and Zhu 2008;Lamb 2012) and limited to goals and attitudes (Gardner et al. 1999;Kormos et al. 2011) rather than language achievement (Alderson et al. 2015). Similarly to parents, peers' influence is rather limited, although Kyriacou and Zhu (2008) noted that peers differ in that respect, with some of them having a remarkably positive effect on language learning. Importantly, Ryan (2001) argued that peers of similar characteristics seem to cluster together, creating an environment that fosters certain goals and behaviours, for example, motivated students might create a group, in which they support each other in language learning.
Another factor often considered to influence investment in language learning is language learning attitudes. Positive attitudes have been found to be related to higher scores on scales of motivational intensity and motivated behaviour (Csizér and Kormos 2009;Islam et al. 2013;Kormos and Csizér 2008;Kormos et al. 2011;Lamb 2012;Papi 2010;Taguchi et al. 2009). Positive language learning attitudes tend to be an indication of intrinsic motivation, which is defined as "an inherent tendency to seek out novelty and challenges, to extend and exercise one's capacities, to explore, and to learn" (Ryan and Deci, 2000: 70) and, as such, is considered to be crucial for the individual's development. Intrinsic motivation is accompanied by positive feelings of interest, curiosity, excitement, enjoyment, competence and self-determination and is only activated when people experience a free choice of activity. Furthermore, intrinsically motivated behaviour is spontaneous, creative and flexible (Deci and Ryan 1985). Shah and Kruglański (2000) specify that an intrinsically motivated activity is perceived as its own end and is pursued in the absence of a reward contingency. In this study, attitudes towards language learning are included in the intrinsic motivation scale.
In the majority of motivational studies, motivated behaviour is chosen to be a criterion measure. A common definition of the concept includes effort as well as persistence in learning a language (Kormos et al. 2011), the elements that are clearly related to Dörnyei's (2001) tripartite definition of motivation. Yet, the concept of self-regulation also mirrors the nature of motivation. According to Zimmerman (1994) , selfregulation is a process controlled by language learners themselves, who first set goals and plan their actions, which are subsequently performed and reflected upon, in order to further adjust goals and improve the efficiency of their effort investment. Therefore, the concept of self-regulation differentiates between active learners, who take responsibility for their own learning process, and passive learners (Zimmerman, 1989). At the same time, the motivated behaviour scale seems to fail to take into account the purposefulness of effort investment and its quality, which is often a factor in how successful language learning is. Thus, this study employs a scale of self-regulation as its criterion measure.
To summarize, students from different contexts may endorse a variety of language learning goals. The contextual factors will also affect the development of intrinsic motivation, the creation of learners' ideal L2 selves and the extent to which students self-regulate. Finally the roles played by peers and parents might also differ. Thus, in order to understand the extent to which the context plays a role in language learning motivation, the two language learning contexts need to be compared.

The motivational questionnaire
Data were collected in Poland and Finland in 2011-2012 using a single motivational questionnaire (Iwaniec 2014) translated into the learners' first language, i.e. Finnish and Polish. Both versions of the questionnaire were validated and their functioning was confirmed by piloting tests prior to testing and by using Factor Analysis. In the questionnaire, a 5-point Likert scale was used. The development of the scales was facilitated by the inclusion of items from already existing scales. Items used by Csizér and Kormos (2009) were incorporated into scales of international orientation, knowledge orientation, motivated behaviour, parental encouragement, and ideal L2 self. The scales of motivated behaviour, parental encouragement, peer encouragement, and instrumental orientation were further influenced by Gardner and Lambert's (1972) instrument. Finally, the scale of self-regulation was adapted from Self-regulating Capacity in Vocabulary Learning Scale (Tseng et al. 2006). The full version of the questionnaire in English was published by the Iwaniec (2014).The following constructs were examined:


Ideal L2 self (6 items) -students' vision of themselves using English successfully in the future. Example: I often imagine myself writing emails in English.


Instrumental orientation (7 items) -students' perceptions of the utilitarian benefits associated with the knowledge of English such as a better job or higher salary. Example: The things I want to do in the future require me to use English.


International orientation (6 items) -students' attitudes towards English as a lingua franca in today's world, enabling them to communicate with other speakers of English. Example: If I could speak English well, I could get to know more people from all over the world.


Intrinsic motivation (9 items) -students' internal motivation to learn English characterized by the urge to learn, explore, and extend one's capacities, accompanied by the feelings of pleasure and satisfaction. Example: I study English because I would really like to be good at it.


Knowledge orientation (6 items) -students' views of English as a necessary tool that enables them to obtain knowledge and information, in addition to it being an important part of education. Example: I think in today's world English is a very important means to get information.


Parental encouragement (7 items) -parents' views on the role of English in today's world and the influence students' parents exert on the processes of learning English. Example: My parents have stressed the importance English will have for me when I leave school.


Peer group pressure (7 items) -the influence of friends and classmates on studying English. Example: My friends have a positive impact on my studying English.


Self-regulation (11 items) -the degree to which students are motivationally, metacognitively, and behaviourally active participants in the process of learning English (adapted from Zimmerman 1989: 4). Example: If there is something I don't understand in English, I do my best to find the answer in a variety of resources.

The participants
The teenage participants from both countries were the same age and had studied English approximately for 7 years as their first foreign language. We examine the motivational data from the Finnish lower secondary school: 115 ninth graders (49 male and 66 female pupils, aged 15-16) from seven different schools and from three Polish gymnasiums: 236 students (112 males, 122 females, 2 with no gender data, aged 14-16). In Poland 166 participants were enrolled in year three, which is the final year of primary education in Poland with students aged 15 or 16-years-old, whereas 70 participants were enrolled in year two (14-15-yearolds). The Polish and Finnish schools were both urban and rural. In Finland four schools were from bigger towns, whereas three were from smaller municipalities, some of which, however, were situated rather close to a larger town. In Poland, the first school was medium sized and located in an urban area, the second one was a medium sized school from a relatively big village and the third one was a small school in a small village. In Finland the data were collected during the DIALUKI project (see e.g. Alderson et al. 2015). The questionnaire was administered via an on-line tool (Webropol) in students' free-time, a paper version was made available for those unwilling or unable to fill in the on-line questionnaire but nobody opted for that. To make responding to the questionnaire easier, it was split into two separate parts that all the students completed. In Poland, the questionnaire was administered during English classes with the class teacher and the researcher present at administration. In both contexts, the completion of the questionnaire was voluntary and the respondents were allowed as much time as they required to do it.

The procedures
We analysed the constructs of the questionnaires using factor analysis. If the Bartlett test of sphericity and the KMO index showed that the factor solutions were adequate, the Principal Components extraction of factors was undertaken. Then the reliability analyses of the scales were conducted and the least reliable items were removed until satisfactory Cronbach alphas resulted (see Table 1). T-tests for independent samples were used to compare the differences in means between the groups, i.e. Polish and Finnish learners. Correlation coefficients were computed to investigate significant links between the variables.

The hypothesized model
Few studies have used SEM to draw models of language learning motivation (Csizér and Kormos, 2009;Kormos et al. 2011). The current model has been influenced by these studies as well as Ford's (1992) Motivational Systems Theory, in which he posits that personal goals, together with emotional arousals (emotional states -here represented by intrinsic motivation) and personal agency beliefs (beliefs about context and one's capability to do something, here self-constructs) are some of the most important motivational components. In the hypothesized model (see Figure 1), we suggest that goals are the basis of language learning motivation with two groups of important others having a direct impact on them. Goals influence intrinsic motivation and the ideal L2 self, which, in turn, affect the levels of self-regulation. Additionally, there is a link between important others and intrinsic motivation. The decision to put goals at the basis of the model was informed by Ford (1992), who argued that goals behave like anchors, structuring one's behaviour. Similarly, this could be applied to contextual influences (Ford 1992), which in the model are, to a certain extent, mirrored in the scales of parental encouragement and peer pressure. These groups are considered by Williams and Burden (1997) to be important others who play a substantial role in language learning, affecting not only language learning attitudes, subsumed under the intrinsic motivation scale, but also language learning goals (Gardner et al. 1999;Kormos et al. 2011). As parental encouragement and peer group pressure are both important, they are interconnected in the current model. Yet parents might affect peers as peers are other language learners influenced by their own parents' views. Similarly, the three goals, instrumentality, knowledge orientation and international orientation, can be assumed to be correlated. In many ways, they might reinforce each other, for example both instrumentality and international orientation represent utilitarian benefits associated with language learning (Gardner 1985;Gardner et al. 1999). Likewise, the sheer definition of knowledge orientation as a tool to obtain broader access to information (Csizér and Kormos 2009;Kormos and Csizér 2008) implies utilitarian values. To be truly motivational, goals need to be considered attainable (Ford 1992). Thus, there exists a link between goals and the ideal L2 self that, according to Dörnyei (2005), is a possible self. Further, the ideal L2 self implies intrinsic motivation (Dörnyei 2009). Finally, ideal L2 self and positive language learning attitudes and enjoyment spur language learning behaviours, measured here using self-regulation (Ajzen 2005;Csizér and Kormos 2009).

Figure 1: Hypothesised model of language learning motivation
The hypothesized model was analysed in Mplus 6.11 (Muthén and Muthén 2010) separately for Finnish and Polish data. During SEM analysis, non-significant relationships (paths) between variables were removed. In the final step of the analysis, the goodness fit of the models was determined 4. Results

Factor analysis
First, we analysed the items for all the constructs using Principal Component extraction of factors separately for Polish and Finnish data. None of the constructs loaded on multiple factors, which indicates that the questionnaire was measuring the given constructs well. Also the Bartlett test of sphericity and the KMO indexes showed that the factor solutions were adequate. Items with low loadings in, at least, one dataset were removed from both datasets until satisfactory loadings (>.63) of each item were obtained. Then, we ran the reliability analyses on the resulting scales and removed all such items from datasets that decreased the reliability of the construct in, at least, one dataset. The resulting data has 40 items and 10 scales. The results of factor analysis and reliability analyses are presented in Table 1. As items were removed until they obtained satisfactory loadings for both populations and the number of items had to be limited to conduct SEM, the number of items used in the analysis is substantially lower than the number of items used in the questionnaire. Table 2 shows the results of independent samples t-tests. Language learners from two countries differ significantly in mean values in seven out of nine scales. On the one hand, the Finnish pupils have on average a stronger ideal L2 self, intrinsic motivation, knowledge orientation, motivated behaviour and peer pressure than the Polish pupils. The Polish pupils, on the other hand, have on average a stronger international orientation and parental encouragement than the Finnish pupils. The effect sizes (Cohen's d) of these differences are small, which means that the variance between samples is less than 5% (Pallant 2010) with the exception of the ideal L2 self scale where it is large. The scales on which the participants from the two countries did not differ significantly were instrumental orientation and self-regulation.

Models of language learning motivation in Finland and Poland
We conducted SEM analyses on the hypothetical models and removed insignificant links. The fit indices of both models can be found in Table 3. The link that was excluded from both models was the one illustrating an impact of parents on intrinsic motivation. There were several differences between the models. Whereas the model of motivation in the Polish context (see Figure 2) retained most of the links hypothesized, a number of links in the Finnish model (see Figure 3) were found not to be significant, and therefore, had to be removed from the model. Consequently, in the Finnish model, the ideal L2 self affected self-regulation only indirectly via intrinsic motivation. Whereas peer pressure was significantly related to the three goals and intrinsic motivation in the Polish model, its influence was limited to international orientation in Finland. Similarly, parental encouragement retained more significant links in Poland than in Finland. In Poland, it was related to all three goals, while in Finland to instrumentality and knowledge orientation only. Finally, there are some differences between the behaviour of goals in the model. In the Polish models, there were significant links between all the goals and the ideal L2 self as well as intrinsic motivation. However, in the Finnish model, none of the goals was significantly related to intrinsic motivation. Moreover, international orientation was the only goal to significantly impact the ideal L2 self.

Differences between the scores on motivational scales between the two populations
In our first research question, we asked whether there were any significant differences on the motivational scores between the two populations. The results of the t-tests suggest that the Finnish students were more motivated than their Polish counterparts as they scored significantly higher than the Polish students on five scales, although their results were lower on two scales. The generally higher levels of motivation among Finnish learners than Polish learners might be also attributed to the sampling as the Polish sample included more learners from typically rural areas than the Finnish sample. Previous research (Lamb 2012) shows that the prevalence of working class population in the countryside often leads to lower motivational scores of rural learners as compared to learners from towns and cities.
The Finnish students seemed to enjoy studying English more. Higher results on the intrinsic motivation scale coincide with higher effort investment in language learning on the part of Finnish learners, as implied by the results on the motivated behaviour scale. It needs to be noticed that the effect sizes of both results are small. This suggests that context plays a role but it is limited. A relatively small discrepancy between the two populations might explain the lack of significant differences on the self-regulation scale. The mean scores on self-regulation for both populations are similar to those of Hungarian learners on the subscales of self-regulation (Csizér and Kormos 2014).
Finnish learners also scored higher than Polish learners on the knowledge orientation scale. The higher levels of knowledge orientation in the Finnish population compared to the Polish population might be explained by the differences in access to information in their L1s. Polish is a relatively vital language with approximately 40 million speakers, whereas the number of native speakers of Finnish is just below five million people (Lewis 2009). Therefore, there may be more sources of information available in the Polish language, whereas Finns might need to resort to the use of English in order to find important information, particularly online. In fact, as many as 36% of English-speaking Finns reported using English daily or nearly every day compared with 11% of Poles (Europeans and their languages 2012).
The largest gap was identified between the scores on the ideal L2 self scale of the two groups with the mean for the Finnish population at 3.74 and the Polish population at 2.71. This result may be explained by the fact that English is much more widely used in Finland than in Poland due to urbanisation, modernisation and new media. Young Finns encounter situations in which they need to use English on a daily basis (Leppänen et al. 2011). This makes English an important part of the life of young Finns, thus preparing the necessary basis for the creation of the ideal L2 self (Markus and Ruvolo 1989). In contrast, young Poles are less likely to use English outside the classroom, as Polish is the dominant language and the number of English speakers is relatively low. Thus, the Polish students are less likely to perceive English as an important aspect of their life and construe an ideal L2 self.
The Finnish students seem to receive more support from their peers than their Polish counterparts. It needs to be noted that the difference is small and the means suggest a rather limited amount of encouragement from peers in both countries. Yet the slight advantage on the part of the Finnish students might be attributed to generally higher motivational scores achieved by this population compared to the Polish learners.
Whereas in most cases, higher scores were observed in the Finnish population, Polish learners were found to score higher on scales of parental encouragement and international orientation. High scores on the parental encouragement scale have been previously reported for secondary school learners of English in context historically similar to Poland, namely Hungary (Kormos and Csizér 2008). In the same study, Kormos and Csizér (2008) also found levels of international posture similar to those from the current study. Yet again, the effect sizes for both results are small, suggesting that the differences between the two populations are limited.

Differences in the structure of motivation between the two populations
In our second research question, we asked whether there are any differences in the structure of the models for the two populations. As described in the results section, a number of such differences were found. In the Finnish model the only variable directly related to self-regulation is intrinsic motivation. It accounts for over 77% variance in self-regulation. In the Polish model, self-regulation is linked to both intrinsic motivation, which accounts for 38% of variance in self-regulation, and ideal L2 self, which explains a further 5% of variance in self-regulation. Thus, the motivational force of the ideal L2 self seems to be mediated via intrinsic motivation in the Finnish model. This could be because, as has been found in the current study, Finnish learners demonstrate more positive attitudes and higher enjoyment of English language learning than their Polish peers.
Another observable difference is the role of important others between the two countries. In Poland both parents and peers affect the three language learning goals measured. Additionally, peers influence intrinsic motivation. In Finland, however, peers affect only international orientation whereas parents influence knowledge orientation and instrumental orientation. The Polish results are consistent with those reported by Csizér and Kormos (2009), whose model of language learning motivation for Hungarian secondary school learners of English linked parental encouragement to both knowledge and international orientation. Thus, it seems that whereas in Poland parents and peers seem to support the endorsement of a number of language learning goals, in Finland parental encouragement is connected to goals that can bring about direct benefits such as better jobs or access to information, whereas peers influence international orientation, which is more internalised by Finnish learners.
The two models differ in the influence of goals on the ideal L2 self and intrinsic motivation. In Poland, all goals are directly related to the ideal L2 self. Further, knowledge orientation and instrumentality influence intrinsic motivation, whereas in Finland international orientation is the only goal directly related to the ideal L2 self and no goals have direct links with international orientation. However, it needs to be noticed that in Poland the links between instrumentality and intrinsic motivation and knowledge orientation and ideal L2 self are negative, which means that the low score on one variable predicts a high score on another variable. The finding that there might be a negative relationship of knowledge orientation to the ideal L2 self is not entirely surprising as previous research has showed that knowledge orientation is not directly related to ideal L2 self but its influence is rather mediated by other goals, such as international orientation (Csizér and Kormos 2009;Kormos et al. 2011). The negative links between the two goals and ideal L2 self and intrinsic motivation might suggest that these goals are not fully internalised by Polish learners, but rather imposed by the context. This is supported by the fact that in the Polish model both instrumentality and knowledge orientation are clearly related to parental encouragement. As such, it might be concluded that international orientation is the main goal fostering motivation in both contexts.

Conclusion
In our questionnaire study, we compared the motivation of Finnish and Polish learners of English by examining mean scores on motivation variables and the structure of language learning motivation in both contexts. Context was found to affect both scores on the individual scales as well as the overall structure of language learning motivation, although most differences were of small effect size. The analysis of means scores revealed that the Finnish students have stronger ideal L2 self, enjoy studying English more, endorse knowledge orientation to a greater extent, receive more peer support and claim to invest more effort into studying English than their Polish counterparts. However, the Polish students reported more parental encouragement and stronger endorsement of international orientation than their Finnish peers. Similarly to Taguchi et al. (2009), we observed some differences in the SEM models of language learning motivation in the two countries. Whereas in Finland, self-regulation is predicted by intrinsic motivation only, in Poland both intrinsic motivation and ideal L2 self contribute to self-regulation. Parents and peers affect all goals in Poland but in Finland parents' impact is limited to instrumental and knowledge orientation, whereas international orientation is related to peer pressure. Finally, international orientation is the only language learning goal affecting ideal L2 self in Finland, whereas in Poland all goals examined are related to ideal L2 self and intrinsic motivation, although the influence of international orientation is the clearest.
This research has shown that language learning motivation differs from context to context. Future research employing the same motivational questionnaire might examine the extent of these differences between various contexts. Considerably more work will need to be done to determine the relevance of language learning goals for different learner groups and the relationship between important others and other motivational constructs. Moreover, it might be useful to employ qualitative methods in order to examine the proposed models with smaller groups of students.
Our study is not, however, without limitations, the most important of which is sample size as only 115 Finnish students filled in the questionnaire, compared to 236 Polish students. The number of schools involved was also different (three in Poland and seven in Finland). The questionnaire was also administered in different ways, namely online in Finland and in paper form in Poland when the researcher was present, which could possibly affect the results, although participation in the study was in both cases voluntary.
The study provides a number of pedagogical implications. First, the role of intrinsic motivation is vital in fostering self-regulation in both contexts. Therefore, teachers should strive to provide high quality positive language learning experiences, where students' needs of autonomy, relatedness and competence are met, as it can facilitate the development of intrinsic motivation. Second, the ideal L2 self is also closely related to intrinsic motivation and self-regulation. Providing time in English classes for students to imagine themselves as successful learners on a regular basis might stimulate the development of the ideal L2 self (Dörnyei 2009). Finally, the students are motivated to learn English because they believe it is an international language that will aid communication with a large number of other English speakers. Creating such opportunities in the language classroom or outside of them, for example via e-technologies, inviting English speaking guests, organising student exchanges, might have a positive effect on students overall motivation.