EXPLORING THE RELATIOSHIP BETWEEN LANGUAGE LEARNING BELIEFS AND FOREIGN LANGUAGE CLASSROOM ANXIETY

: Language learning beliefs and foreign language classroom anxiety present two extremely important factors in language learning. This paper explores these two factors and their complex relationship in students at the Preschool Teacher Training College in Vrsac and Teacher Training Faculty in Belgrade (Vrsac campus). Both quantitative and qualitative research models were implemented in the study. The quantitative analysis used BALLI (Beliefs about Language Learning Inventory) and FLCAS (Foreign Language Classroom Anxiety Scale) as instruments, and latter implemented a correlational analysis to see which language learning beliefs had a strong connection with anxiety levels. Ten language learning beliefs showed a significant correlation with foreign language classroom anxiety levels. These beliefs were placed into five belief categories which were used as a starting point in the qualitative analysis. For the qualitative analysis language learning histories were used. The process produced two additional language belief categories which showed a strong connection with foreign language classroom anxiety. The research proposes which language learning beliefs teachers should first promote in order to establish a positive attitude towards language learning without high levels of anxiety along with other suggestions for the change and promotion of language learning beliefs and the lowering of classroom anxiety.


Introduction
The success of language learners is connected to a myriad of factors which influence their language acquisition process. Foreign language anxiety and beliefs about language learning influence the foreign language learning process in all of its stages (Mohebi & Khodadady, 2011;Kralova, 2016). Horwitz highlighted the importance of focusing on the learners affective factors since they influence their willingness to engage in activities, without which foreign language learning in the classroom is impossible (Basaran & Cabaroglu, 2014). The study of learners' perceptions and feelings help us answer an important question: What causes language learners to approach a specific language learning task differently (Mori, 1999). As important individual differences, foreign language anxiety and language learning beliefs help 1 palinkasevic@gmail.com 2 suncokrety@gmail.com us understand how learners perceive the foreign language classroom, the activities within it, themselves as active participants, their classmates, teachers and the language itself.

Beliefs about Language Learning
Beliefs are central constructs in every discipline which deals with human behavior and learning. Individuals define and comprehend the world and themselves through beliefs. They are of key importance for defining tasks and our behavior (Mohebi & Khodadady, 2011 p. 291).Believing is giving meaning to the world and to ourselves and as we do this we are also forming our identities in this world (Barcelos, 2015).
Beliefs are present in every segment of our lives and are essential to our functioning. Language learning beliefs are just one subgroup of beliefs. Mori's study (1999) on the relationship between epistemological beliefs and language learning beliefs displayed the multidimensionality of personal language learning beliefs and the independence of each belief dimension.
As research on language learning beliefs developed through the years so did the definitions of the concept of language learning beliefs. Studies on language learning beliefs started with early research in individual differences in order to better understand the differences between successful and unsuccessful learners (Mori, 1999). In these studies beliefs were understood to be mental fixed constructs in one's mind (Barcelos, 2015). An example of another early definition is given by Cotteral (1999 as cited in Mohebi&Khodadady, 2011) who defined language learning beliefs as individual characteristics whose investigation is useful for making teachers aware of different learner types.
As the traditional approach made room for the appearance of the contextual, discursive and sociocultural approaches so did the following definitions occur (Kalaja et al. 2018).Berant and Gwozdenko (2005 as cited in Basaran & Cabaroglu, 2014) defined language learning beliefs as learners' metacognitive knowledge about themselves, their goals and needs. Barcelos (2000Barcelos ( , 2001Barcelos ( , 2004 as cited in Aragão, 2011) defined beliefs as dynamic and contextualized cognition which shapes understandings and influences action. She also indicates that "belief is a way of perceiving the world that generates confidence to act upon matters accepted as true, but which may be questioned in the future" (p. 304). Basaran & Cabaroglu (2014) propose that learners beliefs are highly complex, diverse and interrelated, that each learner brings his/her own unique set of beliefs, perceptions, knowledge, experiences and preferences to the classroom. Kalaja et al. (2018) indicate that "Beliefs are conceptualized as shared, necessarily both social and individual, and consequentially contextual. Beliefs reflect individual, social and institutional viewpoints; the learner has interacted with during his/her lifetime" (p. 2). They also state that "Beliefs are co-constructed in interaction with others and are related to language, since language mediates all interactions and actions. As such, they are part of a socialization process of learners becoming members of specific communities. Thus, they are connected to the micro context of ideologies, power structures and status in a given society as these relate to the learning of L2 and their use." (p.5).
The importance of beliefs is reflected in the role that they play in the language learning process. Beliefs effect attitudes, motivation and the learners' efficiency in the classroom settings. If the learners and the teachers have very different beliefs about how languages are learned and should be taught, it may lead to poor confidence in the teacher, dissatisfaction in the course and low achievement of the learners (Basaran & Cabaroglu, 2014). Because of differences in language learning beliefs students with similar proficiency may approach the same task differently. Thus, beliefs are important for understanding individual differences in learners (Mohebi & Khodadady, 2011). From learners beliefs we can understand their expectation of, commitment to, success in, and satisfaction with foreign language classes (Hou, 2013). Barcelos (2015) listed five possible reasons for exploring language learning beliefs of our students: 1. to understand any potential resistance to new methodologies; 2. to understand the differences between teachers and learners beliefs; 3. to better understand language learning difficulties; 4. to become familiar with their motivation; 5. to understand the use of certain language learning strategies.
Beliefs can be divided into core beliefs and peripheral beliefs.Core beliefs have four main characteristics: 1. they are more resistant to change; 2. they are more interconnected with other beliefs, because of this they "communicate" better with and influence other beliefs; 3. they are more connected to a person's identity and self; 4. they are connected to our emotions (Barcelos, 2015). Peripheral beliefs are not so stable nor intertwined and are more easily changeable. The more an individual thinks about certain beliefs the more embedded they become in the neural pathways, making them more resistant to being changed.
Because of the strong influence of language learning beliefs on the students learning process, teachers should promote insightful, healthy and facilitative language learning beliefs. The effects of language learning beliefs on learning exist independantly from ones ability to learn. Students who believe, for example, that language learning is a product of hard work might outperform those who are equal or better initially but who believe that language learning depends on an inborn ability which they do not have in the desired extent (Mori, 1999). Successfull learners develop insightfull beliefs about the language learning processes, their own abilities, and the use of effective learning strategies which have a facilitative effect on learning (Mohebi & Khodadady, 2011). When learners change their beliefs and with them their emotions, they are constructing new identities within the social and power bounds of their context (Barcelos, 2015).
Various authors proposed different strategies for changing foreign language beliefs and promoting positive language learning beliefs, some of which are listed below: • Focusing on active learning and goal setting (Sim, 2007 as cited in Basaran & Cabaroglu, 2014); • Detrimental beliefs should be confronted as early as possible and self-regulated activities should be actively encouraged (Cabarogly & Roberts, 2000 as cited in Basaran & Cabaroglu, 2014); • Aragao (2011) proposes talking to and becoming familiar with students individually so that idealized models could be questioned and a positive classroom environment promoted. He stresses that students should reflect on their progress expectations and beliefs, that language learning narratives should be written, read and shared in the classroom so that learners could see what they have in common and that the challenges that they may be facing are not uncommon. This could also be achieved through visual representation of their feelings and beliefs which would be interpreted and presented or debates could be organized about language learning beliefs, learning styles and strategies; • Promoting discussions by writing language learning histories and reading texts which help learners and teachers metacognitively reflect about their own emotions and beliefs (Andres & Arnold, 2009 as cited in Barcelos, 2015); • Using innovative technology in foreign language learning classes. But also enabling learners to experience favorable learning conditions (Basaran & Cabaroglu, 2014).
At the beginning of language learning beliefs research the traditional approach was active and the studies mostly implemented quantitative research methods with questionnaires as the main instrument. They focused on the cause-effect relationship between affective variables and learning procedures (Kalaja et al., 2018).The Beliefs about Language Learning Inventory (BALLI) is the most widely used questionnaire for studying language learning beliefs (Horwitz, 1988;Horwitz, 1999). With emergence of the contextual and sociocultural approaches the focus moved to qualitative research models where learner narratives became central (Kalaja et al., 2018). Learner narratives are important for the sociocultural approach since it focus on the fact that beliefs are constructed and mediated in social interactions in which the role of other people is central. (Kalaja et al., 2018) For the contextual approach they are important since beliefs are constructed through language and are reconstructed through our communication with others (Aragão, 2011).

Foreign Language Learning Anxiety
Anxiety has attracted a lot of attention form researchers in the field of education because of the debilitative effect that it can have on learners. If anxiety is not properly understood the learners could mistakenly be labeled as lazy, uninterested and uncooperative which would further hinder their learning (Tsiplakides & Keramida, 2009). In the educational context there exist various types of anxiety such as test anxiety, communication apprehension, fear of negative evaluation and foreign language classroom anxiety (Zheng, 2008). Although all of these anxieties have commonalities they are distinctive and separate. Foreign language classroom anxiety was first identified as a special type of anxiety by Horwitz at al. (1986). They defined it as "a distinct complex of self-perceptions, beliefs, feelings and behaviors related to classroom language learning arising from the uniqueness of the language learning process" (p. 128).
As with any other anxiety foreign language classroom anxiety can have both physiological and behavioral consequences. The physiological symptoms typically include: sweating, palpitations, trembling and increased blood pressure (Kralova, 2016). Behavioral symptoms might include: squirming in the seat, playing with hair, clothes, objects, stuttering and generally appearing nervous, avoidance behavior (learners pretending to be sick, hiding in the last rows, missing the classes to alleviate the anxiety etc.), image protection behavior (smiling and nodding frequently, seldom interruptions and frequent communication feedback), giving up, procrastination, freezing up or experiencing memory lapses in test situations (Young, 1991;Kralova, 2016). As Zheng (2008) noted, foreign language anxiety affects learners academically, socially and cognitively; it lowers their quality of communication output and influences their personality primarily influencing self-esteem and self-confidence as a learner. From the cognitive perspective anxious learners experience more task-irrelevant cognitive processing which hampers the capacity of their working memory.
Three important characteristics of foreign language classroom anxiety are: 1. High feelings of self-consciousness; 2. Fear of making mistakes; 3. Desire to be perfect while speaking (Foss & Reitzel, 1988). The foreign language classroom is a place where to expect heightened feelings of self-consciousness is natural since the learners are aware that their ability to express themselves is as limited as their foreign language skills; even very skilled foreign language speakers express difficulty in being witty and expressing themselves fully in the foreign language. (Horwitz at al., 1986).
The number, complexity and interconnectedness of factors which influence foreign language classroom anxiety have sparked a plethora of research on this topic (Tsiplakides & Keramida, 2009;Hashemi&Abbasi, 2013;Rouhani, 2013). Factors which influence foreign language anxiety can either be inhibiting (those which prevent anxiety from occurring) or activating (those which influence the occurrence of anxiety).Only the framework of factors presented by Kralova (2016) will be included here since the focus of the paper is on the factor of language learning beliefs. Foreign language classroom anxiety factors can firstly be divided into lingual (factors dealing with the language itself) and extralingual (factors related to the language learning process). Lingual factors can further be divided into intra-lingual (factors within the language itself) and inter-lingual (factors connected to the relationship between two languages -usually the mother and foreign languages) factors. The extralingual factors are divided into impersonal and personal factors. Impersonal factors incorporate non-personal aspects of foreign language learning mainly pertaining to the foreign language teacher's behavior. Personal factors are divided into interpersonal and intrapersonal factors.
Interpersonal factorsdeal with interaction with others during the foreign language learning process. Intra-personal factors can be divided into four main categories: 1. Personal beliefs -these include language learning beliefs,unrealistic concepts about language learning, self-rating, self-perception, self-presentation concerns, perception of language aptitude, perception of the difficulty of the task, selfperceived foreign language proficiency, perception of low ability in relation to peers (Foss &Reitzel, 1988;Horwitz, 1988;Young, 1991;Zheng, 2008;Von Worde, 2003;Tsiplakides & Keramida, 2009); 2. Attitudes and motivation (Gardner, 1985); 3. Personal characteristics -extroversion and introversion, verbal intelligence, emotional intelligence, perfectionism, tolerance to ambiguity, stylistic preferences (Zheng, 2008;Rouhani, 2013); 4. Psycho-social dimension -perceived social distance, societal interference, demographic differences (Kralova, 2016) 5. Experiential differences (Curry, 2011); In the fight against foreign language classroom anxiety teachers can implement teaching strategies which refer to standard classroom procedures, implement additional in class activities and teach learning strategies for the reduction of anxiety (Hauck &Hurd, 2005). When it comes to standard classroom procedures teachers should build a cooperative and safe learning environment (Cutrone, 2009); include music centered, group, problem solving, drama and role-play activities (Sağlamel & Kayaoğlu, 2013;Dolean & Dolean, 2014;Crookall & Oxford, 1991); encourage learners to continue speaking despite making mistakes, friendship and cooperation among learners, success and satisfaction, risk taking and effort (Sağlamel & Kayaoğlu, 2013;Hashemi & Abbasi, 2013), implement verbal and non-verbal immediacy behavior, portfolio assessment and fluency centered correction (Tsiplakides & Keramida, 2009). Additional in class activities aim to: help learners express their anxiety and problems in a non-threatening environment and share these experiences with peers; form coping strategies for specific language learning situations which are anxiety provoking for the learner; encourage learners to understand their feelings and actions and form a sincere communication with the teacher; help learners set reasonable goals (Crookall & Oxford, 1991). For detailed information regarding additional in class activities please refer to Foss & Ritzel, 1988;Crookall & Oxford, 1991;Young, 1991;Foroutan, & Noordin, 2012;Curry, 2014. Foreign language anxiety can be measured in three major ways: behavioral observation/rating, physiological assessment and self-reports (Zheng, 2008). The most widely used means of measuring foreign language classroom anxiety is the Foreign Language Classroom Anxiety Scale (FLCAS) developed by Hoewitz, Horwirz and Cope (1986).

Language learning beliefs and foreign language classroom anxiety -interaction
As it could have been seen from the previous segment of this paper, language learning beliefs present just one of many factors which may influence foreign language anxiety. However the importance of this factor may be explained by the strong interconnectedness between beliefs and emotions. "Emotions validate and provide evidence for beliefs and guide our attention towards information that is relevant to our goals. Emotions awake, intrude into, and shape beliefs, creating, amplifying and making them more resistant to change" (Frijda et al. 2000 as cited in Barcelos, 2015 p. 311). As Barcelos (2015) notes,feeling is believing because our emotions are believable. They are the root of an interesting phenomenon called the motivational bias which presents a challenge in the process of changing unrealistic language learning beliefs. Motivational bias is the process in which individuals harbor beliefs and look for, notice and produce only the information which supports their emotional goals or misconceptions. Such a process creates a vicious circle of emotions and beliefs where emotions generate and sustain certain beliefs while these beliefs further support emotions.
Beliefs are difficult to change as well as the negative self-perceptions and negative sentences that the learners say to themselves that are one of the causes of language learners' anxiety. When looked at from a psychological angle, mere specifically the rational-emotion behavior therapy (REBT) approach, we function through dialogue with ourselves, through the rational and irrational beliefs that we harbor about ourselves and the world around us. Each individual has his/her own belief system and it is proposed that anxious learners have irrational beliefs which provoke the feelings of anxiety (Ellis & Harper, 1975;Ellis, 2003). The affective contextual approach to studying language learning beliefs functions in much the same way since it proposes that beliefs and emotions are articulated and constructed in discourse. It distinguishes affect from feelings and defines feelings as "part of the domain of language by which people construct their worlds, coordinate actions with other people and are able to reach self-awareness." (Aragão, 2011 p. 303). In both cases the focus is on beliefs and emotions which are expressed through language, and on the fact that through the understanding of these beliefs we can better understand ourselves, our behavior and therefore have a better chance of changing our behavior in language learning classes but also on some wider planes.
Among others Bernat (2006 as cited in Sioson, 2011), Horwitz (1988 and Hou (2013) explored the relationship between language learning beliefs and foreign language classroom anxiety. Brent deduced that negative and unrealistic language learning beliefs lead to foreign language classroom anxiety. If we go through the three basic characteristics of foreign language classroom anxiety (1.High feelings of self-consciousness; 2.Fear of making mistakes; 3. Desire to be perfect while speaking (Foss &Reitzel, 1988)) they may be connected to certain language learning beliefs. High feelings of self-consciousness arise when the learners believe that others are better than they are in language learning. It could also be connected to the belief that there exists a foreign language learning aptitude that they had just not been born with and are therefore doomed to be humiliated in language learning classes. This is connected to the ideal modelconcept (Arago, 2011). The self-conscious learners fear to be embarrassed in front of teachers or other learners who they believe to be superior to them in knowledge and skill. Self-conscious learners perceive their communication as less effective than that of their peers, and expect continued failure no matter what feedback they actually receive, they feel incompetent to grasp the language and present themselves in a way that is consistent with their self-image (Foss & Ritzel, 1988). Because of the motivational bias such beliefs are deep rooted and very hard to change. The second and third characteristics are connected to the belief that nothing should be said before it can be said perfectly. Such a belief is contradictory to two facilitative language learning beliefs which should be nurtured in the foreign language classroom -Mistakes are an inevitable part of learning a foreign language andRisk taking is a positive strategy in language learning (Tsiplakides & Keramida, 2009).

Research
The aim of the study was to explore the relationship between language learning beliefs and foreign language classroom anxiety in learners studying at the Preschool Teacher Training College "Mihailo Palov" in Vršac. For the main aim to be achieved the learners foreign language classroom anxiety levels and language learning beliefs needed to be explored.
It is hypothesized that a complex relationship exists between language learning beliefs and foreign language classroom anxiety and that a stronger correlational relationship exists between certain language learning beliefs and the occurrence of foreign language anxiety.
The results of the study could help researchers better understand this complex relationship but also help language teachers focus on the beliefs which would yield the most efficient and fastest results in helping learners avoid and overcome foreign language classroom anxiety. They would also help us better understand this specific demographic group of learners regarding their individual differences concerning anxiety levels and language learning beliefs. The study incorporates both quantitative and qualitative analysis. For the quantitative analysis two instruments were chosen: BALLI (Beliefs about Language Learning Inventory (Horwitz, 1988)) and FLCAS (Foreign Language Classroom Anxiety Scale (Horwitz et al. 1986)). Both of these questionnaires implement the Linker scale and both have been internationally used and accepted as valid and reliable for the measurement of language learning beliefs and foreign language classroom anxiety, respectively (Basaran & Cabaroglu, 2014;Dolean & Dolean, 2014). Correlational analysis was implemented to see whether a significant correlation existed between certain language learning beliefs and high levels of foreign language classroom anxiety.
The instrument of the quantitative analysis werethe students'Language learning histories. They have been chosen since they allow learners to express their views on a wide variety of topics in depth and since they are frequently used to explore attitudes and behaviors (Tse, 2000). The results from the quantitative analysis were used as the starting point for the qualitative analysis.

Quantitative statistical analysis
The complete statistical analysis was conducted in SPSS for Windows, version 20.0. In accordance with the stated aim of this research, out of all available procedures of statistical analysis, the following were used: Descriptive statistical indicators: In order to render descriptive indicators of category variables, the frequency of their value was calculated (e.g. a student of the Preschool Teacher Training College/Teacher Training Faculty) and percentages.
Examining the differences between these two groups: the nonparametric technique was used in order to ascertain the differences between the category data -chi square test of independence (X2 test) with cut-off significance of p≤0,05. The justification for the utilization of nonparametric techniques was assessed with the normality of distribution test, that is, by utilizing the Kolmogorov-Smirnov test which has indicated that there is a significant distance of results from the normal distribution with all assessed variables (the degree of anxiety, individual beliefs) p= ,000.
Correlation: the correlation describes the strength and direction of the linear relationship between the two groups of variables, and the Spearman's rank correlation coefficient (Spearman's rho) will be used in this paper, which is appropriate for ordinal variables or variables that can be ranked (Likert scale). When analyzing the results of the Spearman's correlation, one must take into account the sign preceding the coefficient of correlation which describes its direction. The negative sign means that large values on one scale are followed by low values on the other, while the positive sign means that large values have been detected on both scales. One should also pay attention to the magnitude of the coefficient of correlation which can have values from -1.00 to 1.00 (Pallant, 2007: 137). The guidelines for determining the strength of the relationship are the following: low correlation= 0.10 to 0.29, medium correlation=0.30 to 0.49 and high correlation=0.5 to 1.

The Research Sample
In accordance with the determined aim, a total of 101 students of the Preschool Teacher Training College in Vrsac and Teacher Training Faculty of the University of Belgrade (Vrsac campus) participated in this research. The sample's composition is depicted in Chart 1.

Chart 1.The composition of the research's sample
Fifty students of the Teacher Training Faculty (Vrsac campus) participated in this research, which represents 49.5% of all participants and 51 students of the Preschool Teacher Training College, which represents 50.5% of all participants. By relying on the data presented in Chart 1 we can surmise that the sample of this research is balanced.
The degree of student anxiety has been reserached by means ofthe Foreign Lnaguage Clasroom Anxiety Scale.The division of learners according to the anxiety levels measured in the FLCAS had been done in the following way depending on the mean score from the aforementioned instrument: • M= 1-2.5 group of learners who did not exhibit anxiety • M= 2.5-4 group of learners who exhibited moderate anxiety • M= 4-5 group of learners who exhibited strong anxiety Psychometric properties have been assessed by utilizing Cronbach's alpha, after reversing negatively worded items. By analyzing the internal consistency of the scale, Cronbach's coefficient of 0.956 has been determined, which implies a high degree of reliability.
The frequency of the degree of student anxiety is presented in Table 1. Out of the total number of 101 participants, 30 did not manifest anxiety, 64 exhibited moderate anxiety, while 8 students exhibited strong anxiety. In the next table (Table 2) we will present the degree of anxiety relative to study group which students attended, so that we could assess the existence of potential discrepancies among them.

Chart 2. The degree of anxiety relative to the group of survey participants
From the table and chart above, we can discern the distribution of the degree of anxiety according to the participants' study group. Approximately an equal number of primary school teachers and preschool teachers was present in all three manifested degrees of anxiety (not present; moderate; strong). There was no anxiety among 11 prospective preschool teachers and 19 prospective primary school teachers; there was moderate anxiety among 36 prospective preschool teachers and 27 prospective primary school teachers; there was strong anxiety among only 4 preschool teachers and 4 primary school teachers. The calculated chi square demonstrates that there is no statistically significant difference between prospective primary school teachers and preschool teachers in the degree of anxiety: X2(2, N=101)=3,409, p=0,182

Correlation
In order to determine the connection between language learning beliefs and foreign language classroom anxiety in students, who are studying to be preschool teachers and primary school teachers, the Spearman's correlation was used. It had been selected since the gathered results do not fulfil the requirements for Pearson's corellation (there does not exist a normal distribution of data which had previously been calculated using the Kolmogorov-Smirnov Test ) and since we are dealing with ordinal data. The tests did not show any significant difference between the two groups of students, and therefore they will not be viewed seperately. From the 34 tested language learning beliefs, a significant correlation occured only in the following (view Table 3.) On the basis of the data presented in Table 3., it can be concluded that there occur both positive and negative correlations. For a detailed analysis each belief will be separately analyzed: 1. A strong negative correlation exists between the belief a) and the level of foreign language classroom anxiety (r=-,407).The higher the anxiety level the stronger the belief that English is a hard/very hard language. 2. A strong negative correlation exists between the belief b) and foreign language classroom anxiety (r=-,515). Learners with high levels of anxiety believe that they will not ultimately learn to speak English well. 3. A positive correlation exists between the belief c) and foreign language classroom anxiety (r=,317). Learners with high levels of anxiety believe that they should not say anything in English until they can say it correctly. 4. A small negative correlation exists between the belief d) and foreign language classroom anxiety (r=-,244). Learners with high levels of anxiety would not join a conversation if they heard people speaking in English.

A negative correlation exists between the belief e) and foreign language classroom
anxiety (r=-,327). Learners with high anxiety levels believe that they do not have foreign language aptitude. 6. A strong positive correlation exists between the belief f) and foreign language classroom anxiety (r=,466). The higher the level of foreign language anxiety the more self-conscious the learners feel speaking in front of others. 7. A weak positive correlation exists between the belief g) and foreign language classroom anxiety (r=,230). Learners with high anxiety levels believe that learning a foreign language is mostly a matter of learning a lot of grammar. 8. A weak positive correlation exists between the belief h) and foreign language classroom anxiety (r=,264). The higher the level of foreign language classroom anxiety the stronger the belief that learning a foreign language is a matter of translating from one language to another. 9. A weak positive correlation exists between the belief i) and foreign language classroom anxiety (r=,217). Learners with high anxiety levels believe that people who are good at math and science are not good at learning foreign languages. 10. A negative correlation exists between the belief j) and foreign language classroom anxiety(r=-,357). Learners with high levels of anxiety believe that not everyone can learn a foreign language.

Qualitative analysis
Language learning histories have been written by 25 language learners who participated in the study. The learners could choose if they wanted to write them in English or in Serbian. This choice was offered so that the learners would feel free to express themselves in the language in which they felt comfortable so as to give their opinions unhindered. For the learners who wrote in English the examples given in the further text will be citations with any and all grammar and vocabulary mistakes, while the researcher will provide translations of the examples from the language learning histories written in Serbian. In order to protect the learners' identity only the initials will be given with the extracts from language learning histories.
The data from the quantitative analysis was used as the starting point for the qualitative analysis. Namely, the ten beliefs which showed a high correlation with foreign language classroom anxiety were divided into five broader categories: 1. Beliefs regarding the difficulty of language learning -The English language is: 1. very difficult, 2. difficult, 3. a language of medium difficulty, 4. easy, 5. very easy language. 2. Beliefs regarding foreign language learning aptitude -I have foreign language aptitude; People who are good at math and science are not good at learning foreign languages; Everyone can learn to speak a foreign language. 3. Beliefs regarding learners expectations -I believe that I will eventually speak English very well. 4. Beliefs regarding language learning strategies -Learning a foreign language is mostly a matter of learning a lot of grammar; Learning English is mostly a matter of translating from one language to another; If I heard someone speaking English, I would join the conversation to practice. 5. Beliefs regarding feelings of inadequacy -You should not say anything in the foreign language until you can say it correctly; I feel self-conscious speaking the foreign language in front of other people.
These categories are similar to the ones Horwitz (1988) used in her analysis but not identical. Two more categories were added to the qualitative analysis since they were prominent in the language learning histories: • Incongruence between the teachers' and learners' language learning beliefs • Beliefs about teachers' behaviour Beliefs regarding the difficulty of the English language were mentioned in eleven language learning histories. It is interesting to note that most of the learners believed that English can be hard or easy to learn depending on the type of instructions. As is illustrated in the following extracts: K.B. -I wasn't lucky enough to have my first experience in English presented in an interesting and easy way; T.O. -I'm always hated English because was hard for me (during her first encounter)… English is easy when you have practice and if you have wish and love. I.T. -…playing interesting games to learn English easily. Two of the learners expressed that English was hard for them because their previous experiences did not offer them the right conditions for language learning. Only two from the eleven who discussed the difficulty of the English language considered it inherently easy or difficult -e.g. M.K.

-I could never understand people who don't speak English at all because you hear English every day and you can't not remember it.
It should be noted that the anxiety levels of these two learners were in extremes, the learner who believed that english was inherently difficult had very high anxiety schors while the one who thought that it was inherently easy had very low anxiety scores. Two learners believed that some parts of learning the language may be more difficult in the beginning but get easier over time with practice -A.S. Later on we started learning grammar which was difficult to start, but after a while we understood the principles.
Beliefs regarding foreign language aptitude were mentioned in seven language learning histories. Four of the learners believed that a predisposition for learning a language is not needed. They believed that the most important thing for learning languages is to love the language, to have the will to do it and for the language learning to be interesting -M.R. As for the predispositions -they are not necessary. You just have to love the language! K.B.I realized that English language learning was interesting and that no predisposition is necessary for learning it, only a wish. Four of the learners believed that they could not learn English because of a lack of ability (M.S. Learning a foreign language doesn't go well for everyone, I belong in this group), two of which changed this belief during the course of their study -M.I. Then I realized that I can learn English better and have new friends.
The first two categories could be connected to a belief that the learners expressed that if English language learning is interesting it is easy and possible. i.e. The best way to learn English is through fun activities be they in class or outside of it.
When it comes to beliefs regarding learners' expectations of success, seven learners expressed them in their language learning histories. Six learners expressed their hopes for success in the future, four used the expression I hope and two used an expression which contained a bit more confidence want. Three of the learners placed their hopes of mastering the language in the teacher's hands -J.T. I hope that you will help me to learn English well and help me to be proud on myself. Only one learner mentioned that she believed that she would never be able to learn the English language and that such a belief caused her to further dislike this subject (T.O. I've always hated English because was hard for me, and my opinion was that I would never know English).
A total of fourteen learners mentioned beliefs about language learning strategies in their language learning histories. Five of the learners believed in the usefulness of in class strategies such as active participation in class, doing homework regularly, asking questions when something is unclear, writing essays, doing dictations etc. The other ten learners believed in the usefulness of learning strategies outside of the classroom such as watching movies, TV series and cartoons in the foreign language (usually without subtitles), listening to foreign music, translating lyrics of favorite songs, playing video games in the foreign language, speaking English with siblings, meeting friends from abroad on the internet and while traveling.
Learners also complained about the usage of insufficient teaching strategies and more generally the differences in the teachers' and learners language learning beliefs. Six learners addressed this issue in their language learning histories. The learners were mostly unsatisfied with the fact that teachers focused too much on grammar, reading aloud and translating texts and dictations, they did not promote speaking enough and did not pay attention to all of the learners equally. The learners considered such approaches boring and inefficient -R.J. During the whole secondary school we didn't work productively as we could. We focus only on the simple tenses and such trivial things so we didn't get so far. Seven learners discussed feelings of inadequacy in their language learning histories. Five of the learners believed that making mistakes in language learning classes should be avoided. They felt ashamed and irritated after making mistakes and frightened at the possibility of making them -D.V. -I wish to speak better but always when I talking with somebody, I'm blocking it, because I am afraid. I not to say good some words, and do not know some words. Two of the learners expressed concerns that they would not be able to meet the demands of the tasks in class and that they would not understand the teacher and classmates -K.B. It was a problem for me to actively participate in class and discussions because I did not understand what everyone was talking about and therefore felt discarded and unsure in myself. I felt ashamed because most of my classmates knew English and I did not. I was in constant fear that the teacher would ask me something and that I would not understand her and be able to answer.
In three of the learners such a negative approach to mistakes was formed as a consequence of a negative experience in which they were mocked in class for making a mistake. One learner attributed the fear of making mistakes to her character which is perfectionistic in nature.

Discussion
From the quantitative analysis we have gathered information regarding which language learning beliefs have the strongest relationship with foreign language classroom anxiety. In the qualitative analysis these beliefs were explored in more detail and two more categories of language learning beliefs which had a strong relation with anxiety were added. This information is significant so as to direct language teaching practice to focus on the promotion of beliefs which are the most important for the formation of a healthy relationship with language learning without high levels of anxiety.
The quantitative analysis showed that learners who believed that the English language is hard showed higher levels of anxiety. On the other hand, the qualitative analysis revealed that when learners wrote about this difficulty level, a small group considered the language as being inherently difficult or easy, but rather the majority considered that the difficulty level depended on the type of instruction. In other words, it was indicated that the learners believed that the language could be learned in either a difficult or an easy way. Such results could be connected to the importance of the congruence between the language learners' and language teachers' beliefs regarding foreign language teaching and learning.
The belief in the existence of foreign language learning aptitude was positively correlated with high levels of anxiety in the quantitative analysis. In the qualitative part of the research it was shown that this belief is open to change when a different language learning belief regarding the fun factor of language learning also changed, namely when they started viewing English language learning as fun they started believing that it could be learned even without some special language learning aptitude. From these results it can be concluded that teachers should promote the language learning belief that everyone can master the foreign language with appropriate instruction and effort. They should also focus on making their classes fun for the learners.
From the results on the learners' beliefs regarding their expectations of success it can be concluded that learners with lower expectations exhibit higher levels of anxiety and that the learners rely on the teachers support in this segment. Teachers should try to encourage learners as often as possible and try to instill in all learners the belief that language learning success is possible for everyone with hard work, a realistic plan and realistic expectations. This could be connected to the anxiety lowering strategy (Negotiation of objectives) proposed by Curry (2014) which relies on Critical Behavior Therapy (CBT) and the focus of which is realistic goal setting. Furthermore, active learning and goal setting were proposed as strategies for promoting positive beliefs by Basaran & Cabaroglu (2014).
When it comes to beliefs regarding learning strategies the quantitative analysis showed that learners who believed that learning a foreign language was mostly about learning a lot of grammar and translating exhibited higher levels of anxiety. In the qualitative part of the analysis a great majority of learners expressed strong positive beliefs towards language learning strategies which they considered fun (mostly outside of the classroom but also within the classroom) such as watching foreign movies, listening to foreign music, learning through in class games or video games, discussions etc. Learners expressed a negative view of the teaching approaches which put too much focus on grammar and reading and translating texts aloud in class. It is the researcher's belief that the learners who participated in the study understood the aforementioned learning strategies as common practices implemented in class and that the positive correlation between these beliefs and anxiety exists because these learners had more encounters with teachers who had such a teaching style.
Incongruence between the learners' and the teachers' language learning beliefs regarding the best ways and strategies in which the foreign language should be taught were the source of dissatisfaction and anxiety in many learners who participated in this study. Such a state can be avoided by exploring the students' learning beliefs at the beginning of the language teaching experience and explaining why certain strategies which are not in line with their beliefs are implemented and explain the usefulness of these strategies (Barcelos, 2015). Teachers who do this can also focus more on the strategies which they know that their language learners prefer in order to get the optimal results.
Language learners' beliefs regarding the teachers' behavior mainly focused on the positive characteristics which the teachers need in order to form a warm and encouraging environment. We can conclude here that the importance of rapport in language learning classes cannot be stressed enough when it comes to providing a safe environment which lowers the learners' anxiety (Tsiplakides & Keramida, 2009).
It was concluded in the quantitative analysis that learners who strongly believed that one shouldn't say anything until they can say it correctly exhibited high levels of anxiety. In the qualitative part of the study the learners expressed that one should avoid mistakes at all costs since making mistakes is scary because they will seem dumb or incompetent to others. Since some of the learners stated that negative experiences of mocking after a mistake had been made as the root of their fear of making mistakes we here again stress the importance of an accepting and warm classroom environment and the promotion of the belief that mistake making is an inevitable, normal and useful occurrence in foreign language learning. There exists an extensive literature on mistake correction practices which positively and negatively influence the levels of learners' anxiety as well as a few anxiety lowering strategies which focus on promoting a positive attitude towards mistakes which should be consulted for details on practical implementation (Crookall & Oxford, 1991;Hashemi & Abbasi, 2013;Kralova, 2016 etc.).

Conclusion
The paper explored foreign language classroom anxiety, language learning beliefs and the relationship between these elements in students at the Preschool Teacher Training College in Vrsac and the Teacher Training Faculty in Belgrade who study at the Vrsac campus. In order to get the fullest picture regarding these two factors and their complex interrelatedness both a quantitative and a qualitative analysis were implemented. The results were complementary and the qualitative analysis gave detailed information regarding what the learners considered to be the most important. Seven learning belief categories had been singled out as having the strongest relation to the occurrence of foreign language classroom anxiety: beliefs regarding the level of language difficulty, language aptitude, learners' expectations, language learning strategies, teachers' behavior, incongruence between the teachers' and learners' beliefs and the learners' feelings of inadequacy. On the basis of these results recommendations were given regarding teaching practices and exercises which could help learners get a learning experience in which anxiety is not present.

Bibliography
Jovana Brkic is an English language teacher. She runs a company called YLEARN which deals with educational practices and different forms of translations. She obtained her BA and MA degrees at the Faculty of Philology, University of Belgrade, and is now in the process of completing her PhD thesis the title of which is Ethical Values in Ursula Le Guin's Novels. She researches in the fields of literature, English language teaching methodology, e-Learning and translation.