The relationship between anxiety and attitude of students learning Turkish as a foreign language and their achievement on target language

The purpose of this study is to assess the anxiety connected with target language of the high school students learning Turkish as a foreign language. In this study, descriptive relational screening model was used. Two scales were used for collecting data. First scale was FLCAS Foreign Language Classroom Anxiety Scale; it was developed by Horwitz et al. and its validity and reliability was done by Horwitz. Alpha reliability coefficient of this scale was .97. The second scale was FLAS Foreign Language Attitude Scale; it was developed by Briem and its validity and reliability was done by Corbin and Chiachiere. The scale was translated into Turkish and alpha reliability coefficient of this scale was .88. The participants are 314 students (from 27 countries) learning Turkish in GİH High School in Kayseri-Turkey. 314 students’ responses were studied. The data were analyzed using means, standard deviation, Pearson correlation analysis, regression, and Mann Whitney U test. The findings show that there are in a tight relationship the success of Turkish course with the students’ foreign language anxiety and attitude. Students who fail have a negative higher evaluation score than successful students. In addition, students who fail have more total anxiety score than successful students (p<0.05).


INTRODUCTION
Teaching of Turkish as a foreign language has gained significant momentum in recent years.Prepared by the Turkish Council of Higher Education, Teaching Turkish graduate program is given course of training Turkish as a Foreign Language.In addition, many universities have started to give training to graduate programs for the teaching of Turkish as a foreign language (Göçer, 2009(Göçer, , p. 1301)).
At Gazi, Ankara etc. universities established The Turkish Learning Centres and have issued a book set to be used in training (TÖMER a , 2002;TÖMER b , 2006).Turkish Cooperation and Development Agency (TIKA) has prepared a set of language learning syllabus to be used in teaching foreigners Turkish.
Authors agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License or indirect factors affecting foreign language learning/ teaching are anxiety and attitudes.Many studies have been done on the effects of anxiety and attitudes on language learning.They are: "Students' and Teachers' Beliefs about Language Learning" (Kern, 1995); Anxiety and Foreign Language Learning: Towards A Theoretical Explanation (MacIntyre and Gardner, 1989); 'Influencing High School Students learning English Foreign Language Anxiety' (Doğan, 2008); 'Anxiety in the Process of Foreign Language Learning' (Ergür, 2004); 'Relationship Between Students' Foreign Language Anxiety Level and Achievement' (Gülsün, 1997); 'A Study on the Role of Anxiety in Second Language Learning' (Dalkılıç, 2001).In addition, many researchers have done research on the impact of anxiety in learning a foreign language.According to Na, anxiety has been considered among the most important emotional factors in second language acquisition (2007: 22).Sertçetin (2006), Na (2007), Xu and Li (2010) etc. have studied the relationship between foreign language anxiety and other variables.
Motivation and attitudes were first studied made by Gardner and Lambert (1959) in second language learning (cited from Gardner and Lambert, 1959;Saif and Lori, 1999, p. 23).This work was later studied by Smythe (1981), Pierson et al. (1980) and Salasiah (1999).According to these studies, there is a significant relationship between students' achievements and attitudes towards the target language.
According to Öncül (2000, p. 1082-1083), attitude is the consistent behavior of persons, objects that invariably leads to belief.In Gardner's attitude, an emotional property (1978, p. 1) explains the motivation and attitude in learning the language acts as a complement.According to Brown (1994, p. 168), attitude is the human mental and physical development of every aspect in childhood.Attitudes of people vary.The relationship within human experiences occurs as a result of emotional factors.Starks and Paltridge (1996, p. 218), in their work titled learning a language, say that language is closely related to attitudes.
To be successful in foreign language learning, the students' faith, motivation levels, emotional state, attitudes and expectations of individual properties in closely related variables count.Affective states have an important place in foreign language anxiety; similar to other individual differences in learning language, it appears to be an important variable affecting success (Aydın and Zengin, 2008, p. 82).
There are three types of anxiety in language learning: anxiety due to persona; worry about having a creature (cited from Ellis, 1994, Sasson, 2010).The second type of anxiety is caused by exams or oral presentations.The third type of anxiety occurs in reading or speech understanding, where the necessary competence is not available.This occurs from a learning situation (Sasson, 2010).Depending on the situation of anxiety, it is an integral part of foreign language learning process.Students are forced to learn the use of language if they could not express themselves well enough, think well, and lack the real intelligence.This kind of anxiety obstructs the right to use the target language by students (Ergür, 2004, p. 49-50).Ellis, concern about the following factors, indicates that (cited from Ellis, 1994;Sasson, 2010): 1. Learners have a competitive nature.2. Teachers' questions are threatening.3.There is lack of a better second-language learning environment.

Problem statement
1. Is there a relationship between academic achievement and, anxiety and attitudes towards TFL learners'?2. Is there a difference in terms of attitude and anxiety scores of successful and unsuccessful students'?

METHOD
In this study, descriptive relational screening model was used.

The working group
The working group of this study composed of high school students who are foreign nationals.The research was conducted in the 2007-2008 academic years.The participant are 314 students (came from 27 countries) learning Turkish in State High School (GİHL) in Kayseri-Turkey.

Data collection and analysis
Two scales were used for collecting data.The first (FLCAS -Foreign Language Classroom Anxiety Scale) was developed by Horwitz et al. (1986).Validity and reliability of the scale was made by Horwitz (1986).Turkish version of the scale was adapted by Gülsün (1997), Dalkılıç (2001) and Batumlu and Erden (2007); it consists of two parts.The first part contains personal information such as Turkish as a foreign language learning goals of participants, grade level, gender, and nationality.The other part is the Turkish version of the scale.The scale consists of 33 items.Eight of them (1, 9, 14, 18, 24, 27, 29, 32) communicate concerns, nine (3, 7, 13, 15, 20, 23, 25, 31, 33) fear of negative evaluation, five (2,8,10,19,21) for the test anxiety.The remaining eleven items were collected in a group under the name of Turkish-class anxiety.The 5 Likert scale ranged from "1 -Strongly Agree" to "5 -Strongly Disagree".The Turkish version of the scale was made by Gülsün (1997).Using the Turkish version of the scale reliability coefficient was calculated.The scale was applied twice in 30 days, with 32 students.For the data, Cronbach's alpha () reliability formula for calculating the reliability coefficient () was .97.
The second scale (FLAS -Foreign Language Attitude Scale) was developed by Briem (1974).The scale consists of 32 items.Validity and reliability study of the scale was made by Corbin and Chiachiere (1995).To respond to the questions of the scale, 'Certainly Agree', 'Agree', 'Undecided', 'Disagree' and 'Strongly Disagree' options were presented in the form of five-degree.The scale was translated from English to Turkish by two English teachers and then Turkish to English translation was built and made suitable.The scale was applied twice in 30 days, with 33 students.For the data, Cronbach's alpha () reliability formula is used to calculate the reliability coefficient (), found as .88.
In this study, the relationship between achievements of students and anxiety and attitudes was investigated.The scores of anxiety and attitudes were examined in terms of the following variables: differences in gender, nationality, grade level and purpose of learning a foreign language.9 th grade level of students in the study group was '1'; 10 th , '2'; and 11 th , '3'.In terms of nationality, the Turkish republics students were '1', and '2' for arrivals from other countries.Coding, analysis and investigations were made accordingly.
The first semester students studying Turkish lessons receive intensive semester grade point averages of scores; achievement status is taken as the criterion variable.Students' success and status of transcripts showing the progression of book data were obtained with permission from National Education Directorate of the School Administration.Note scores are based on the average, calculated in accordance with relevant regulations, the principles of assessment and evaluation.0-44 was accepted as 'not successful'; 45-100 was accepted as 'successful'.Analyses were made in accordance with these ranges.By using SPSS 11.5 package program on the data obtained, means, standard deviation, Pearson correlation analyzing, regression, and Mann Whitney U test were analyzed.

The findings about personal information of the students learning Turkish as a foreign language
As shown in Table 1, total number of the students learning Turkish as a foreign language is 314 (13 students from Turkish Republics; 301 students from other countries).81 of the students are in 9 th grade ; 107 in 10 th grade and 126 in 11 th grade.

The findings obtained with two scales and the success situation of the students
The relationship with total anxiety scores and dimensions of the anxiety scale of students' success was determined using Pearson's Correlation Coefficient (Table 2).According to results, the relationship between the success and level of total anxiety (sub-dimension of anxiety: anxiety of communication, anxiety of negative evaluation, anxiety of Turkish classroom) was close to zero.There is negative and weak correlation between success and test anxiety.As can be seen in the table, Pearson's Correlation Coefficient is close to zero.
Is predictor of success attitude and sub-dimension of anxiety significant?Answer to this question was investigated with multiple regression tests (Table 3).Looking at the variable R 2 it is predicted only by 0,026 percent with the success of the variable sub-components and attitudes.
The relationship of the success of the students with sub-dimensions of anxiety and attitude is evaluated using regression models.The test result shows this model is not significant (p> 0.05) (Table 4).
Whether or not different in terms of anxiety, subdimension of anxiety and attitude of successful and unsuccessful students was investigated with the Mann Whitney U test.Due to the big difference between the number of successful and unsuccessful students, a nonparametric test was preferred.The following are Mann Whitney U test results related to descriptive value of variables (Table 5a and b; Table 6).
According to results of the comparison, there is in term of negative evaluation a significant difference (p> 0.05) between the two groups.Successful students score higher than unsuccessful students who have a negative evaluation.In addition, unsuccessful students have a higher total anxiety score than successful students (p<0.05).

RESULTS AND DISCUSSION
The relationship between students' success and both total anxiety scores and dimensions of the anxiety scale was determined using Pearson's Correlation Coefficient.
According to the results of the relationship between success and level of total anxiety (sub-dimension of anxiety: anxiety of communication, anxiety of negative evaluation, anxiety of Turkish classroom), correlation was found close to zero.According to MacIntyre and Gardner, Students' language learning anxiety increases as a result of the adverse situation in the process (1989, p. 257).
According to Öner and Gedikoğlu, anxiety can lead to difficulties in learning the language as well as pronunciation; students also lack information about the structure of language etc. (2007, p. 69).Very high numbers of students have success and students' anxiety levels in Turkish are not high.Turkish anxiety level is low.So level of students' achievement is high.Very few of the 314 students failed.Most students have a successful outcome, which is a natural indicator.According to Aydın and Zengin, students with high anxiety levels have low achievement levels (2008, p. 87).In other words, students with low anxiety level were more successful in learning a foreign language (Öner and Gedikoğlu, 2007, p. 77).
Success with test anxiety, which is one of the anxiety sub-dimensions, was observed as negative correlation.Test anxiety, students' knowledge and skills were seen as a negative factor.
There are close to zero correlation between success and attitude.The attitudes of the students to the target language affect the success in part.
Anxiety sub-components and attitude, 'success' variable were predicted by only 0,026 percent.
Anxiety, sub-dimension of anxiety and attitude of successful and unsuccessful students were investigated with the Mann Whitney U test.Accordingly, there is a significant difference between the two groups in terms of negative evaluation (p <0.05).Unsuccessful students have a negative evaluation score higher than successful students.In addition, unsuccessful students have more total anxiety score than successful students (p <0.05).

RECOMMENDATIONS
Anxiety is an effective factor in students' learning and effective use of their target language.Teachers who teach Turkish as a foreign language must be comfortable with the students in classroom and extracurricular environments.To do this, teachers who teach Turkish as a foreign language, first of all, should correct the concerns of the target audience.This is because the achievements of individuals with low levels of anxiety are higher than those of over-anxious students.
Test anxiety directly affects the success of students.For that, the examinations of students' knowledge and skills in full shifts should be line with psychological and philosophical perspective.
Success of individuals learning a foreign language, with a request for a particular purpose, depends on an adequate level of learning.In addition, a positive attitude towards the target language has greater importance.According to Göçer, Teachers who teach Turkish as a foreign language should have a positive attitude towards Turkish.This is because, the success of individuals with a positive attitude is higher than those without it (2009, p. 1309).Students learning language with a positive attitude use the target language effectively in everyday life and have higher achievement levels.
As we have seen, the success or failure of students is effective in anxiety states.For this, teachers should communicate effectively with their students, and should provide a comfortable environment in which they behave.And most importantly, the feelings, thoughts, experiences and proposals should create a media to express verbally and in writing.Failed students have a high negative rating point.In addition, unsuccessful students have more total anxiety score compared to the successful students (p <0.05).
Turkish as a foreign language teaching task must be carried out by teachers who have adequate pedagogical formation.In addition, students learning Turkish as a new language should use well-prepared custom compiled texts, the story books that fit the purpose, etc. Students should be engaged in cultural tours in different social environments.This will increase their opportunity to use and learn the target language.

Table 2 .
Pearson Correlation Coefficients between Students' Achievement and, Anxiety and Attitudes against Turkish as a Foreign Language

Table 3 .
Related To the Multiple Regression Test Whether Significant Predictors of Success of Attitude and Sub-Dimension of Anxiety

Table 4 .
The Relationship the Success with Sub-Dimensions of the Anxiety and Attitude

Table 5a .
The Situation of Successful and Unsuccessful Students Related To Anxiety, Sub-Dimension of Anxiety and Attitude Scores

Table 5b .
The Situation of Successful and Unsuccessful Students Related To Anxiety, Sub-Dimension of Anxiety and Attitude Scores

Table 6 .
Descriptive Values of Variables and the Mann-Whitney U Test Results Turkish Cooperation and Development Agency; TÖMER: Turkish Education, Research and Application Center; YÖS: The Examination for Foreign Students; YÖK: Higher Education Council; TFL: Turkish as a Foreign Language; FLCAS: Foreign Language Classroom Anxiety Scale; FLAS: Foreign Language Attitude Scale.