The Correlation among EFL Learners’ Test Anxiety, Foreign Language Anxiety and Language Achievement

The main aim of this study was to investigate the correlation among test anxiety (TA), foreign language anxiety (FLA) and language achievement of university preparatory students learning English as a foreign language. The sample of the research consisted of 301 (211 females, 90 males) attending a one-year EFL preparatory school at Ondokuz Mayıs University, School of Foreign Languages. The TAI (Test Anxiety Inventory) developed by Spielberger (1980), FLCAS (Foreign Language Classroom Anxiety Scale) developed by Horwitz (1986) and the students’ language achievement scores were used to collect data. In addition, this study aimed to determine the effect of gender on both TA and FLA. The findings of the study revealed that there was no relation between students’ TA and their language achievement. It was also found a significant positive correlation between students’ TA and FLA. Besides, the result demonstrated that there was a significant reverse correlation between students’ FLA and their language achievement. Following t-test results, it was found out that females exhibited higher mean scores than males in both TAI and FLCAS. Finally, certain suggestions were set for EFL teachers to reduce text anxiety and language learning anxiety in foreign language classroom settings.


Introduce the Problem
Language learning is undoubtedly viewed as a quite complex and multifaceted process in nature.Over the past few decades, there is a widespread agreement that language researchers aim to explore the reasons rendering language learning process more difficult.The related literature on affective factors in EFL learning reveal that one of the most widely studied emotional concepts in psychological and educational research area is the notion of anxiety.Anxiety has become a major research interest of a wealth of studies in literature.In essence, it is almost impossible to define anxiety in a simple sentence.Anxiety is associated with feelings of uneasiness, frustration, self-doubt, apprehension, or worry (Brown, 1994).In a similar vein, Spielberger (1966), defines anxiety as subjective, consciously perceived feelings of apprehension and tension, accompanied by and associated with activation or arousal of the autonomic nervous system.Anxiety can also be referred to as "relatively stable individual differences in anxiety-proness as a personal trait" (Spielberger, 1972, p. 482).From the holistic views of language learning, learner's emotional characteristics cannot be separated from his cognitive features.Therefore, the anxiety construct has been an outstanding focus of attention in foreign language learning research.Dörnyei (2009) described anxiety as the five best-known individual factors -language aptitude, motivation, learning styles, learning strategies, and anxiety as a curious variable.
According to Horwitz, Horwitz and Cope (1986), foreign language anxiety is the separate complicated phenomenon of self-perception, beliefs, feelings and behavior related to classroom language learning process.Foreign language anxiety is classifed as a situational specific anxiety (MacIntyre & Gardner, 1991;Horwitz, 2001) rather than a personality trait.Situation specific anxiety is an individual tendency to be anxious in a particular time and situation.According to Horwitz et al., (1986), the situation specific characteristic of foreign language arises from the uniqueness of foreign language learning in classroom settings.Horwitz et al., (1986) defines three components of Foreign Language Anxiety (FLA) as follows: the fear of negative evaluation, communication apprehension and test anxiety.Fear of negative evaluation consists of the feelings of apprehension related to other people's evaluations, refraining from the situations that need evaluation and false perception that they will always be evaluated in a negative way.Foreign language learners often experience this fear as they are evaluated both by their teachers and by their peers.Communication apprehension is a type of shyness characterized by fear or anxiety about communicating with people.It derives from the personal knowledge that one will almost certainly have difficuly understanding others.Communication apprehension occurs when learners have immature communication skills although they have mature ideas and thoughts.Test anxiety is an apprehension over academic evaluation.It is a fear of failing in test situation and an unpleasant experience held consciously or unconsciously by learners.It is a type of anxiety concerning apprehension over academic evaluation arising from a fear of failure.
A wealth of research indicates that Test Anxiety (TA) is regarded as one of the most remarkable constructs in various fields.Zeidner (1998) asserts that test anxiety refers to phenomenological, physiological, and behavioral responses that accompany concern possible negative consequences and failure on an exam or similar evaluative situation.Hancock (2001) states that TA is a disturbing feeling that has physiological and behavioural sides, and that is experienced in testing or evaluation situations.Sarason (1972, p. 383) identifies test anxiety as "proneness to emit self-centered, interfering responses when confronted with evaluative conditions".In addition, TA can be conceptualized as "a perceived intellectual incompetency and a defensive ploy to ward off low self-evaluation" (Zeidner, 1998, p. 81).Liebert & Morris (1967) proposed that TA consists of two major components : "worry" and "emotionality".Worry is the cognitive component of TA referring to concerns about being evaluated and about the consequences of failure.In contrast, an affective component labelling "emotionality," refers to the perception of autonomic reactions evoked by the test situation.Worry is more correlated with academic performance than emotionality (Liebert & Morris, 1967).Spielberger (1972) viewed test anxiety as a situation-specific form of trait anxiety.Trait anxiety is a stable personality characteristic, whereas state anxiety is a transitory emotional state determined by the interaction of one's trait with the present situation.According to Spielberger (1983), state anxiety occurs when individual perceives a situation as dangerous and threating.A person with state anxiety shows anxiety only under particular circumstances, such as before an exam.Test-anxious students have a tendecy to experience high levels of trait anxiety during or before testing settings, and they perceive evaluative situations as threatening.Worry and emotionality are symptoms of state anxiety, and these symptoms interfere with students' test performance (Spielberger & Vagg, 1995;Zeidner, 1998).

Importance of the Problem
In recent years, there has been much interest in test anxiety and foreign language anxiety research in literature.This has resulted in a growing body demand for research in foreign language learning process.The research on this issue in Turkey seems too limited.In order to fill this gap, the current study aims to investigate the correlations between test anxiety and foreign language anxiety, and language achievement of preparatory students learning English as a foreign language.Another purpose of the study is to determine whether there is a significant effect of learners' gender on test anxiety and foreign langauge anxiety.

Research Questions
The present study was conducted to investigate the following research questions: 1) Is there a significant relationship between students' test anxiety and language achievement?2) Is there a significant relationship between students' gender and test anxiety?
3) Is there a significant difference between students' foreign language classroom anxiety and language achievement?4) Is there a significant relationship between students' gender and foreign language anxiety? 5) Is there a correlation between students' test anxiety and their foreign language anxiety?

Method
The study adopted a descriptive kind of general survey research design with a quantitative method.As it is known, the descriptive approach aims to determine the condition of interest.Survey models stand for research approaches that aim to define past or existing situations.The research was conducted at School of Foreign Languages, Ondokuz Mayıs University in the spring term of 2015-2016 academic year.The subjects of the study were chosen randomly and they all participated in the research voluntarily.

Participants
The participants of the study were 301 (211 females and 90 males) students attending a one-year EFL preparatory school at Ondokuz Mayıs University, School of Foreign Languages, Samsun, Turkey.

Data Collection Instruments
TAI (Test Anxiety Inventory) developed by Spielberger (1980) was used to collect the data.The TAI is a self-report scale that was designed to measure individuals' test anxiety as a situation-specific trait.The TAI consists of 20 items which are to be responded by using a four-point rating-scale format ranging from 1 (almost never) to 4 (almost always).Minimum and maximum points students got from TAI are 20 and 100.In addition to a total score, separate scores for worry (8 items) and emotionality (12 items) can be obtained.In the present study, a total Test Anxiety score that TAI yielded was used.Higher scores on the TAI indicate higher levels of test anxiety.Spielberger (1980) reported studies on the validity and reliability of the instrument.Reliability of the TAI Total score for college students over a three week period was found to be .80,which indicated strong reliability.In a study of the internal-consistency reliability of the TAI Total score, the alpha coefficients were found to be .92.Turkish version of TAI adapted by Öner (1990) was used (Cronbach Alpha= .87) in this study to specify students' test anxiety.The Cronbach Alpha was reported to be .91for the present study.Another instrument used to gather data for the research is Foreign Language Classroom Anxiety Scale (FLCAS) developed by Horwitz et al., (1986).FLCAS is a self report with 33 items which measures the situational anxiety directly associated to the specific context of the foreign language classroom.FLCAS was a 5 point Likert type scale from strongly agree to strongly disagree.Cronbach Alpha of the FLCAS demonstrated high internal consistency (.93) and test reliability of the scale over 8 weeks .83,p=.001 (Horwitz, 1986).In order to prevent misunderstanding, translated version of FLCAS was used in the present study.Aydın (1999) translated FLCAS into Turkish and item 27 "I feel more tense and nervous in my language class than in my other classes" was eliminated from the questionnaire as all the classes the subjects of this research participated in the preparatory program were EFL classes.This item would not be suitable for their situation.Possible scores of the translated version of FLCAS in this study ranged from 32 to 160.The internal consistency of this final version was .91.Cronbach's Alpha for the FLCAS used in this study was .94 which manifested a satisfactory reliability.

Data Analysis
The research methdodology applied in this study is quantitative design.The analysis of data was carried out using SPSS (Version, 22.00).Pearson Correlation, t-test, mean, percentages were used to test research problems.In this study, the students' language achievement was determined by the average of their academic achievement scores at the end of 2015-2016 academic year.Thus, as the average academic score 60 and above refer to achievement in English.On the other hand, the scores below 60 stand for poor achievement in EFL.In order to determine gender effect on the scores of test anxiety and foreign language anxiety, an independent t-test was conducted.The pearson correlation analysis was performed to test the relationship between test anxiety and foreign language anxiety.

Findings
This research is carried out to define the correlation among university students' test anxiety, foreign language anxiety, and language achievement.In addition, it is aimed to assess the influence of gender on EFL students' test anxiety and foreign language anxiety.The first research question investigated whether there is a significant relationship between students' test anxiety and their language achievement.The findings are presented in Table 1.The findings in Table 1 show that there is no relationship between students' test anxiety and their English achievement.
The second research question examines if there is a significant relationship between gender and test anxiety.The related findings are delineated in Table 2.As it is viewed in Table 2, the mean of test anxiety scores for females (M= 45.91) is greater than that of the males (M= 37.53).The results of the independent t-test show that this difference is higly significant (p=.000).
Based on this finding, it can be reported that female students are significantly more anxious than male counterparts.
The third research question aims to determine whether or not there is a relationship between foreign language anxiety scores and language achievement.In Table 3 independent t-test result is displayed.As illustrated, there is a significant correlation between students' foreign language anxiety and their academic achievement.Comparing the mean of foreign langauge anxiety scores between successful students (83.41) and unsuccessful students (94.12) shows that there is a meaningful negative relationship between foreign language anxiety and achievement.The mean of test anxiety scores for unsuccessful students is higher than the mean of test anxiety score for successful students.The correlation between the FLCAS scores and language achievement is found to be negatively correlated (p=.001) indicating that as the level of anxiety increases, the language achievement decreases.In other words, this finding suggests that high levels of language anxiety is associated with low academic achievement.
In the fourth analysis, the researcher investigates the effect of gender on foreign language anxiety.Table 4 demonstrates related findings.Based on the research findings in Table 4, it is observed that female students' foreign language anxiety (M=92, 13) is higher than that of males (72.01).It is displayed that there is a statistically significant difference (p=.000) between female and male students' test anxiety.This finding points out that females are significanly more anxious than males.
In the last research question the researcher intends to test whether there is a relationship between students' test anxiety and their foreign language anxiety.To this end, a pearson correlational procedure is conducted.Table 5 displays related findings to that research question.Pearson product-moment correlation coefficients is run to explore the relationship between TA and FLA.The results from Pearson product moment correlation coefficients between FLCAS scores and English scores are designated in Table 5.The findings in Table 5 represent that there is a significant positive correlation between students' test anxiety and foreign language anxiety ( r= .61,p= .000).A positive ratio is observed between TA and FLA.The finding of a positive correlation between students' test anxiety and foreign language anxiety suggests that TA and FLA are related constructs and are not independent of each other.

Discussion
This study was conducted to determine the correlation among university students'test anxiety (TA), academic language achievement, and foreign language anxiety (FLA).In additon, it was aimed to define gender effect on both TA and FLA.Based on the results of the study, it was reported that there was no relationship between students' TA and language achievement.This finding of current research is consistent with previous studies.Yoğurtçu and Yoğurtçu (2013) found no correlation between TA and preparatory university students' language achievement.Cheraghian, Freidooni Moghadam, Baraz-Pardjani and Bavarsad (2008) observed no meaningful relationship between test anxiety and academic performance.The correlation between test anxiety and academic achievement does not mean that there is a direct cause-effect between these two variables.As Goonan (2003) aptly put it, "test anxiety does not have a direct impact on academic achievement, but it may manifest differently based on many factors, including familial background, level of achievement, motivation, and intellectual giftedness" (p. 7).In the same vein, Cassady and Johnson (2002) stated that it is not clear whether academic performance causes test anxiety or it is caused by a low academic proficiency.
TA has no statistically significant correlation with English achievement in this study.This finding is compatible with many similar studies examining the effects of anxiety on test performance (Gardner, 1989;Aida, 1994;In'nami, 2006;Birjandi & Alemi, 2010).The researchers pointed out that "test anxiety is an anxiety problem in general and not specifically related to the foreign language learning context" (In 'nami, 2006, pp. 329-330).In that sense, test anxiety is not a single factor that causes low language achievement.It is possible to propose the facilitative effect of test anxiety on achievement exists in the study group of this research.Some researchers support the idea that small amount of anxiety has positive effects on learning (Scovel, 1991;Horwitz, 2001).
Facilitative anxiety helps the learner to be more alert in cognitive individual thoughts and the individual's ability to construct and organize thoughts or to concentrate on the task or testing.The students may need to become anxious to perform a task to the optimum level (Scovel, 1991).Facilitating anxiety or an optimal amount of anxiety which can mobilize human beings to respond rapidly and efficiently.It is noteworthy that facilitating anxiety motivates learners to fight the new learning tasks and prompts them to make extra efforts to overcome their feelings of anxiety (Simpson, Parker, & Harrison, 1995).Additionally, worry component of TA is more correlated with academic performance.Based on the aforementioned considerations, it is claimed that the scores that the students obtained from worry dimension may also be very low.It is clearly evident that the students in the sample perceive testing situation as undangerous and nonthreating.The students did not fear taking test and the negative consequences of getting a bad mark.The present study suggests that test anxiety is not a salient predictor of language achievement.According to Gardner, Tremblay and Masgoret (1999), the students' personalities should be taken into consideration to determine the reasons of being unsuccessful.In addition, current research revealed a significant negative relation between students' FLA and their language achievement.Thus, it is possible to ascertain that high FLA may lead to low academic achievement.
In the current research it was found a significant relationship between gender and test TA.This research revealed that female learners exhibited higher levels of TA scores than male counterparts.This finding is in harmony with almost all relevant previous studies on gender effects on test anxiety in literature (Spielberger, 1980;Hembree, 1988;Di Maria & Di Nuovo, 1990;Zeidner, 1990;Seipp, 1991;Feingold, 1994;Bandalos, Yates & Thorndike-Christ, 1995;Berger & Schecter, 1996;Bishop, Baner & Becker, 1998;Zeidner, 1998;Ferrando, Varea & Lorenzo, 1999;Aysan, Thomson, & Hamarat, 2001;Brown, 2002;Cassady & Johnson, 2002;Altermatt & Kim, 2004;Stober, 2004;Chapell et al., 2005 ;McCarthy & Goffin, 2005;Aydın et al., 2006;Bacanlı & Sürücü, 2006;Lashkaripour, 2006;Mousavi et al., 2008;Rezazadeh & Tavakoli, 2009;Önem, 2011;Karataş, Alçı & Aydın, 2013;Saedi & Khaliliagdam;2013;Şener, 2015;Erzen & Odacı, 2014).Test anxiety and gender are significantly correlated in this study.This result demonstrated that female EFL learners experienced more anxious than males.One explanation for the gender differences in TA is that males and females experience similar levels of test worry, but females have higher levels of the emotionality component, producing higher general test anxiety scores (Deffenbacher, 1980;Mueller, 1980).The difference in the socialization patterns of males and females has been the primary explanation for this difference.More pressure is placed on females to succeed in school than males.This leads to the increase in TA scores because girls essentially are afraid to fail; each testing situation is seen as another possible chance of failure.Females are more apt to suffer from test anxiety.Another possible explanation is that "males are more defensive about admitting anxiety because it might be seen as threatening to their masculinity; they are trained to cope with anxiety by denying it or by finding ways to overcome it" (Mousavi, Haghshenas, & Alishahi, 2008).A little research findings reported no difference in the scores for girls' test anxiety levels (Aida, 1994;Chang, 1996;Aydın, 2013;Crişan & Copacı, 2015).
In the current research, it was reported a significant inverse relationship between FLA and language achievement.Young (1999) averred that anxiety impairs performance in language classes.There was a widespread consensus that highly anxious students obtain lower exam scores than more relaxed students.The high level of anxiety will be associated with low academic performance (Luigi et al., 2007).Text-anxious students can not concentrate on the test and thus they may get lower scores because they feel insecure about their own proficiencies and capabilities during the test (Pintrich & De Groot, 1990;Pintrich Smith, Garcia, & McKeachie, 1993).The present study reconfirmed that high anxious foreign language learners tend to perform relatively and significanly poorly than those who are less anxious.However, some researchers proposed that poor language learning is a cause rather than a result of language anxiety (Horwitz, 2001).
A considerable amount of research has been conducted to investigate the relationship between FLA and gender.
As for the correlation between TA and FLA, the results pointed out there was a significant positive correlation between students' TA and FLA.The fact that relatively strong positive correlation was observed TA and FLA indicated that students' TA and FLA were in line with each other.Namely, TA and FLA are related constructs and they are not independent of each other.Current finding is also consistent with previous research (Covington, 1992(Covington, -1998;;MacIntyre, Noels, & Clement, 1997;Bailey, Onwuegbuzie & Daley, 2000;Koralp, 2005;Önem, 2010;Tsai & Li, 2012;Tsai, 2013;Salehi & Marefat, 2014).Few previous research contended that test anxiety was not related to language anxiety (MacIntyre & Gardner, 1989;Aida, 1994;Matsuda & Gobel, 2004).The result of this study also argued that, in case of the presence of a type of anxiety, another type of anxiety may appear.In other words, foreign language anxiety and test anxiety, along with state and trait anxiety, seems to accompany each other.One possible explanation for that may be related to research which reported that individuals with high anxiety levels have a tendency to direct their attention towards possible anxiety provoking stimuli (Vasey, El-Hag, & Daleiden, 1996;Zeidner, 1998;Zeidner & Matthews, 2005).Therefore, if total of state-trait anxiety may be considered as a sign for the general tendency of anxiety, it can be suggested that when an individual has high level of generalized anxiety, s/he may reflect it to any cases felt as a threat.As mentioned earlier, a foreign language class or tests can be the source of anxiety.Thus, state and trait anxiety may serve as the common point between foreign language anxiety and test anxiety and it can be argued that a high level of state anxiety may accommodate other types of anxiety.FLA is classifed as a situational specific anxiety (MacIntyre & Gardner, 1991;Horwitz, 2001) rather than a personality trait.Situation specific anxiety is an individual tendency to be anxious in a particular time and situation.However, this may have resulted from the participants'being first year students at a preparatory class of English.This situation may have an effect on the level of state anxiety as every participant were studying English only (Önem, 2010).The reason for discrepancy between the results of the present research and other studies considering the relationship between TA and FLA might be the fact that other studies conducted factor analysis on FLCAS.These studies were based on Horwitz et al's (1986) study.It was clearly evident that one of the components of language anxiety is test anxiety and some of the items in FLCAS measure test anxiety.The previous studies conducted factor analysis on these items and found low factor loadings in them.The difference between results of this study and those of others may be due to using different statistical procedures and different number of items measuring test anxiety (Salehi & Marefat, 2014).
In the light of these data observed from this research, it is highly recommended that EFL teachers should develope effective anxiety reduction stategies to cope with anxiety-provoking situations appropriately.Test-taking strategy training program need to be included in EFL teaching, especially for highly text-anxious students to reduce the potential harmful effects of TA.Additionally, EFL teachers should determine the factors that cause students' high anxiety in all of different EFL skills and the reasons behind it.Efficient techniques are designed to prevent the increasing anxiety and promoting positive coping skills to improve academic performance.By practicing the techniques, students learn how to identify the unpleasant symptoms that appear with study anxiety and replace them with more relaxed feelings.In addition, EFL teachers should be involved in the pre-service and inservice training programs on both general psychology and language anxiety.Such programs might enhance FLA awareness in a way that enables teachers to deal effectively with the problems of stress and anxiety in the FL classroom setting.Upon an in-depth review of the exisiting literature on test anxiety and foreign language anxiety, it is suggested that EFL teachers should take into consideration learner differences in determining appropriate methods, techniques, material selection, and content of the course etc.
In conclusion, the present investigation provided valuable insight on TA and FLA in EFL context.It should be underlined that there are important avenues for future research and it is hoped that this study will foster additonal work in this exciting new educational realm.Specifically, it is recommended for future studies that the debilitative and facilitative factors should be examined elaboratively in EFL context.Hence, further longitudinal studies could be carried out to examine the relationships among TA, FLA, and foreign language achievement with different samples and dealing with certain related variables.

Table 1 .
Independent t-test results for test anxiety and language achievement

Table 2 .
Independent t-test results for test anxiety and gender

Table 3 .
Independent t-test results for foreign language anxiety and language achievement

Table 4 .
Independent t-test results for gender and foreign langauge anxiety

Table 5 .
The results of pearson correlation for test anxiety and foreign language anxiety