Sources of Turkish Efl Learners' Foreign Language Reading Anxiety

The purpose of this study was to investigate possible sources of reading anxiety experienced by Turkish EFL learners. The participants of the study were 50 first year university students studying at the English Language Teaching program of a public university in Turkey. The study employed the explanatory sequential mixed methods design. First, the quantitative data were collected through Foreign Language Reading Anxiety Scale (FLRAS), and then the qualitative data were collected through semi-structured interviews for the purpose of better understanding quantitative data findings. For analyzing the quantitative data, each participant's mean score from FLRAS was computed. As for the qualitative data, the transcribed interviews were analyzed using the constant comparison method, for which the researcher identified the recurring themes through coding, prioritizing and connecting the pieces of data. Results indicated that the participants of the study, in general, experience a moderate level of FL reading anxiety. The following three main categories of FL reading anxiety sources were identified with their ten sub-categories: (1) personal factors (inappropriate reading strategy use, lack of self-confidence, high expectations), (2) features of the reading text (topic familiarity, unknown vocabulary, complex reading structure, text length, figurative language), and (3) reading course (compulsory reading, exam).


INTRODUCTION
Why do some learners succeed in foreign language learning and others fail? In an attempt to find an answer, foreign language (FL) researchers have examined numerous factors that may affect VWXGHQWV· DFKLHYHPHQW LQ IRUHLJQ ODQJXDJH OHDUQLQJ 7KHVH VWXGLHV addressed cognitive variables (e.g. language aptitude, language learning strategies, cognitive ability), affective variables (e.g. anxiety, self-perception), personality (e.g. locus of control, individualism) and demographic variables (e.g. age) for the reason that these variables seemed to be related to FL achievement (Subasi, 2014;Onwuegbuzie, Bailey, & Daley, 2000). Among them, affective variables have received the attention of researchers in the way to understand their potential for success or failure in language learning (Kuru-Gonen, 2005;Rodriguez & Abreu, 2003). As being one of the affective factors, anxiety has also drawn the attention of FL researchers. Notably, they have conducted research on FL anxiety as they have recognized the existence of FL anxiety and its potential for significant interference with language learning and production (Saito, Garza, & Horwitz, 1999).
FL anxiety can be deILQHG DV ´D GLVWLQFW FRPSOH[ RI VHOIperceptions, beliefs, feelings, and behaviors related to classroom language learning arising from the uniqueness of the language OHDUQLQJ SURFHVVµ +RUZLW] +RUZLW], & Cope, 1986, p.31). This implies that FL anxiety is a phenomenon distinct from general anxiety; it is unique to foreign language learning (Horwitz, 2001;Oxford, 1999). As far as FL anxiety is concerned, it is claimed that this DQ[LHW\ KDV D SRWHQWLDO QHJDWLYH HIIHFW RQ OHDUQHUV· ODQJXDJH XVH (MacIntyre, 1999). According to Worde (1998), one third to one half of language learners examined indicated experiencing debilitating levels of FL anxiety.
At first glance, reading would appear to be the least anxietyprovoking part of the FL performance (Saito, Garza, & Horwitz, 1999). Unlike speaking a FL, reading is done silently with unlimited possibilities for reconsideration and reflection (Saito, Garza, & Horwitz, 1999). FL reading also does not depend on spontaneous interaction in the way speaking does (Zhao, Guo & Dynia, 2013). From a cognitivist and pedagogical perspective, however, FL reading would seem to have great potential for provoking anxiety (Subasi, 2014;Kuru-Gonen, 2005). From a cognitivist view, FL reading elicits anxiety by interfering with processing capacity and thus decreases the amount of attention that the student has to pay to the reading task (Kuru-Gonen, 2005). When FL anxiety is examined from pedagogical perspective, it becomes obvious that FL reading process is further complicated where additional factors are considered such as learner motivation, language ability, cultural backgrounds and unfamiliar scripts and writing (Saito, Garza, & Horwitz, 1999). Many students learning a FL can experience the feeling of tenseness and frustration when they try to understand reading texts in FL; that is, they can experience anxiety. From this aspect, FL reading anxiety is the anxiety that students experience throughout the FL reading process and therefore, is related to the specific language skill of reading (Zhao, Guo & Dynia, 2013).
As previously mentioned, FL anxiety has been studied a lot and the results of these studies generally revealed that FL anxiety has a negative effect on FL performance (Yan & Horwitz, 2008;Young, 1991, MacIntyre & Gardner, 1991Horwitz, Horwitz, & Cope, 1986). FL reading anxiety, on the other hand, has taken the FL researcKHUV· attention recently (Subasi, 2014;Kuru-Gonen, 2005). Recent research on language anxiety problems in reading has shown that FL reading anxiety is prevalent among FL learners (Subasi, 2014;Zhao, Guo & Dynia, 2013;Shi & Liu, 2006;Kuru-Gonen, 2005;Saito, Garza, & Horwitz, 1999;Oh, 1992). Nevertheless, many questions such as whether FL anxiety can be distinguishable from general anxiety and what could be the sources of FL reading anxiety, await answers (Subasi, 2014;Zhao, Guo & Dynia, 2013;Shi & Liu, 2006;Kuru-Gonen, 2005;Saito, Garza, & Horwitz, 1999). In order to provide additional insights into the existence of language skill specific anxiety, further examination of the relationship between FL anxiety and reading and exploration of the possible sources of FL reading anxiety are needed (Subasi, 2014;Kuru-Gonen, 2005). In this sense, the purpose of the present study is to investigate possible sources of reading anxiety experienced by Turkish EFL students.

REVIEW OF LITERATURE
$Q[LHW\ LQ LWV VLPSOHVW WHUP FDQ EH GHILQHG DV ´DQ XQSOHDVDQW emotional state or condition which is characterized by subjective feelings of tension, apprehension, and worry, and by activation or DURXVDO RI WKH DXWRQRPLF QHUYRXV V\VWHPµ 6SLHOEHUJHU et al., 1983, p.125). Studies on anxiety propose that anxiety can be experienced at different levels (Oxford, 1999). In this sense, two types of anxiety are characterized: trait anxiety, a more permanent disposition to be anxious, and state anxiety, anxiety that is experienced in relation to some particular event or act.
One of the first attempts to treat FL anxiety as a separate and distinct phenomenon specific to language learning was made by Horwitz, Horwitz and Cope (1986). They differed general communication anxiety from FL anxiety, focused particularly on the situational nature of state anxiety, and concluded WKDW )/ DQ[LHW\ LV ´a distinct complex of self-perceptions, beliefs, feelings, and behaviors UHODWHG WR FODVVURRP ODQJXDJH OHDUQLQJ SURFHVVµ (p. 128).
As Horwitz (2001) states how anxLHW\ DIIHFWV )/ OHDUQHUV· performance is only one of the issues to concern when discussing DQ[LHW\ DQG ODQJXDJH OHDUQLQJ DQG ´SHUKDSV QRW HYHQ WKH PRVW importantµ S much importance should be given to understand what causes the frustration and uneasiness that many FL learners experience in language classrooms. Many studies conducted to explore the sources of FL anxiety have revealed that speaking in front of others is one of the strongest sources of anxiety (Phillips, 1990(Phillips, , 1992Young 1990Young , 1991Young , 1992. Apart from the sources of general foreign language anxiety, researchers recently have paid attention to the relationship between FL anxiety and four language skills, namely listening, speaking, writing and reading (Horwitz, 2001). In this regard, Zhao (2009) pinpoints that different sources of FL anxiety associated with different language skills exist and it should be explored to have a deeper insight into the nature of FL anxiety.
Reading as one of the major aspects of language learning also involves certain difficulties (Tsui, 1996). According to Sellers (2000), reading is not just difficult when it is performed in the target language. Actually, reading in any language is difficult since the process is cognitively demanding and includes minimally the coordination of attention, memory, perception, and comprehension (Sellers, 2000). As far as foreign language context is considered, reading process is even more complex due to the additional factors such as language ability, learner motivations and cultural background (Sellers, 2000). Authentic texts and unfamiliar context of these texts can be possible causes that make reading an anxiety provoking skill.
Authentic texts can be anxiety provoking because learners may not attach the meanings of unfamiliar FL words in a reading passage. Cultural background may also be anxiety provoking because learners unfamiliar with the target language culture may not make sense of the reading passage even if they recognize the words (Saito, Garza, & Horwitz, 1999).
Saito *DU]D DQG +RUZLW]·V (1999) study is one of the first studies to construct FL reading anxiety as a phenomenon distinguishable from the general FL anxiety. The findings of their study revealed that FL reading anxiety is distinguishable from general FL anxiety and general FL anxiety is independent of target language. However, levels of reading anxiety are revealed to vary by target language, and appear to be related with the two possible sources of reading anxiety: specific writing systems and unfamiliar cultural materials. Zhang (2002) reports grammar and worry about the reading effect as sources of FL reading anxiety. In the study conducted with 90 overseas students learning Chinese in China, the results showed that when compared to Korean and Japanese students, reading was more anxiety provoking for European and American students. While European and American students experience high anxiety in general reading comprehension, Korean and Japanese students get worry more when they face with unknown grammar structures. Also, the results indicated that European and American students have rather less anxiety in relation to cultural related items compared to students in 6DLWR *DU]D DQG +RUZLW]·V (1999) study.
Furthermore, in another study by Oh (1992), reading task type was indicated as a source of FL reading anxiety. Oh (1992) examined the relationship between reading assessment methods and the level of FL reading anxiety. The findings of the study revealed that when compared to comprehension and recall tasks, cloze and think-aloud tasks caused more anxiety. Oh proposed three possible reasons for WKLV UHVXOW OHDUQHUV· IDPLOLDULW\ ZLWK WKH UHDGLQJ DVVHVVPHQW PHWKRGV FOR]H DQG WKLQN DORXG PD\ EH OHVV NQRZQ OHDUQHUV· SHUFHSWLRQ LQ relation to the validity of an assessment method, and the difficulty level of the task.
In Turkish context, Kuru-Gonen (2005) and Subasi (2014) aimed to examine whether FL reading anxiety exists as a phenomenon distinguishable from general FL anxiety for Turkish learners of English. The results of both of the studies revealed that FL reading anxiety exists as a distinct phenomenon and students were experiencing FL reading anxiety. For the sources of FL reading anxiety experienced by the participants, Kuru-Gonen (2005) reported the following three main sources: personal factors, reading text, and reading course; Subasi (2014), on the other hand, reported six main sources: personal reasons, teaching procedures used in the class, WHDFKHU·V PDQQHU LQ WKH FODVV IHDWXUHV RI UHDGLQJ WH[W UHDGLQg text anxiety, and previous experience.

METHOD Participants
The participants of this study were 50 first year university students studying at the English Language Teaching program of a public university in Turkey. All participants were monolingual speakers of Turkish between the ages of 17 and 20. When the data of this study were collected, all the participants had completed an EFL UHDGLQJ FRXUVH FDOOHG ¶Academic Reading· DQG WKH\ ZHUH taking DQRWKHU ()/ UHDGLQJ FRXUVH FDOOHG ¶Critical Reading in English Language Teachingµ (seven weeks of the course had been completed).

Instruments, Data Collection, and Data Analysis
This study employed WKH ´H[SODQDWRU\ VHTXHQWLDO PL[HG PHWKRGV GHVLJQµ &UHVZHOO Accordingly, first, the quantitative data were collected through Foreign Language Reading Anxiety Scale (FLRAS), and then the qualitative data were collected through semistructured interviews for the purpose of better understanding quantitative data findings.
FLRAS was developed by Saito, Garza, and Horwitz (1999) to assess the level of anxiety directly linked to FL reading. It includes 20 items that are answered on a 5-point sFDOH UDQJLQJ IURP ´6WURQJO\ $JUHHµ WR ´6WURQJO\ 'LVDJUHHµ. Lower mean scores from FLRAS indicated more anxiety. FLRAS was administered during the SDUWLFLSDQWV· FODVV KRXUV After the data coming from FLRAS was analyzed, semi-structured interviews were conducted with six participants who were found to be high anxious in order to find out the possible sources of foreign language reading anxiety. Interview questions were prepared in the light of the related literature, and expert opinion was taken in order to ensure the validity of the questions for the research purpose. For analyzing the quantitative GDWD HDFK SDUWLFLSDQW·V PHDQ score from FLRAS was computed. As for the qualitative data, the transcribed interviews were analyzed using the constant comparison method, for which the researcher identified the recurring themes through coding, prioritizing and connecting the pieces of data (Dantas-Whitney, 2002).

RESULTS AND DISCUSSION
The mean score on the FLRAS was found as 2.74 (SD = 0.44). This result indicates that the participants of the study, in general, experience a moderate level of FL reading anxiety. The participants with high, medium, and low FL reading anxiety were determined by adding and subtracting the standard deviation to the mean score. Accordingly, the participants whose mean FLRAS scores were more than 3.18 (2.74 + 0.44) were identified as experiencing low anxiety, the ones whose mean FLRAS scores were less than 2.3 (2.74 ² 0.44) were identified as experiencing high anxiety, and the ones whose mean FLRAS scores were between 2.3 and 3.18 were identified as experiencing moderate anxiety. As a result of the data analysis, 4 participants were identified experiencing high anxiety, 45 participants were identified as experiencing moderate anxiety, and 1 participant was identified as experiencing low anxiety.
In order to find out the sources of FL anxiety, semi-structured interviews were conducted with the four participants identified as experiencing high FL reading anxiety.

FL Reading Anxiety Caused by Personal Factors
Participants of the study addressed some personal factors as the sources of their FL reading anxiety. A total of three sub-categories were revealed under the concept of personal factors: (1) inappropriate reading strategy use, (2) lack of self-confidence, and (3) high expectations.
The analyses of the interviews demonstrated that students might experience reading anxiety due to not using effective strategies or using some certain strategies inappropriately such as starting reading from the very first line and going word by word, stopping to look up every unknown word instead of getting the meaning from the context, and relying mostly on linguistic knowledge. In addition to these, focusing on little detail and missing the main idea of each paragraph and how the ideas are connected were also detected as the sources of reading anxiety related to inappropriate strategy use. The following participant stated this feeling as: Furthermore, some participants stated that they lack selfconfidence in FL reading because they think that not having the good habits of FL reading affects their development of reading skill in the target language; and therefore, the sense of being inadequate in FL reading leads them to escalate the anxiety. The following excerpt shows how lack of self-confidence causes anxiety:

´$FWXDOO\ , OLNH UHDGLQJ %XW , DP QRW D JRRG UHDGHU
Especially, I rarely read in English. This makes me upset. I know the role of reading in proficiency. Because of not having good habits of UHDGLQJ , IHHO P\VHOI LQDGHTXDWH , WKLQN WKDW·V why I lack self-confidence in reading in English. In time, this sense of insufficiency turns into anxiety and negatively affects PH µ High expectations emerged as another anxiety provoking part of the reading process related to the personal factors. Some participants expressed having such a belief that they have to set their standards in relation to reading skill as high as the level of native speakers. According to them, these high expectations lead to frustration due to their inability to attain the high standards and clash the outcomes in reality. The following student expression demonstrates how students might be affected in a negative way by high-expectations:

´, NQRZ WKDW LW LV LPSODXVLEOH WR VHW P\ VWDQGDUGV DV KLJK DV
the level of native speakers in terms of speaking, listening and writing skills. But, I believe that I can become proficient in reading as much as they are. However, when I have difficulty in comprehending the main idea conveyed through the text, I see that I am not able to attain the standards I set myself for reading. This makHV PH IUXVWUDWHG µ

FL Reading Anxiety Caused by Features of the Reading Text
Some participants reported that during the reading process there is an interplay between cognitive and affective conflict. That is, while trying to understand the written content, some factors make them anxious and hamper their understanding. These factors were formed under the category of the reading text and they are: (1) topic familiarity, (2) unknown vocabulary, (3) complex reading structures, (4) text length, and (5) figurative language.
According to the participants who identify unfamiliar topics as one of the main factors for FL reading anxiety, their background knowledge or familiarity with the reading topic influences their reading comprehension. In relation to this, they stated that once they are faced with a text on which they have no background information, they get anxious, and even sometimes this anxiety blocks their understanding. One of the participants stated: Another source of anxiety related to the reading text is unknown vocabulary. According to some participants, comprehension of the text which is embedded with too many unknown words creates gaps in the meaning of the text, which, in return, causes difficulty in inferring the meaning from the context. Because of constructing insufficient amount of meaning, they cannot make sense of the text and start experiencing anxiety. The following excerpt illustrates how unknown vocabulary affects reading: ´:KDW PDNHV PH QHUYRXV D ORW LV FRQIURQWLQJ ZLWK XQNQRZQ vocabulary items. Those words generally are the key words that you need to know in order to understand the passage. When I feel that I cannot infer the meaning of the unknown words from the context, I gLYH XS UHDGLQJ µ Complex linguistic structures in a reading text appeared to be another source of FL reading anxiety for the participants of the study. With respect to this, participants stated that reading texts containing relatively complex linguistic structures and features such as complex grammar cause high level of anxiety. Accordingly, they expressed that when they confront with such a complex pattern, they suddenly get a feeling of failure in understating and as a result, they are negatively affected. One of the participants expressed this feeling as As shown in Table 1, some participants think that the length of a text is another source of anxiety for them. That is, confronting with a whole page full of English paragraphs leads to anxiety. They have such an intuitive belief that the longer a reading text, the more difficult it will be. One of the participants expressed this feeling as follows: ´/RQJ WH[WV RU HYHQ long paragraphs are enough to frighten me. )URP WKH YHU\ ILUVW OLQH , IHHO DV LI , GRQ·W XQGHUVWDQG ZKDW , DP reading. At that moment, I start to read quickly to forget that the text is long. But, that makes me remember it more. Within these attempts to read, I JHW IUXVWUDWHG µ The last sub-category related to anxiety caused by features of the reading text is figurative language used in the texts. Some participants expressed that when they face with figurative language in a reading text, they experience difficulty in making the right inferences due to their inability to decode figurative expressions. Accordingly, they reported that figurative expressions in English disturb their reading comprehension and trigger the release of anxiety. The following excerpt is an example:

FL Reading Anxiety Caused by the Reading Course
Concerning the sources, the reading course appeared as the last main category of FL reading anxiety. Two sub-categories were appeared under this category: (1) compulsory reading, and (2) exam.
In relation to compulsory reading, some participants reported that attending a reading class as a compulsory course diminishes pleasure of reading and makes them get away from the reading, and therefore, leads to anxiety. The following participant stated this feeling as: ´6RPHWLPHV , GR QRW IHHO OLNH UHDGLQJ IRU HGXFDWLRQDO purposes. If it is something that I take pleasure ²like reading comics, then I can read somehow. But, when I am forced to read by someone, for example by the teacher in the reading course, reading becRPHV WRUWXUH IRU PH µ Exam is another sub-category emerged from the interviews. According to some participants, the concern of passing the class is one of the chief sources of anxiety for them. One of the participants expressed this feeling as:

CONCLUSION
The purpose of this study was to investigate possible sources of reading anxiety experienced by Turkish EFL learners. Results indicated that the participants of the study, in general, experience a moderate level of FL reading anxiety. Of the 50 participants of the study, four participants were identified experiencing high FL reading anxiety, 45 participants were identified as experiencing moderate FL reading anxiety, and one participant was identified as experiencing low FL reading anxiety.
Semi-structured interviews were conducted with the four participants identified as experiencing high FL reading anxiety in order to reveal the sources of FL anxiety. According to the results of the analysis of the interviews, the following three main categories of FL reading anxiety sources were identified with their ten subcategories: (1) personal factors (inappropriate reading strategy use, lack of self-confidence, high expectations), (2) features of the reading text (topic familiarity, unknown vocabulary, complex reading structure, text length, figurative language), and (3) reading course (compulsory reading, exam).
The results of this study not only would enable FL educators to gain better understanding of the nature of FL reading anxiety, but it would also make them more sensitive to organize anxiety free FL reading programs by taking into account the identified sources of reading anxiety in learning a new language. Through re-designing programs and creating a positive learning environment in FL classroRPV LW ZRXOG EH PRUH SRVVLEOH WR UHGXFH OHDUQHUV· IHHOLQJV RI anxiety.