Relationship Between Language Anxiety and English Speaking Performance

,


Introduction
Speaking is one of four language skills, along with listening, reading, and writing. However, various researchers have proposed that speaking skills are more difficult to master than the other language skills. Speaking occurs in real time, meaning that the person you are talking to is waiting for you to speak right away (Bailey, 2005). Moreover, compared with writing, which can be revised or edited, speaking is more spontaneous. discovered that most learners aspire to speak fluently. Using assignments that require students to attempt real communication to develop their speaking ability, even when their English skills are limited (Hermaniar, 2021).
Speaking is a vital skill because it not only connects people speaking the same language but also helps them to express their thoughts, ideas, feelings, and emotions. Moreover, people who are illiterate deliver and transfer their ideas mainly through speech. The scope of speaking extends from simple conversation to formal public speaking (Salem & Dyiar, 2014), while other researcher argued that speaking is a fundamental skill essential for success in life (El-Basel, 2008). However, as speaking performance has gained much importance in L2 pedagoy, learners usually felt anxious when they needed to produce language and get involve in interpersonal communication. This anxiety may lead fear of speaking. Therefor, such fear may affect learners' speaking performance (Rofida, 2021). undesirable academic outcomes. The American College Health Association reported that more than 23% of college students believe that anxiety affects their academic achievement. Anxiety and performance are negatively correlated, and poor performance may lead to further anxiety and to a vicious cycle of decreasing performance (Susanti, Nabilah, & Irasanti, 2018).
Anxiety among learners of English as a foreign language has drawn the attention of numerous researchers in psychology and education. However, no consensus definition of anxiety has been reached. The current study considered anxiety to be a feeling of uneasiness, frustration, uncertainty, or worry (Brown, 1994;Cakici, 2016).

Speaking and Anxiety
Learners' proficiency and performance are factors related to FLA (Habisbuan & Irzawati, 2019). FLA hinders students' speaking performance and is a debilitative factor in language learning (Pamungkas, 2018). FLA is especially problematic for oral communication (Amini, Elfrida, & Kasmaini, 2019). Horwitz et al. (1986) identified speaking as one of the main sources of anxiety, which manifests as the fear of making mistakes and the avoidance of communication in testing contexts. Similar findings have been reported of anxious students in language learning contexts. A high level of FLA engenders communication apprehension, which can make learners unwilling to communicate (Hasibuan and Irzawati, 2019).
Both low-proficiency and high-proficiency learners are susceptible to anxiety. A study using the Foreign Language Classroom Anxiety Scale (FLCAS) discovered a moderate negative correlation between FLCAS scores and second-language achievement, implying that, when students start feeling anxious, they tend to avoid learning the target language (Hasibuan & Irzawati, 2019).
Although most researchers support the view that language anxiety in speaking classes may negatively influence language acquisition, others believe that anxiety can increase learners' ability to master a foreign language (Julianingsih, 2018) and their motivation to perform well in the language learning process (Pamungkas, 2018). The relationship between anxiety and performance remains debated. In review of the literature, it was found that there seems to be little exploring FLA in the context of technology-assisted language learning. Considering this attempt, the correlation between FLA and language performance in the online-based speaking test was an important issue that needs to be further explored. This study investigated this relationship during a computer-based oral exam by attempting to answer the following research questions: a) What are students' anxiety levels when taking the computer-based oral exam? b) Are foreign language speaking ability and oral exam anxiety related?

Participants
The participants of the present study were 59 seniors majoring in English who were enrolled in a one-semester course entitled English Communication Skills at a university in central Taiwan. Majoring in applied foreign languages, the participants (44 female students and 15 male students) had studied English for more than 12 years. All participants needed to take this course for two hours every week, giving a total of 36 hours. The participants' English proficiency level ranged from A1 to B2 levels according to the Common European Framework. In particular, four participants got A1 level; 22 participants obtained A2 level; 31 particiapnts reached to B1 level; two of the all got B2 level.

Instruments
The FLCAS was used as the main instrument in this study. It is a self-report questionnaire designed to measure learners' anxiety in the foreign language classroom and comprises 33 items scored on a 5-point Likert scale, with items rated from 1 (strongly disagree) to 5 (strongly agree). Horwitz (1986) measured the reliability of the FLCAS, revealing high internal consistency (Cronbach's alpha = .93) and high test-retest reliability of .83 (p = .001). To prevent any misunderstanding, the English version of the FLCAS translated into Chinese was used. In order to examine students' oral performance, an online-based mock Test of English for International Communication (TOEIC) Bridge Speaking Exam was delivered. The mock speaking test was designed based on the items from the TOEIC Bridge speaking section, including eight questions grouped into six task types. The qustions and tasks include two questions for reading a short text aloud, two questions for describing a photograph, one question for listening and retelling, one question for short interaction, one question for telling a story, and one question for making and supporting a recommendation.

Procedure
The study was conducted at the university in central Taiwan. The English Communication Skills course aims to enhance students' oral competency by delivering knowledge, promoting oral skills, and having students practice those skills. All participants completed the FLCAS questionnaire and took a mock Test of English for International Communication (TOEIC) Bridge Speaking Exam in the first week of the class. The Chinese version of FLCAS questionnaire was delivered to every student at the very beginning. After finishing all the questionnaire, the participants were asked to take a computer-based speaking mock exam in the lab in about 20 minutes. The researcher made sure every participant's microphone is positioned correctly and speak in the normal volume, and asked all participants to say as much as they can in the time allowed. After finishing this online speaking assessment, the students uploaded their audio files onto the computer-based speaking platform. Getting all the audio files of the speaking test, two native English speakers assigned scores based on the criteria of the TOEIC Bridge Speaking Exam.

Data Analysis
To examine the relationship between speaking performance and language anxiety, the researcher used SPSS software to analyze the participants' speaking and FLCAS scores. The linearity of the relationship between oral performance and anxiety was determined using the Pearson product-moment correlation statistic.

Results and Discussion
The mean FLCAS score of all participants was 3.25 (SD = 0.71), indicating a moderate level of anxiety during the speaking test. The participants' mean oral score on the mock TOEIC exam was 77.19 (SD = 6.81), indicating a high-intermediate level of oral proficiency (Table 1). According to Horwitz (2008), a mean score of approximately 3 on the anxiety scale indicates that a student is slightly anxious, and a score of less than 3 indicates low anxiety (Horwitz, 2008). However, students with a mean score of 4 and above are considered to be very anxious (Pamungkas, 2018). The finding of a moderate level of anxiety (M=3.25, SD=0.71) during the speaking test indicates that most of the students seemed to be slightly anxious when performing speaking test. Such finding echoes the result of previous studies (Yao and Dong, 2019;Lui, 2006), which also reported that students have a moderate level of anxiety in speaking context. As shown in Table 2, many participants (27/59, 45.76%) had moderate speaking anxiety, and some (20/59, 33.90%) had high speaking anxiety. Eight (13.56%) had low speaking anxiety, whereas only four (6.78%) had very high speaking anxiety. The results indicated that most of the participants reported moderate to high speaking anxiety when they take the speaking test. These findings corroborate those of previous studies (Yao & Dong, 2019;Martin & Valdivia, 2017). When Yao and Dong (2019) pointed out that English majors commonly experienced moderate anxiety in learning English, Martin and Valdivia (2017) stated that students reported higher levels of anxiety when doing oral tasks.

Minimum Level 64
Mean 77 The Pearson product-moment correlation between speaking anxiety and speaking performance was negative (−.082), implying that higher speaking scores are associated with lower FLA and vice versa. Yet, this relationship was nonsignificant (p > .05; Table 4). Several studies have reported the same tendency regarding these two variables (Hasibuan & Irzawati, 2019;Huang, 2018;Liu, 2006;Rofida, 2021;Tien, 2018;Hasibuan & Izawati, 2019). However, when those studies showed that foreign language anxiety has significant negative correlation with speaking performance, the researcher in the present study only found negative correlation. As Hewitt and Stephenson (2012) indicated, the correlation between foreign language anxiety and speaking exam scores measured by FLCAS and TOEIC Bridge speaking test was negative, suggesting that students exhibiting higher levels of language anxiety performed more poorly on their oral exam than did their less anxious counterparts.

Conclusion
This study investigated the second-language speaking anxiety of students during computer-based oral exams and its relationship with oral performance. Many students taking the computer-based speaking exams had moderate speaking anxiety and high-intermediate speaking proficiency. In line with previous studies (Hasibuan & Irzawati, 2019;Huang, 2018;Liu, 2006;Pamungkas, 2018;Rofida, 2021;Tien, 2018), this study found a negative correlation between FLA and speaking performance, but this relationship was nonsignificant, implying that anxiety may influence students' speaking performance to some extent in certain contexts. Although the relationship between FLA and speaking performance in the present study was negative, the p value did not reach to a significant level. This result is different from most of the literature, motivating more researchers to conduct studies on relevant topics.
Pedagogically, there is an urgent need to alleviate students' level of anxiety in order to improve their oral performance. The findings highlighted the importance of providing speaking anxiety-coping strategies with learners in the L2 online testing context (Huang, 2018). Those anxiety-coping strategies may allow test-takers to perform better when they are taking the online speaking test. Also, speaking skills or strategies are suggested to be taught in oral training program before the tests.
Further research should investigate the relationship between FLA and oral performance in different computer learning situations. Furthermore, qualitative methods, such as open-ended questionnaires, interviews, and learning logs, may help to provide a more comprehensive understanding of the relationship between anxiety and speaking performance in foreign language learning.