University Students ’ Self-Efficacy : A Contributing Factor in TOEFL Performance

Self-efficacy is one of the often-underestimated factors, which in fact, has an influence in determining human action. This study aims to find out the correlation between self-efficacy and the TOEFL performance of the students in two universities in Banda Aceh, namely Syiah Kuala University and Serambi Mekkah University. This study employed the correlational design to find out the correlation between the two main variables (selfefficacy and TOEFL performance). The sampling technique used in this study was the stratified sampling method with a total sample of 200 university students, both English, and Non-English majors. Furthermore, to collect the data, two instruments were used, namely the self-efficacy questionnaire and the TOEFL test. In analyzing the data, Pearson’s Correlation was used to find out the correlation. The result shows that there is a positive and low correlation between university students’ selfefficacy and their TOEFL performance (r=.249). Despite the correlation is low, self-efficacy is still considered to contribute to the TOEFL score. Therefore, it is suggested that TOEFL instructors consider the variable of self-efficacy in their teaching. Furthermore, it is expected that future research continues this study by enlarging the sample size in different levels of education and in different areas.

Since English has been claimed as the global language in this modern era (Crystal, 2003), it plays a more significant role in the development of knowledge and technology.In Indonesia, for example, English proficiency has been used as the requirement to achieve many important stepping stones in life.This includes a requirement to enrol in certain universities (Abboud & Hussein, 2011;Sharpe, 2004), a requirement to be able to do the thesis defence in order to graduate from the university, a requirement to register for a scholarship and receive the scholarship in order to continue the study in the higher level either in the country or abroad, as well as to join the exchanging students or internship program, a requirement to apply for a job, particularly in the national and international standard company, and also a requirement to have a higher position and certification in working institutions.This is supported by Ananda (2016) who affirms that TOEFL mastery is very crucial, particularly to show that people have a good English proficiency and to easily get access to further the study either abroad or in Indonesia.Therefore, English should be mastered by all academicians and some other professions, especially to be able to reach the prestigious life milestone and future career.
A number of tests are frequently used to measure one's English capability.Some of them include TOEFL (Test of English as a Foreign Language), TOEIC (Test of English for International Communication), and IELTS (International English Language Testing System).However, one of the often-used tests in Indonesia is TOEFL; a test that measures the English ability as the foreign language through a set of multiple-choice questions.This 105-minute test measures the English ability in three sections, namely listening, structure-written expression and reading.The score of TOEFL varies from the lowest of 217 to the highest of 677 (Phillips, 2008).In terms of the TOEFL capability, some previous research claims that the TOEFL scores of the university students in Indonesia were still below the average of 450.Zaitun (2014) reports that only 5% of the non-English major university students in Teacher Training and Education Faculty at Sriwijaya University had a score of above 450.In fact, the minimal requirement that has been decided to enable the students to register for the thesis defense, graduate from the university, as well as to apply for a job or scholarship is generally above 450.This shows that it is essential to find out the solution to help improve the university students' TOEFL score so that the university students, particularly in Aceh, Indonesia, will be able to achieve the next higher educational degree and the future career based on their skills and interests.
There are many different factors influencing university students' low score of TOEFL.One of the possible factors is self-efficacy.Bandura (1997) defines selfefficacy as people's personal beliefs or expectation of their capability to accomplish certain things.He further explains that some people might have a positive self-efficacy in which they will think positively about their ability in doing something; whereas some others have negative self-efficacy.Theoretically, self-efficacy plays an important role not only by directly influence the action but also by giving the influence on the chosen strategy, consistency to do the efforts in order to achieve what has been set, as well as the ability to face the possible challenge that may hinder people to achieve certain target or goal (Bandura, 1995(Bandura, , 1997)).
Furthermore, some studies related to self-efficacy have been conducted, especially about the relationship between self-efficacy and the students' English ability limited to certain English skills (speaking, listening, reading, and writing).The studies in the speaking skill, particularly related to the correlation between self-efficacy and speaking performance revealed a significant and positive correlation and the strength of the correlation was medium (Kitikanan & Sasimonton, 2017;Mastur, 2016;Maulida, 2016;Zulaikha, 2015).In terms of the listening skills, the medium positive and significant correlation was also found between self-efficacy level and the students' listening skill (Isty et al., 2017;Kitikanan & Sasimonton, 2017).In the studies about reading skill, Kitikanan and Sasimonton (2017) found a medium positive correlation, while Ningsih (2014) found a low positive correlation between self-efficacy and the students' reading capability.Finally, in the writing skill, a medium positive correlation between self-efficacy and the students' writing ability was also found by Kitikanan and Sasimonton (2017).
Moreover, some other studies focusing on the relationship between the students' self-efficacy and English performance as general, which was taken from the English score in the students' report or the English score from the school national examination were also found (Ish & Rozali, 2016;Kitikanan & Sasimonton, 2017;Mahyuddin et al., 2006;Yusuf, 2011).Specifically, a study by Ish and Rozali (2016) found a low correlation r=0.347, while a study by Kitikanan and Sasimonton (2017) found a medium correlation r=0.60 in terms of the relationship between self-efficacy and English performance in general.However, the study that focuses on the correlation between the university students' level of self-efficacy and their TOEFL capability was still limited, particularly in Aceh, Indonesia.Therefore, this study aims to find out how far is the correlation between the two variables: self-efficacy and TOEFL performance.

Definition of Self-Efficacy
Self-efficacy is a familiar term in the world of education nowadays.Self-efficacy is defined as the personal beliefs of capabilities to conduct and accomplish certain tasks (Bandura, 1995).Kreitner and Kinicki (2005) further define self-efficacy as the beliefs of every individual about their capacity to face and solve a problem effectively.In addition, Bandura (1997) also claims that self-efficacy influences how people think (cognitive), feel (affective), motivate themselves and act (Bandura, 1995).A study by Lerner and Locke (1995) reveal that people with a high level of self-efficacy will be able to reach the target because they work maximally compared to those who have a low level of self-efficacy.Self-efficacy itself is formed because of the availability of energy (internal motivation), sources (situational factor), and the ability to understand appropriate action (role perception) as well as competence (capability) in completing tasks (Kreitner & Kinicki, 2005).Some previous studies about self-efficacy have proved that self-efficacy was related to self-control, resilience in facing failure, as well as effort and performance in solving the problem (Cherian & Jacob, 2013).
Based on the above definition, it could be concluded that self-efficacy is a personal belief about self-capability in facing different situations and conduct certain things as well as to succeed in accomplishing the actions.Therefore, people with a high level of self-efficacy will be able to force themselves to find out the solution to the problems they face and reach the target goal effectively.
2.2 Sources of Self-Efficacy Warsito (2009), in his study, claims that there are some sources of self-efficacy.They are explained as the following.1. Magnitude Self-efficacy of every people is based on their life demand and is different from each other.Every person has a different perception of the demand for every task given to them that they have their own strategy in order to face the challenge and perform maximally.There must be a variety of difficulty level to be able to perform actions maximally.When the challenge to reach the target is relatively small, the actions will be easier to do so that the individual will have a higher self-efficacy level.

Strength
Another factor influencing self-efficacy is the experience the people have.Lack of experience will result in a lack of self-efficacy, while people with a lot of relevant experience will have a strong belief on themselves to face all the challenge and perform the action effectively.Because experience forms people's self-confident, which is a very important aspect to succeed in life.

Generality
Generality is related to the work standard demand, which is varied across professions, including types of activities, the people in the workplace, and the surrounding areas.Each person is used to work or act with the standard; therefore, whenever they are given the task similar to the generality, they will have a high level of self-efficacy to conduct the task.
To sum up, magnitude, strength, and generality are the three important sources of self-efficacy that could help improve the self-efficacy.Therefore, it is suggested to develop those three sources, for example by providing more relevant experiences in order to help increase the personal self-efficacy.Bandura (1997) states that there are some dimensions of self-efficacy.They are explained as the following.

Past Performance
This dimension considers new things that people receive as the accumulation of their previous works.By having past performance, people become more confident in their own capability (self-efficacy is higher), particularly in performing similar tasks.

Vicarious Experience
Vicarious experience refers to the observation of other people's success.The successful performance by colleagues who are in the same field or who likely have the same capability will increase people's self-efficacy in conducting the tasks.Therefore, the modeling performance by peers is considered important to enhance self-efficacy.

Verbal Persuasion
Verbal persuasion refers to advise, suggestion, recommendation, and guidance either from the leader or colleague.The people's self-efficacy towards the capability of conducting certain tasks will increase by listening to advice, suggestion, recommendation, and guidance from others.Therefore, they can accomplish the target more easily if the surrounding people give them advice and suggestion.

Emotional Cues
Emotional cues include emotional feeling when conducting certain activities.The negative feelings, such as fear, anxiety, and stress will decrease self-efficacy.
Meanwhile the increased of emotions, which is not excessive will increase selfefficacy.Therefore, the ability to control emotion is essential to help people increase self-efficacy in reaching the target.
In conclusion, these four different dimensions of self-efficacy (past performance, vicarious experience, verbal persuasion, and emotional cues) also contribute to varieties of ways that can help increase self-efficacy.For the past performance and vicarious experience, enhancing self-efficacy could be done by providing an opportunity to perform similar tasks in different occasions.Meanwhile, for verbal persuasion, the self-efficacy could be enhanced through being with a community who has a positive feeling that supports each other.Finally, for emotional cues, stress control is one of a solution that could improve self-efficacy.

Definition and Kinds of TOEFL
TOEFL (Test of English as a Foreign Language) is a standardized test that is generally used to measure people's English capability.TOEFL was first created and developed by ETS (Educational Testing Service) in New Jersey.Previously, this test was only used by the surrounding community to measure their English capability and was particularly used to continue the study in the higher education level (Tumijo & Riyanto, 2009).Tumijo and Riyanto (2009, p. 10)  There are some language centers in Aceh, for example, the language center in Syiah Kuala University and the language center in Ar-Raniry State Islamic University located in Banda Aceh, as well as KIES (Kangguru International Education Service) located in Greater Aceh.This kind of PBT (ITP) is frequently used as the international benchmark of people's English capability and is generally used as a requirement to pass educational scholarship, such as to continue to higher education or to apply for jobs.

CBT-TOEFL or Computer Based Test TOEFL
This kind of TOEFL test used formal software from ETS (Educational Testing Service) as the test medium.In this test, the test takers answer the TOEFL questions by using computers in four sessions, namely listening, structure, reading, and writing.Besides, this test is done between 2 until 2.5 hours.Furthermore, the score in this CBT-TOEFL is ranging from 0 to 330.This test is still rarely used in Indonesia.

iBT-TOEFL or Internet Based Test TOEFL
iBT-TOEFL has been introduced and applied internationally since 2005.Therefore, this kind of TOEFL test should be conducted worldwide and is the TOEFL test that is admitted globally.This iBT-TOEFL is done by using a computer online, without a special TOEFL software as the test media.In this test, the test takers answer the TOEFL questions provided by ETS (Educational Testing Service) online.This TOEFL test also has 4 sessions, namely listening, writing, reading, and speaking, with a maximum score of 120, and is conducted within 4 hours.Furthermore, this test is divided into two parts including the individual test and the integrated test.
In short, the different kinds of TOEFL test has a different format of application and evaluation.It is suggested that the test takers prepare themselves with strategies to be able to achieve the targeted score.

TOEFL Scoring Rubric
The process of marking TOEFL is not the same as marking other types of multiple-choice test.It uses a converted table that is usually provided in many TOEFL book.Having seen the converted score, the process of the calculation of TOEFL score is well-explained in the following table.

The number of correct answers
The converted score In conclusion, this process of marking requires a process of score conversion and calculation.This is why the TOEFL score starts from the lowest of 217 to the highest of 677.

METHODS
This study is one of the quantitative studies that use correlational research design.Specifically, the design of this study is the explanatory correlational research.Creswell (2014) defines explanatory correlational research design as the study that analyzes the relationship between two main variables or more, and to what extent the changes in one variable influence the changes of another variable.The relationship between the variables should be clearly explained by using appropriate statistical calculation.In addition, Borg and Gall (2003) explain that the correlational study mainly focuses on collecting data of two variables or more and finding the relationship or r value (correlation coefficient) of the variables.There are two main variables focused in this study, namely the level of self-efficacy and TOEFL performance.
The population in this study is the university students in Banda Aceh who are from Syiah Kuala University and Serambi Mekkah University.The sampling technique used is stratified sampling method, in which a group of students should firstly fulfill the criteria including: (1) a second-year student of Syiah Kuala University or Serambi Mekkah University (English and Non-English majors), and (2) has taken a subject of TOEFL in campus, such as TOEFL Preparation class (Creswell, 2014).The use of random group in stratified sampling technique eased the researchers to generalize the result of this study.The total sample participated in this study were 200 university students.This number has satisfied the correlational study requirement, where Field (2014) states that there should be a minimum of 30 participants in the correlational study.
There are two instruments used in this study.The first instrument is the questionnaire to measure the students' level of self-efficacy at the universities under 123 | Studies in English Language and Education, 6(1), 117-130, 2019 study.This questionnaire was designed by the researchers by following the guide of self-efficacy questionnaire designed by Bandura (2006), which is by using several statements related to the ability on performing certain tasks in TOEFL test and each statement starts with" I can…".
The number of 0-100 was then used to measure the self-efficacy of each statement.These statements have been analyzed and consulted with a colleague (expert judgment), particularly in terms of the language use and meaning to avoid misunderstanding, as well as whether or not it has covered all essential components in TOEFL self-efficacy.This questionnaire was then distributed to 30 students with the same characteristics of the targeted sample (but not the real sample) for the validity and reliability check, especially to investigate whether the students could easily understand the statements as intended and to find out the total time needed to complete the questionnaire.This process was done in accordance with the guide of designing a new research instrument as suggested by Lietz (2010).This questionnaire has two parts, where the first part focuses on questions related to personal data of participants.Meanwhile the second part focus on 30 statements about performing TOEFL; in this part, the participants were expected to circle the appropriate level of their self-efficacy on performing those tasks from the lowest of 0 to the highest of 100 with the allocated time of ±10 minutes.
In addition, to find out the university students' capability in TOEFL, the TOEFL test was used as the second instrument.The complete test was taken from the book entitled Longman Complete Course for the TOEFL Test: Preparation for the Computer and Paper Tests by Phillips (2008).This book is considered valid and reliable because it has been used as the main guide in the process of teaching and learning TOEFL in many places, particularly in Aceh.Additionally, the level of the items in the complete TOEFL test of these books has been adjusted to meet the standard TOEFL test.This test is a multiple-choice test (A, B, C, D) with the total items of 140, and the allocated time to complete the test is 110 minutes, with 50 items in the listening section (30 minutes), 40 items in the structure section (25 minutes), and 50 items in the reading section (55 minutes).The students needed to shade the answer on the provided answer sheet.
The collected data were further analyzed by using SPSS 17.0, that is by productmoment correlation or Pearson correlation (r) formula to find out the relationship between the two variables (self-efficacy and TOEFL performance), as follows: From the formula, rxy stands for the correlation coefficient between self-efficacy and TOEFL performance, N refers to the total sample, X means the score of selfefficacy level while Y means the TOEFL score, ∑X refers to the total score of selfefficacy level while ∑Y means the total score of TOEFL, and ∑X2 refers to the sum of squared scores from self-efficacy level while ∑Y2 means the sum of squared scores from TOEFL (Sugiyono, 2007, p. 228).
Furthermore, in order to investigate the significance level of the two variables, the researchers use the following guide Table 2. rxy value interpretation by Borg and Gall (2003, p. 320)  Finally, the t-test was used to test the hypothesis of the study is as the following: H0: There is no correlation between university students' self-efficacy and their TOEFL performance.H1: There is a significant correlation between university students' self-efficacy and their TOEFL performance.The t-test formula used is as suggested by Sugiyono (2007): From the formula, t is the t-test value between self-efficacy level and TOEFL performance, r refers to the correlation coefficient between self-efficacy and TOEFL performance, while n is the total sample.

The Level of University Students' Self-Efficacy
The students' level of self-efficacy varies (low, medium, and high) both in the English or non-English majors in both universities.Figure 1 depicts the students' selfefficacy level.

Figure 1.
The university students' level of self-efficacy on the TOEFL test.Figure 1 shows that most of the university students have a medium level of selfefficacy on doing the TOEFL test, which is around 154 respondents.However, the number of students who have high self-efficacy level is higher (about 35 respondents) compared to the number of those with a low level of self-efficacy (about 11 respondents or 5.5%).

The University Students' TOEFL Scores
In terms of the TOEFL score, it is found that the scores also varied, ranging from the lowest score of 303 to the highest score of 540. Figure 2 shows the university students' TOEFL capability in general.From Figure 2, it is obvious that the university students of Syiah Kuala and Serambi Mekkah who have low scores of TOEFL (below 400) are dominant (83 respondents).While those who achieve the medium scores (401-449) are around 67 respondents, and those who get high scores (> 450) are only 50 students, which is around 25% from the respondents.

The Correlation between Self-Efficacy and the University Students' TOEFL Score
In order to check the data distribution, the test of normality was conducted, with results shown in Table 3.It can be seen that the significant value of both variables, which is self-efficacy and the students' TOEFL capability, is .200.This value is bigger than the p-value of .05,which means that the data is normally distributed.Therefore, the product moment correlation or Pearson Correlation could be used to see the relationship between the two variables in this study, as shown in Table 4.   From the calculation of the Pearson Correlation, it is found that there is a relationship between the university students' level of self-efficacy and the students' TOEFL performance.The r of both variables is r=.249.In order to clearly see the relationship between the two variables (self-efficacy and the TOEFL capability), see Figure 3. Figure 3 indicates that both variables (self-efficacy and TOEFL performance) are related to each other, even though the relationship is not very close to each other.This means that the university students' self-efficacy do not completely influence the students TOEFL performance, but is still needed to be considered because the correlation is not very scattered as seen in Figure 3.
Furthermore, in order to find out the correlation strength of both variables, the following interpretation table by Borg and Gall (2003, p. 320) is used, such as shown in Table 5.The result of r=.249 indicates that the correlation between self-efficacy and the university students' TOEFL capability is categorized as low.

The Correlation between Self-Efficacy and the University Students' TOEFL Score
In order to test the formulated hypothesis, the t-test formula is used, with the result of t=3.616.The t-test of t=3.616 shows that the t-test is higher than t-table (1.972).It means that there is a significant relationship between the two variables tested, which is self-efficacy level and TOEFL capability.Therefore, in this study, the null hypothesis (H0) is rejected, while the alternative hypothesis (H1) is accepted.

DISCUSSION
In terms of the university students' level of self-efficacy in doing TOEFL test, it is found that the university students of Syiah Kuala University and Serambi Mekkah University averagely have a medium level of self-efficacy, which is 77%.Meanwhile, the university students with a high level of self-efficacy (those who have a strong selfbelief in completing the TOEFL test) is only 17.5%.However, Kitikanan and Sasimonton (2017) found that Thai students' self-efficacy is considered high.This should be further investigated of why Thai students have a higher TOEFL self-efficacy compared to Indonesian students.
Related to the university students' TOEFL performance in both universities, the highest percentage of the respondents have a low TOEFL score (below 400), which is 42%.While, the university students who can reach the target of 450 and above has the smallest proportion, which is only 25%.This shows that an intensive effort is needed, particularly to improve the university students' TOEFL score in Banda Aceh to be able to achieve the standard TOEFL score with a minimum of 450.This result is in line with the previous study by Zaitun (2014) who confirms that very few students (about 5% can achieve 450 in the TOEFL test.This suggests that there should be extra work for both test-takers in TOEFL as well as the TOEFL instructors to help the students achieve the targeted score of TOEFL. Furthermore, the r-value of this research result also shows a positive correlation between the university students' level of self-efficacy and their TOEFL performance.It means that the higher the self-efficacy of the university students, the higher the TOEFL score they can achieve.The result of this study has supported the previous studies by Ish and Rozali (2016) as well as a study by Kitikanan and Sasimonton (2017), in which those studies also revealed a positive correlation between the university students' level of self-efficacy and their TOEFL performance, but with different strength levels (r-value), where the medium correlation was found in the study by Kitikanan and Sasimonton (2017), r=0.60.While, the low correlation was found in the study by Ish and Rozali (2016), r=.347.In this study, the r value found (r=.249) is closer to what is found in the study by Ish and Rozali (2016).Hence, the result about the correlation implies that in fact, there is a positive correlation between the two variables (self-efficacy and TOEFL performance) and it does play a small contribution to improve the students' TOEFL performance.Therefore, self-efficacy should be considered to develop when learning and teaching TOEFL, for example by believing that when the students think they are able to do it, then they are able to do it.This is because thinking influences actions in life.

CONCLUSION
Based on the study about the correlation between self-efficacy and TOEFL performance, it could be concluded that 77% of the two universities' students (i.e.Syiah Kuala University and Serambi Mekkah University) have a medium level of selfefficacy, but only 25% from the students can achieve 450 above for the TOEFL scores.In addition, the result of the study revealed that there is a positive correlation between self-efficacy and TOEFL performance, but with a low correlation, r=.249, and there is a significant correlation between self-efficacy and TOEFL performance since the t-test (t=3.616) is higher than t-table (1.972).
Nonetheless, the limitation of this study lies in the difficulties of the standard use to self-evaluate when answering the questionnaire.This is proved when the researchers conducted the informal interview with the students with the low, medium, high selfefficacy, they tended to play safe by choosing the medium level, and some of them are indeed overestimated or underestimated by their self-efficacy.Therefore, further analysis of the same issue is still encouraged, particularly by enlarging the sample size in different education levels and in different areas in Indonesia to cover a wider context.The future results are also important to confirm the present study, especially to conclude to what extend self-efficacy contributes to English proficiency.Zaitun, Z. (2014).The relationship between students' self-perceived English proficiency and their performance in TOEFL.MORPH Journal of PBI FIP UMJ, 1(1), 11-26. Zulaikha, L. (2015).The correlation between the level of students' self-efficacy and English speaking skill at tenth grade of MA Bilingual Krian.(Bachelor), Sunan Ampel State Islamic University, Surabaya.
Figure 2. The TOEFL scores of the university students.

Figure 3 .
Figure 3.The relationship between the self-efficacy level and the university students' TOEFL capability.
state that there are several kinds of TOEFL that are conducted with different methods.They are explained as the following.1. PBT-TOEFL or Paper Based Test TOEFL Paper Based Test TOEFL is a test that is often conducted in Indonesia.The test, which is done by rounding or shading the correct answers in the answer sheet, aims to measure people's English capability in three sections, namely listening skill, structure, and reading skill.This kind of TOEFL test has a score ranging from a minimum score of 217 to a maximum score of 677.The allocated time for completing this test is about 2 hours.The PBT TOEFL, especially ITP/Institutional Testing Programme TOEFL, is generally conducted by language centers in various universities in Indonesia, particularly which have cooperation with ETS (Educational Testing Service) or by other private language centers which have been specifically appointed by IIEF (Indonesian International Education Foundation).

Table 4 .
Correlation of self-efficacy and the TOEFL score.

THE AUTHORS Marisa Yoestara
was born on January 16, 1989, in Banda Aceh.She completed her Master's degree as an Australia Awards Scholarship (AAS) recipient in the Humanities, Art, and Social Science Department of Flinders University South Australia, specializing in TESOL in 2015.She is now working as a lecturer at the English Department, Faculty of Teacher Training and Education, Universitas Serambi Mekkah, Aceh, Indonesia.Her research interests focus on innovative English teaching and learning process, including assessment to make effective learning.She is currently the secretary of the Quality Assurance Board of the English Department at Universitas Serambi Mekkah.Zaiyana Putri is a lecturer at the English Department of Teacher Training and Education Faculty of Universitas Serambi Mekkah, Aceh, Indonesia.Her research interests focus on teacher training, teachers' competence and professional development, especially for pre-service and in-service teachers.Currently, she is the head of the language laboratory of the English Department.Additionally, in 2017, she won the grant for a two-year research funded by DRPM-Kemenristekdikti in collaboration with Syiah Kuala University.Besides, she also works for the Acting Official Governor of Aceh as one of the technical assistance in supporting the Aceh Government's programs.