THE EFFECT OF SELF-EFFICACY ON ENGLISH TRAINING RESULTS

The aim of this study is to explore how much influence the self-efficacy of English training results in the form of the TOEFL score. This research used the quantitative approach, with multiple linear regression analysis. The sampling technique used a saturated sample in which the researcher uses 23 persons who are carrying out the TOEFL training. Data collection techniques using the questionnaire in the form of google form and then analyzed using SPSS 24. Data collection tool in the form of the validated questionnaire with all item coefficients greater than 0.3 (r = 0.316 — 0.885) and declared reliable with alpha value 0.945. The results of this study indicate that there is a significant influence simultaneously between self-efficacy with ability variables, the ability to organize, the ability to perform a task, the ability to achieve a goal and implement actions to achieve proficiency in training results in the form of TOEFL score. But partially no significant effect. Effect of self-efficacy with these variables has an effect on the training results in the form of TOEFL score of 64.5%.


A. INTRODUCTION
Training is one of the non-formal education programs, it is explained in Law Number 20 of 2003 concerning the national education system part 5 of article 26 paragraph 3(Kementerian Pendidikan dan Kebudayaan RI, 2003).The training has a strategic position in a program or activity.Research states that the training phase has an important influence on productivity.The results of the analysis show the training and development of employees of PT.Air Manado has a significant influence on employee work productivity by 74% (Kandou, 2013).Even so in the process of recruitment of experts in a particular field of work also requires training, then the term employee training arises.Training is seen as being so strategic that at present there are training institutions that are initiated by the community and the government.Training has a purpose (Mangkunegara & Anwar Prabu, 2006), among others: 1) increasing appreciation of soul and ideology, 2) increasing work productivity, 3) improving work quality, 4) improving human resource planning, 5) improving moral attitudes and work morale, 6) increasing stimulation so that employees are able to perform optimally, 7) improving health and safety, 8) avoiding uniformity,and 9) improving employee personal development.
Judging from the expertise to be mastered, there are various types of training institutions which are oriented to skills related to knowledge, skills and attitudes.Currently grow training instructions oriented to mastering foreign language, one of which is popular is mastery of English.This is not strange because the enthusiasm of the peoplr is so high in the need for English mastery.The second, English is an international language.People who have activities related to the international community certainly need English language mastery.Access to educational activities, business and social activities will be easier if people master the international language.Data from Education First stated that the mastery of English in Indonesian society is in the position of 39 of the 80 countries surveyed (Education First, 2017).Furthermore Education First stated that there was a strong relationship between the English proficiency of a country and the number of researchers in research and development per capita, as well as its investment in R & D. Researchers need English to share their findings and to keep abreast of the work of their international colleagues.Therefore, there is a strong correlation between English proficiency and income for the use of intellectual property.Innovation in countries with higher English proficiency tends not to be hampered by language barriers, and is more likely to be known, shared and licensed internationally.The low ability of English language certainly limits people's access to international activities.On the other hand foreign language learning programs in formal education have limitations, because learning hours must be shared with other subjects.The limited formal education institutions in meeting these needs are the potential that is utilized by non-formal education in meeting the learning needs of the community and its role as a supplement, complement and substitution of formal education (Dinisti, 2016).
In the world of education, literature and studies in the country, although the current trend is increasing, the need to look for reputable journals is still high, but for some scientific studies.Scopus indexed Indonesian International Scientific Publications as of October 2, 2017 was at 12,098 publications, this number increased significantly compared to the achievements of Indonesia's international scientific publications in August at 9,501 (Biro Kerjasama dan Komunikasi Publik -Kemenristekdikti, 2017).This requires education actors to access studies and literature that are more global in nature, which of course mastery of English is absolutely necessary.Campus located abroad requires foreign students to master the international language as evidenced by a certificate or ceremony before the applicants are passed to participate in learning activities.This is also required by the domestic campus as a condition for attending lectures, especially at postgraduate level.This is in line with the increasing quality of international publications that are being pushed by the government.Mastery of English in books, journals and scientific studies used in supporting learning is absolutely necessary.This is presumably seen by the sponsor of the postgraduate scholarship as the Indonesia Endowment Fund For Education (LPDP) which provides the prerequisite for mastering the international language.In improving the quality of scholarship recipients, the Indonesia Endowment Fund For Education (LPDP) provides training between 3 -6 months to scholarship recipients from the affirmation registrant pathway.This affirmation registrant consists of scholarship recipients from the poorest performing communities, from 3T (leading, outermost and disadvantaged), international achievers, and alumni of Bidik Misi scholarships.
The opportunity for scholarship recipients to obtain such training must be used wisely.Scholarship recipients who have the opportunity to get training must be able to challenge themselves to be able to organize, perform a task, achieve a goal, produce something and implement actions to achieve skills in the field of English language mastery.This will be examined in this study.Researchers want to explore how much influence self-efficacy has on the results of English language training.This research is important because the results will give an overview to scholarship sponsors, training institutions and scholarship recipients, how much self-efficacy influences how much the training results.This will later become a consideration for scholarship sponsors and training institutions to develop training programs that not only provide material but also condition scholarship recipients to always have strong self-motivation during the training period.

B. METHODS
This study used a quantitative approach, using multiple linear regression analysis using SPSS 24.The study was conducted at the Language Hall of the State University of Malang (UM).This is based on the Language Center of the State University of Malang (UM) in collaboration with the Indonesia Endowment Fund For Education (LPDP) to organize a language enrichment program for recipients of affirmation path scholarships.Data collection tool in the form of a questionnaire in the form of google form that is given to all scholarship recipients who are undergoing a language enrichment program for 6 months totaling 23 people.Researchers used saturated sample techniques, so researchers used the entire population as a research sample.Test the validity and reliability of the instrument using a group of scholarship recipients who are undergoing language enrichment programs elsewhere.Data collection tools in the form of questionnaires were declared valid with all item coefficients greater than 0,3 (r = 0,316 -0,885) and declared reliable with an alpha value of 0,945.

Normality Assumptions Test
Normality testuse the probability plot method, and images show residual points tend to spread between diagonal lines.Thus the residual is declared to spread normally.

Heteroscedasticity Assumption Test
Heteroscedasticitymeans there are variant variables in the regression model that is not constant (the same), on the contrary if it has a constant variant called homoskedasticity.
In order for the model to be said to be good, the regression model must meet homoskedasticity.There are several methods that can be used to detect residual normality, including scatter plot, Glejser, white, etc.In this context researchers use the scatter plot method.The image is known to state that the residual points spread randomly, thus the residuals are declared to be spread normally.Multicollinearity is the near perfect occurrence of linear correlation between two or more independent variables.Multicollinearity test is used to test whether in the regression model that is formed there is a perfect correlation or not on the independent variable.There are several methods that can be used to detect multicollinearity methods, including Tolerance, Variance Inflation Factor (VIF), Wise Correlation Pair, Partial Correlation, etc.In this context the researcher uses the Variance Inflation Factor (VIF) method, in which the analysis shows the value of Variance Inflation Factor (VIF) X1, X2, X3 and X5 smaller than 10, but the X4 variable values more than 10 so that multicollinearity occurs.To overcome this problem the researcher eliminated the X4 variable in data analysis so as not to occur multicollinearity.The adj square value (R2) is 0,645 (64,5%), meaning that the magnitude of the variable variability in training results is the Toefl (Y) value which can be explained by the variable ability to organize (X1), the ability to do a task (X2), the ability to achieve a goal (X3) and implement actions to achieve skills (X5) of 64,5%.This means that the contribution of variable ability to organize (X1), the ability to do a task (X2), the ability to achieve a goal (X3) and implement actions to achieve skills (X5) on the results of training in the form of TOEFL (Y) of 64,5%.While the remaining 35,5% is the contribution of other variables not included in the study.The value of multiple correlation (R) of 0,842 shows a very strong relationship between the variable contribution of the ability to organize (X1), the ability to do a task (X2), the ability to achieve a goal (X3) and implement actions to achieve skills (X5) as a whole with training results in the form of Toefl (Y).The calculation results above show the test statistic Fcount is 10.994 with a probability of 0,000 while the alpha value is 0,05 (5%).This shows that the probability value <alpha, so H0 is rejected.Thus it can be stated that there is a significant influence simultaneously variable ability to organize (X1), the ability to do a task (X2), the ability to achieve a goal (X3) and implement actions to achieve skills (X5) on the results of training in the form of TOEFL scores ( Y).The calculation above shows the value of t on the ability to organize variable (X1) is 0,524 with a probability of 0,606 while the alpha value is 0,05 (5%).This shows that the probability value> alpha, so H0 is accepted.Thus partially the ability to organize variable (X1) has no significant effect on the training results in the form of TOEFL (Y).

Model
The calculation above shows the value of t on the variable ability to do a task (X2) is 1.2292 with a probability of 0,213 while the alpha value is 0,05 (5%).This shows that the probability value> alpha, so H0 is accepted.Thus partially the ability to do a task variable (X2) has no significant effect on the training results in the form of TOEFL (Y).
The calculation above shows the value of t in the variable ability to achieve a goal (X3) is worth 0,933 with a probability of 0,363 while the alpha value is 0,05 (5%).This shows that the probability value> alpha, so H0 is accepted.Thus partially the ability to achieve a goal (X3) variable has no significant effect on the training results in the form of TOEFL (Y).
The above calculation shows the value of t on the variable implementing the action to achieve skills (X5) is 0,227 with a probability of 0,823 while the alpha value is 0,05 (5%).This shows that the probability value> alpha, so H0 is accepted.Thus partially the ability to achieve a goal (X3) variable has no significant effect on the training results in the form of TOEFL (Y).Y = 252,065 + 10,631 X1 + 27,531 X2 + 16,968 X3 + 3,303 X5 The constant of 252,065 states that if the ability to organize, the ability to do a task, the ability to achieve a goal, implementing actions to achieve skills are constant (unchanged) then the measurement of training results in the form of TOEFL (Y) of 252,065.The coefficient of the ability to organize is 10,631 indicating the ability to organize a positive effect on the training results in the form of TOEFL scores.This means that the higher the ability to organize tends to be able to improve training results in the form of TOEFL scores.
The ability to do a task coefficient of 27,531indicating the ability to do a task has a positive effect on the results of training in the form of TOEFL scores.This means that the higher the ability to do a task tends to be able to increase training results in the form of TOEFL scores.The coefficient of ability to achieve a goal is 16,968 indicating the ability to achieve a goal has a positive effect on the training results in the form of TOEFL scores.This means that the higher the ability to achieve a goal tends to be able to improve training results in the form of TOEFL scores.The coefficient implements actions to achieve skills of 3,303 indicating that implementing actions to achieve skills has a positive effect on training results in the form of TOEFL scores.This means that the higher the implementation of actions to achieve skills tends to be able to improve training results in the form of TOEFL scores.Dominant influence can be seen through the largest Coefficients standarized value.It can be seen from the calculation above that the variable that has the most influence is the ability to do a task.
In the results of the analysis in the previous sub-chapter, it is known that self-efficacy has a significant effect on the results of training simultaneously.This means that the higher the level of self-efficacy tends to be able to improve training results in the form of TOEFL scores.The results of this study are consistent with research from (Wahdania, Rahman, & Sulasteri, 2017) which states that self-efficacy has a significant effect on learning outcomes.And corresponds to research conducted by (Pertiwi, 2015) which states that 29.6% of student learning outcomes are influenced by self-efficacy.Selfefficacy has a very important role in relation to predicting learning achievement.Research conducted by Research (Lane, Lane, & Cockerton, 2003) on Postgraduate students supports the results of previous studies that there is a significant relationship between self-efficacy and learning achievement.

C. CONCLUSSION
The results of the analysis can be stated that there is significant influence simultaneously or jointly between self-efficacy variables for organizing ability, the ability to perform a task, the ability to achieve a goal and implement actions to achieve proficiency on the training results in the form of TOEFL value.But partially it has no significant effect.The influence of self-efficacy with these variables impact the results in the form of training TOEFL value by 64.5% while the remaining 35.5% is contributed by other variables not included in the study.
With these findings, researchers can provide advice to the management body of training in order to maintain and even increase self-efficacy during the training process, so that the participants always had the passion to constantly improve themselves.The next suggestion is to sponsor scholarship institutions to pay attention to the psychological state of language enrichment participants in order to have stable self-efficacy, so that the results of training can be maximized.

D. ACKNOWLEDGEMENT
The researcher expressed his appreciation and acknowledge to the Indonesia Endowment Fund For Education (LPDP) for providing postgraduate scholarship support at the State University of Malang.