The relevance of self-efficacy, perception, ICT ability, and teacher performance (Study on Islamic Teachers in Semarang, Indonesia)

The paper aimed to investigate the influence of teacher's self-efficacy and perception on his or her job to performance through ICT ability. The research is a quantitative description using path analysis methods. The population is 1500. They are the Islamic religious teachers in Semarang, Central Java, Indonesia The researcher used Isaac & Michael's table to determine the number of samples. Based on the table, the samples are 290. The Findings are the teachers’ selfefficacy and perception greatly affect to their performance, while perception is also being the critical factor influences teachers to use ICT in the classroom, but self-efficacy is not the vital thing for teachers ability to use ICT. The finding recommends that self-efficacy and perception of Islamic teachers are vital aspects of the quality of performance. The use of ICT in the classroom is the critical variable for the quality of teacher performance.


Introduction
This study aims to reveal the influence of self-efficacy and teacher's perception of ICT ability (information, communication, and technology) and teacher's performance. The base of discussion is the reality in the field of how ICT ability and teacher performance, primarily Islamic religious teachers in Central Java, are relatively low. The ICT ability of Islamic religious teachers is mostly below average. Their ability to use computers as learning aids in class is still low. The low ability of this field of computer use turns out to be in line with the ability to develop teaching methods, using media, and developing teaching strategies in the classroom. In reality, the majority of Islamic religious teachers teach in conventional ways. Very few of them teach using contextual learning methods and strategies. The problem is whether self-efficacy and perception of the teacher's profession influence the ICT ability and the performance of Islamic religious teachers?
Many education experts have studied the close relationship between self-efficacy and perception related to one's performance. Heuer & Keele 1 even specifically discussed the relationship between perceptions and attitudes and ways of working someone. The experts of psychology and education have studied the close relationship between self-efficacy, attitudes, and actions. Selfefficacy is a person's belief in his or her ability to carry out tasks.2 It indicates that self-efficacy and performance have a closeness relationship. An expert has discussed in depth the closeness of the relationship between self-efficacy and one's performance.3 Even Wentzel & Wigfield4 firmly stated, self-efficacy affects motivation, achievement, and self-regulation. In line with these opinions, Bandura & Jacobs5 have revealed that Self-efficacy has a strong influence on performance. Even Gaskil and Murphy6 found in their research that self-efficacy is the most decisive variable in a person's performance. Based on the theory, self-efficacy and perception determined the commitment of teachers to improve their ability in the field of ICT and its performance. How to increase teacher self-efficacy for his profession is an important issue. In the research, Karel Kreijns7

| Ikhrom
found that the creation of a conducive working atmosphere is an essential variable in encouraging teachers to use ICT in classroom learning. The self-efficacy of a good teacher for his profession can encourage him to learn a lot of ICT and how to teach in the classroom. While the proficiency of teachers using ICT is directly proportional to the quality of learning in the classroom.8 To analyze how the relationship patterns of the four variables, namely self-efficacy or/and teacher perceptions of ICT abilities and performance quality, were analyzed using path analysis. The researcher used Path analysis to find out the direct and indirect effects of variable X on variable Y. The two questions are being the basis of all analyses are (1) do teacher self-efficacy and perceptions simultaneous or partial, both directly and indirectly influence ICT ability? And (2) do teacher self-efficacy and perceptions simultaneous or partial, both directly and indirectly influence performance through ICT ability?

Performance is the Key Term.
The quality performance of the teacher largely determines the quality of education. The quality of its performance determines teacher quality. Performance means accomplishment, execution, carrying out, working out of anything ordered, or carried out.9 Performance is about doing meaningful work in effective and Armstrong's statement that performance is a function of capabilities as well as output (the achievement of objectives).11 Bandura 12 insists that performance is causally before outcomes. Bandura's quote indicates that there is no performance except an impact on achievement. Therefore, these two things are not enough to make someone's performance right. Excellent performance requires an environment, conditions, perceptions, and also the right motivation.
Galton and Simon13 say, performance is the result of the interaction of elements of motivation (m), ability (k), and perception (p) in a person, so that formula of the performance is P=(mxk+p). The meaning is, the more positive a person's perception of his job, coupled with his ability to relate to the job, and the motivation of his work is also good, the better the performance will be. The weakness of one of these elements will disrupt the performance.
The teacher's performance is a behavior produced by a teacher in carrying out her or his duties of teaching in the class. Teacher performance is a teachers' activity in carrying out their primary tasks.14 The teacher's performance is daily working conditions and conditions. Teacher performance is closely related to the obligations of the teacher by the main activities, namely planning to learn, implementing learning, assessing learning outcomes, guiding and training students, and carrying out additional tasks inherent in the implementation of the main tasks of the teacher. Thus, teacher performance is a form of implementation of the main tasks and functions of the teacher as a whole.
To find out the quality of teacher performance, Armstrong 15 calls it, 'productive performance,' so we must know how to measure it. To measure teacher performance, people must know the dimensions of teacher performance. According to Michael D Jones,16 teacher performance consists of five dimensions: (1) the quality of work and results of work, (2) speed and accuracy of work, (3) initiative in work, (4) ability to work, and (5) communication in work. The dimensions of work quality and work outcomes consist of several indicators: planning, mastery of teaching materials, management of the learning process, classroom management, and student achievement results. The dimensions of speed and accuracy of work consist of the use of media or learning resources, mastery of educational foundations, and planning of learning programs. The dimensions of initiative in work consist of class leadership, management of learning interactions, and implementation of learning outcomes assessments. The dimensions of workability consist of the application of variations in learning methods, and the implementation of remediation and enrichment.
The dimensions of communication include understanding and administering school administration and understanding and interpreting the results of research to improve the quality of learning.
The conclusion is many factors determine teacher performance. The quality of the surrounding factors determines the quality of teacher performance. These factors include competence, knowledge, attitude, confidence, motivation, and excellent communication skills. The latter is even the critical one because the quality of a teacher's performance is the quality of their communication.17 Self efficacy is a crucial variable to determine teachers' ICT ability and performance. Self-efficacy, not self-image, self-worth, or any other similar construct. Self-efficacy is the judgment of personal capability.18 It means self-efficacy refers to the belief of one's abilities, specifically the ability to meet the challenges and complete a task successfully. Self-efficacy is not just a person's belief in one's ability to carry out tasks that are on his shoulders. However, selfefficacy is a judgment of someone on his ability to plan and carry out, which leads to a particular goal.19 Self-efficacy is a person's belief that he can learn and do activities or work according to the desired level. Self-efficacy affects one's thoughts, feelings, and actions. Every person has a self-efficacy. Some have a high level of self-efficacy, and those who have a low level of self-efficacy.
Thus, self-efficacy is not related to the skills one has, but rather about judgment on what person can do things with the skills possessed. Therefore, self-efficacy also helps determine how much effort a person will spend on an activity, and how long he will pursue his work even though there are many difficulties and challenges. People who have high self-efficacy will be calmer in facing difficulties and strikes in his work. He will continue to struggle to do the work to achieve success. While people who have low self-efficacy, they will effortlessly stop their work when they find many difficulties and obstacles. In short, self-efficacy is the belief that he or she can succeed.
The application of self-efficacy in academic studies is undoubtedly beneficial for improving the quality of education. Bandura20 states, "Perceived academic self-efficacy means personal judgments of one's capabilities to organize and execute courses of action to attain designated types of educational performances." The belief in self-efficacy is a crucial factor in the source of human action, what people think, believe, and feel affects how they act.21 Efficacy of beliefs influencing how people think, feel, motivate themselves, and act.22 Thus, self-efficacy can be as a control variable for commitment and teacher performance. Low or high self-efficacy can be as a control tool to determine the quality of a teacher's performance. Teachers who have low self-efficacy can indicate to low-performance quality. Conversely, teachers with high self-efficacy can be sure the quality of performance is outstanding. Gaskil & Murphy's research findings conclude that 20 Bandura, 202. 21 Bandura, Self-Efficacy: The Exercise of Control, 22. 22 Bandura, Self-Efficacy in Changing Societies, 3. self-efficacy is the most dominant variable in determining teacher performance.23 Self-efficacy is an essential aspect of influencing student learning24 and teacher performance. Many studies reveal the urgency of self-efficacy on teacher performance quality,25 teacher creativity,26 commitment teacher to his profession.27 23 Gaskill and Karen Murphy, "Effects of a Memory Strategy on Second-Graders' Performance and Self-Efficacy." 24 Anubha Rohatgi, Ronny Scherer, and Ove E Hatlevik, "The Role of ICT Self-Efficacy for Students' ICT Use and Their Achievement in a Computer and Information Literacy Test," Computers & Education 102 (November 2016): 103-16, https://doi.org/10.1016/j.compedu.2016.08.001. 25 Honicke and Broadbent, "The Influence of Academic Self-Efficacy on Academi"; Chang et al., "Improving Creative Self-Efficacy and Performance through Computer-Aided Design Application"; Matthew L Bernacki, Timothy J Nokes-Malach, and Vincent Aleven, "Examining Self-Efficacy during Learning: Variability and Relations to Behavior, Performance, and Learning," Many studies on the relationship of self-efficacy with teachers' ICT ability and performance. The teacher's mastering on ICT in the learning process determines the quality of achievement of the students in his class. Thus the findings of the study entitled "Is the way they use it?: Teacher, ICT, and Achievement".28 Research on ICT use and student achievement was also carried out by Cunska and Savicka29 with the title "Use of ICT Teaching-Learning Methods for making School Math Blossom". Although the focus of the study does not directly link ICT skills to achievement, the ultimate goal of the research is ICT empowerment in the learning process can make the classroom atmosphere pleasant and refreshing. Difficult mathematical material can be easily understood by students, thanks to using ICT.
Next, how is the relationship between self-efficacy and teacher performance? The definition of self-efficacy has clearly stated, self-efficacy is the belief in our ability to succeed. People who believe that they can succeed have the readiness to face all obstacles and obstacles in carrying out their profession. He even has a sense of crisis that success in work requires sacrifice. People like this certainly do not give up easily when obstacles and difficulties arise. Theoretically, the relationship between the effect of self-efficacy and performance is powerful. Besides, research has also proven the strong influence of self-efficacy on performance. Research of Gaskil dan Murphy30 concluded that self-efficacy is the dominant variable for performance. A similar conclusion is also 28 30 Gaskill and Karen Murphy, "Effects of a Memory Strategy on Second-Graders' Performance and Self-Efficacy." coming from other research with a focus on 'the relationship of selfefficacy and performance'.31 Perception is a vital thing for the teacher.
Perception is a psychological factor that has essential roles in influencing individuals both in behaving and acting. Perception is not merely a reaction to outside stimulation but a process of creation.32 The definition of this perception changes the current view of the meaning of perception. Until now, perceptions are only reactions to external stimulation. This understanding provides an understanding that perception is a process of creation. Niedenthal33 defines that perception is the transformation of sensations caused by impinging stimulus into an internal representation of that stimulus.
Perception consists of two parts. The basis of dividing is the quality of attitude (affective qualities).34 First, perception in the 31 Honicke and Broadbent, "The Influence of Academic Self-Efficacy on Academi"; Chesnut and Burley, "Self-Efficacy as a Predictor of Commitment to The"; Krista Althauser, "Job-Embedded Professional Development: Its Impact on Teacher Self-Efficacy and Student Performance," form of attitude (perception becomes effective). Second, creative perception. The first one is only talking about attitude,35 but the second is continued to create something. The Good' theory of perception is the kind of perception that impact on individual behavior. This meaning is in line with Rookes & Willson's definition.36 Perception is a process that makes people choose, organize, and interpret stimuli received into a meaningful and complete picture of the world. This kind of perception is the object of study by many researchers. The influence of teacher perceptions proved to have a positive impact on performance. Even based on the finding of the research, the teacher's positive perception can improve his or her performance.37 They recommend, improving the school can not deny the teacher's perception. Perception will help them and finally make decisions to behave and act. The theory states that perception is part of the information processing system. It is a complex phenomenon that provides higher-order processes such as creative choices.38 A teacher who has a positive perception and thoughts of his work, his or her feelings will be calm because he or she enjoys the teaching profession. Her or his behavior is steady and positive. The appearance is also convincing. Some theories support the relationship between teacher perceptions and performance. research finding of Campbell et al.39 has proven that perception affects performance. The four studies recommended that perceptions significantly influence teacher performance.
The following are some studies that recommend that perceptions significantly influence the ability of a teacher to use ICT in the learning process. First, the research conducted by Bas, Kubiatko, and Sunbul40, the results stated that the teacher perceptions are very significant towards the enthusiasm of teachers in using ICT in the teaching-learning process. Second, research of Mura and Diamantini41 with the same focus also found the conclusions that in line with the first study, the teachers' positive perception of the teaching profession positively affected the use of ICT in the teaching-learning process. Third, this study conducted by Kopcha42 resulted that ICT integration in the teaching and learning process was only able to be carried out by teachers who had a positive perception of their profession.

Teacher self-efficacy and perception are dual power for teachers' empowerment.
Combining the belief in one's capabilities and perception will be the productive power for teachers' performance. Miller et al.43;Kung44 dan Chong et al.45 has proven through their research that self-efficacy and perception simultaneously form an effective teacher in carrying out their duties. Even with these twin power, their teaching makes students eager to learn until their achievements increase. However, no much research on the relationship between self-efficacy and perceptions with ICT abilities and teacher performance. Research on self-efficacy and perception is found together, but not much. Miller  The explanation confirms that the relationship between selfefficacy, perception, ICT ability, and teacher performance had theoretical support and empirical data. It is thus significant for the verificative research design. Merten46 insisted the use of path analysis is to test a causal theoretical model. Displaying the previous research findings are mostly needed.

Research Method
This research is an ex post facto type of research. Ex post facto means after the fact.47 This study traces events that have occurred based on respondents' responses, with a quantitative approach. Data was collected using a questionnaire instrument distributed to respondents. Determining respondents through sampling techniques. The sampling technique uses random sampling. The research population was all Islamic religious teachers in Semarang City that come from elementary school, junior high school, and senior high school. From the data obtained from the Head of the Ministry of Religion in Semarang, there were 1500 Islamic religious teachers. The size of the study sample was determined using the Isaac & Michael table. Based on the table, the sample size is obtained with a 5% error rate at the 95% confidence level to a population of 290.48 Furthermore, to determine who the respondents were 300 people, the researcher decided to divide equally according to the level of education, so that 100 Islamic religious teachers represented each level of education. 100 from 46  elementary school, 100 from junior high school, and 100 from senior high school.
The collected data is analyzed using path analysis. The use of path analysis is to test a causal theoretical model. Path analysis is not a method used to find causes but to find an explanation of the patterns of direct and indirect relations of a causal model. The basis of cause and effect relationship pattern is theoretical considerations and the knowledge of researchers.49 Path analysis to analyze the pattern of causal relationships between several exogenous variables towards endogenous variables both directly and indirectly, simultaneously or partially. The path analysis model is only suitable for data that meets the assumptions that apply to regression analysis. The assumptions are including the normality, linearity, and homogeneity. As Teo50 stated, "the assumptions of linearity, normality of errors, and homogeneity are generally assessed by graphical means, but more formal assessments can also be made." Path analysis is to analyze the pattern of causal relationships between several exogenous variables towards endogenous variables both directly and indirectly, simultaneously or partially. The calculation of statistics is performed using SPSS 21. Based on the theoretical framework, that the carrying out path analysis through two steps, namely: (1) do teacher self-efficacy, perception simultaneous or partial influence ICT ability?; (2) do teacher selfefficacy and perceptions simultaneous or partial both directly and indirectly influence performance through ICT ability? Furthermore, the discussion of the results of the study was carried out by confirming field data with theoretical studies as well as theoretical frameworks. The next step is to decompose the correlation to calculate the value of the combined effect, partial effect, direct and indirect influence, total influence, the influence of other factors. Hypothesis testing is by examining the limits of acceptance-rejection of the level of statistical significance and the result of the path coefficient.51

Findings and Discussions
First, path analysis is a stepwise regression analysis that test analysis requirements and is often called assumption testing. Cohen52 emphasized that the path analysis model can be only applied to data meets the classical assumptions include a test of linearity, normality, and homogeneity.
The linearity test is aimed to determine whether two variables have a relationship that is linear or not significantly. If there are two variables X and Y are not linear, then it is not possible to make a prediction. From the results of testing the linearity of X1, X2, X3 on the Y variable obtained data, that all data X1, X2, and X3 against Y are linear. The linearity test is by ANOVA table. The criterium is if the sig scores lower than 0,05, so the data are linear.
From the table shows that the significant score in linearity in the three tables is 0,000. The significance is less than 0.05, and the conclusion that there is a linear relationship between the selfefficacy variable, perception, ICT ability, and performance.
The second assumption test is the normality test. Data normality test is used to determine the standard frequency distribution of research data. If the distribution of data is not standard, the results of the statistical analysis are less able to describe the characteristics of the population.
The criterium, distribution of data standard if Asymp. Sig. > than 0.05. From the table above, it shows that the Asymp. Sig score of the four variables is smaller than 0.05, which means that the data distribution of the population of self-efficacy, perception, ICT, and performance data is normal.
The homogeneity test is to evaluate whether each residual error for the independent variables is known to have the same variance. If the variance is different, it will cause the linear regression equation produced is no longer useful to make a prediction. The homogeneity test is carried out using a histogram, as illustrated below.
Picture 1: Histogram of homogeneity The interpretation of the histogram in the picture above that the variant data from the population is homogeneous. Data shows that the distribution of data forms a standard curve. That explains the research data has met the requirements of homogeneity. Then research can be carried out because homogeneity is one of the requirements for a path analysis model. Second, after checking the assumption, the discussion continued with an examination of the path analysis of Sub Structure 1: do teacher self-efficacy and perceptions simultaneous or partial, both directly and indirectly influence ICT ability?
The basis of discussion of substructure one is the results of the analysis using the SPSS method as below.
In the summary model table, the coefficient of determination (Adjusted R2) is 0.189. The use of score is the basis for determining the residual coefficient score of the sub-structure path analysis 1. The formula for the residual coefficient is √1-R 2 . The results obtained are √1 -0.126 = 0, 935 (93.5%). This score is included in the multiple regression equation analysis of the sub-structure path I (X3 = p31 X1 + p32 X2 + p3.e1). After filling out the formula, the results are (X3 = 0.067 + 0.337 + 0.935). That is, self-efficacy influences ICT ability variable (0.067), perceptions (0.142), and several residual variables not included in the research model of 0.935, or 93.5% influenced the rest, while the table shows that F  Table) has a score of 21,851 with the significance level of 0,000 ≥ 0.05. The meaning, self-efficacy, and perception together significantly influence ICT ability, while partially selfefficacy does not significantly influence ICT, while perceptions have a significant effect.
Then to be easily understood, the above explanation can be summarized in the form of substructure one coefficient Nonsignificanc e Percepti on P32 0,337 0,00 0 Significan ce Third, the discussion continued with an examination of the path analysis of Sub Structure 2: do teacher self-efficacy and perceptions simultaneous or partial, both directly and indirectly influence performance through ICT ability?
The essential discussion of substructure two is the results of the analysis using the SPSS method as below.

| Ikhrom
The table of the model summary shows that the determination coefficient score is 0,394. The score is basic in determining the residual coefficient score of path analysis of substructure two. the key of the residual coefficient is √1 − R 2 and the result is √1 − 0,394 = 0,778 (78,8%). The score will be the basis of the multiple regression equation of path analysis of substructure two (Y = py1 X1+ py2 X2 + py3 X3 + py.e2). After filling out the formula, the results are (Y = 0,298 + 0,468 + 0,009 + 0,778). That is, selfefficacy (0.298), perceptions (0.468), ICT ability (0,009) influenced the variation in performance variable and several residual variables not included in the research model of 0.778, or 78.8% influenced the rest, while F table (Anova Table) has a score of 63.731 at the significance level of 0,000 ≥ 0.05. The meaning, self-efficacy, perception, and ICT ability together significantly influence performance, while partially ICT ability does not significantly influence performance, while self-efficacy and perceptions have a significant effect.
Then to be easily understood, the above explanation can be summarized in the form of substructure two coefficient tables as follows.

Independent Variable
Symbol Coefficient β

Sig. Status
Self-Efficacy Py1 0,298 0,000 Significant Perception Py2 0,468 0,000 Significant ICT Ability Py3 0,009 0,857 Nonsignificant The explanation of path analysis from sub-structures one and two, which contain the path coefficients above, will be the basis for discussing the research findings, and the formula of substructure one and two as follow.
To find out the shape of the causal relationship of exogenous variables to the endogenous variable as a whole, a summary of the coefficient scores of the analysis of sub-structure one and substructure two are in the following Picture 3: Empirical Revised Model and Beta (β) Value, some changes found. First, deleting the arrow from X1 to X3, because it is not significant. Second, deleting the arrow from X3 to Y because it is also not significant.
Then is the process of decomposition of correlations between exogenous variables with endogenous variables. Decomposition aims to find the magnitude of the coefficient on the pattern of direct influence (DE) and indirect influence (ID). The calculation of DE and ID quantities involve exogenous variables on endogenous variables. Decomposition is only done at a significant path coefficientremoval of arrows in the revised model above for the benefit of the correlation decomposition process.
First, the pattern of the relationship between the variable selfefficacy (X1) on the performance variable (Y) = py1 X1 + py3X3 + e3. DE = py1 is significant (0.30), while IE = not significant. From the results of the decomposition, it shows that the pattern of the relationship between exogenous self-efficacy (X1) and endogenous performance variables (Y) does not have an indirect relationship pattern, but has a significant direct relationship pattern, DE = py1 (0.30).
Second, the pattern of relations between perception variables (X2) with performance variables (Y). Y = py2 X2 + py3X3 + e4. DE = significant (0.47), IE is also significant (0.34). From the results of the decomposition, it shows that the relationship between the exogenous variables of perception (X2) and endogenous performance variables (Y), has a direct relationship pattern, also has a significant indirect relationship pattern, namely IE through X3 = py3 p32 (0.47. 0, 34) = 0.16.
Third, the pattern of relations between ICT variable and performance variable (Y). with the equation, Y = py3X3 + e5.
Obtained DE = not significant. Based on the results of the decomposition above, it shows that exogenous variables, ICT ability (X3) do not have an indirect relationship pattern, nor a significant direct relationship pattern with performance.
Correlation decomposition above generates that the pattern of direct and indirect effects of exogenous variables is on the performance variables, which are briefly formulated in the table as follows.
Then how much is the useful contribution of each X variable to the variable Y? For searching for the useful contribution of each variable X is by multiplying the path coefficient (p) with the correlation coefficient (r). The path coefficient score (p) is obtained from the table of decomposition of the causal relationship to the performance variable (Y). while the correlation coefficient score is obtained through the correlation matrix as below.  Table: Effective contribution to performance The table shows that self-efficacy has an effective contribution to the direct effect on performance by 14%. Self-efficacy has no contribution from the indirect effect on performance. Perception has an effective contribution from direct influence on the performance of 27% and also has an effective contribution from indirect effects of 34%. Based on the results of effective contribution, as found in the table above, the conclusion is that 75% of the variation in teacher performance (Y) can be predicted through a variety of variables X. The details are 14% self-efficacy, 27% through perception variable, 34% through ICT ability variable, while the remaining 25% cannot be explained through the three exogenous variables. Other variables may explain some of the remaining effective contributions outside of the three variables, and others are variations due to measurement errors. Overall, the remaining 25% is called an error variable or residual or unexplained variance.
For proving the hypothesis formulated in this study is by explaining as follows. (1) The causal relationship between selfefficacy and teacher performance. Self-efficacy does not have a pattern of indirect influence with teacher performance, but has a significant direct influence pattern (py1= 0.30) and has an effective contribution of 0.14%. That means that the variance of the performance of Islamic religious teachers that can be explained by the variable self-efficacy is only 14%. Although having a small effective contribution, but this contribution means that selfefficacy is an important thing in improving the teacher's performance.
(2) The causal relationship between perceptions and teacher performance variables. Perception has a pattern of direct influence with teacher performance (py2 X2 = 47%), and also has a significant pattern of indirect influence through ICT ability variable (py2 p22 = 0.34%) with an effective contribution of 0.61 (61%). Based on these findings, the conclusion is that the perception variable turns out that the influence is stronger than selfefficacy. Perception has a direct influence on teacher performance, and can also influence ICT ability. This finding confirms that perception has a function on increasing the performance of Islamic religious teachers and can also be a predictor variable for the ICT ability.
Examining the results of the research shows some differences between the literature review and theoretical review with the empirical data. The theory said that self-efficacy is the influential variable for the teacher's performance, but the fact shows the different finding that self-efficacy gives a small contribution to performance.

Conclusion and Recommendation
Based on the discussion above, the conclusions are selfefficacy and perception have proven as the couple variables for performance. Besides, the perception has determined the teachers using ICT in the classroom. In turn, the quality dan effectivity of improving the teacher's performance can be supported through enhancing self-efficacy and perception of the teachers. The finding of the research recommends that self-efficacy and perception of Islamic teachers are vital aspects in determining the quality of performance. The performance teacher's quality can be shown by using ICT in the classroom. Future research may wish to research the relationship of self-efficacy and perception to the performance of the Islamic teachers by using the qualitative design or mixed research design.