Implementation of Performance Accountability System For Government Institution ( SAKIP ) : Determinants and Consequence In Local Government I

Info Articles ____________________ History Articles: Received 1 January 2018 Accepted 15 March 2018 Published 8 July 2018 ____________________


INTRODUCTION
Differences of Characteristic between public sector organizations and private sector have an impact on different performance measurement systems as well as between the two organizations. Private sector organizations that tend to assess performance from a financial perspective different with the public sector, especially the government that pure non-profit oriented on public service. To this day, many government agencies, especially local governments, focus solely on achieving financial performance and seeking "Reasonable" opinions (Sofyani and Akbar, 2015). This is certainly a dilemma because according mandate of Government Regulation No. 8 of 2006 on Financial Reporting and Performance of Government Institution beside the financial perspective, performance accountability is prioritized from a non-financial perspective that is the performance achievement of government programs through the implementation of Performance Accountability System for Government Institution (SAKIP) regulated in Presidential Regulation No. 29 of 2014.
The performance of local governments in Indonesia through the implementation of SAKIP is still low, indicated through the results of the appraisal of the Ministry of State Apparatus and Bureaucracy Reform in 2016 which found that there are only 3 provinces from 34 provinces that have received "A" with satisfactory category, only 1%, while 99% still get the predicate below it. Of the 34 provinces, the province of West Nusa Tenggara received the predicate "B" after eight consecutive years earned the predicate "CC". Despite the increase but the increase is not too significant only by 1.99% from 58.65% to 60.64% from the previous year. Of course this condition requires a more optimal effort again so it is expected for the next year performance achievement of this program can increase even higher. To be able to follow 3 other provinces that have earned the predicate "A", the minimum required increase of 19.46% because the range of values to predicate "A" < 80-90. The result of SAKIP Evaluation in Province of West Nusa Tenggara for six years from 2011 to 2016 can be seen in Table 1. The performance measurement system of government institutioan in Indonesia was first regulated through Presidential Instruction No. 7 of 1999 and after almost 14 years carried out finally replaced with the Presidential Regulation No. 29 of 2014. But until now, the existing system has not been able to give expextacion desired. In contrast to the accountability of financial performance has been good, this is indicated by the number of local governments that obtain "Unqualified Exception (WTP)" opinion which is the highest predicate.
Similarly, for the province of West Nusa Tenggara that has got the WTP opinion for 6 consecutive years.
Based on the perspective of institutional isomorphisms in institutional theory, there are three forms of isomorphism that are mimetic isomorphism or imitation effort, coercive isomorphism as a form of pseudo adherence to applicable rules and normative isomorphism that emphasizes professional organizational work mechanisms. Of the three isomorphisms, performance measurement system through implementation of SAKIP is expected to reach normative isomorphism level which emphasizes on professionalism in organizational management so that touching aspect of substance that is different with the other two isomorphism which is the form of mere symbol and administrative formalities as not touching aspect of substance.
Institutional theory emphasize the legitimacy of an organization, in which case government organizations should be able to give confidence to the society as the main stakeholder of governance so that they believe on government performance that will ultimately support the existence of the government organization itself. Therefore, the implementation of SAKIP at the normative level is expected because the performance achievement of the government programs in the form of services either stuff or service to the society can be realized properly.
The provincial government of West Nusa Tenggara has made concrete efforts to follow up the implementation of SAKIP wich still low among others : first, undertaking training to improve the capacity of human resources and the quality of SAKIP; second, conducted a comparative study to the city of Bandung in Province of West Java which has applied electronic SAKIP (e-SAKIP) based on information technology and third, the government emphasized to all regional apparatus organizations to implement of SAKIP in accordance with the regulations which existing such as determination of strategic planning, strategic porgam arrangement, target to be achieved and appropriate budget allocation.
Understanding the concept of public sector performance can be done with two approaches from the perspective of bureaucracy and the perspective of the target group. Specifically from a bureaucratic perspective, public sector performance can be seen in terms of productivity, accountability and transparency. The implementation of SAKIP resulting in the accountability report (LAKIP) is a form of government accountability and transparency over government performance productivity. Rahmanurrasjid (2008) said that the implementation of the principles of accountability and transparency in the implementation of local government requires the government to provide accountability and information to the public related to government management so that the government strives to provide the best (best performance) to the society.
The performance of government institution is inseparable from the role of manager in managing the organization through managerial functions. Performance measurement has a significant role in managerial or internal controls to ensure the organization is managed with the best attention of all stakeholders. Performance measurement is needed to generate information that can help manager to monitor the achievement of a program.
Individual characteristics such as self efficacy have an influence on the relationship between goal setting and individual performance (Robbin, 1998). Based on that opinion it can be illustrated that the person in this case the managers who have a high self efficacy will try to achieve the objectives set out in the planning document that are part of the SAKIP and will ultimately be able to improve organizational performance. Similarly, intellect (in some literature known as "conscientiousness") is an individual characteristic that directs its actions to normative behavior patterns, eg using performance information to make decision, revise program and evaluate program in order to improve the performance of government institution (Korzaan and Boswell , 2008).
Utilization of information technology has a significant effect on accountability of government performance (Putri, 2015). Likewise Razi (2017) found that utilization of information technology has effect on accountability of government performance while Rofika (2014) found different result. Cavalluzzo and Ittner (2004) and Sofyani and Akbar (2013) found that training has a significant effect on the implementation of SAKIP while Julianri (2017) found training has no effect on the implementation of SAKIP. Zirman (2010), Rofika (2014), Zulharman (2015) and Razi (2017) found that obedience to regulation has a significant effect on government performance accountability while Putri (2015) found opposite result. Guadagno et al. (2007) and Sofyani and Akbar (2013) found that self efficacy and conscientiousness have a significant effect on the implementation of SAKIP.
This research is a development of previous research, where the previous research tested separately between the utilization of information technology and obedience to regulation on government performance accountability by Putri (2015) and Razi (2017) while training and individual characteristics to implementation of SAKIP by Cavaluzzo and Ittner (2004) ) and Sofyani and Akbar (2013). The novelty in this study are first, to test the technical and organizational factors of the previous research which is separate as the determinant of the implementation of SAKIP; second, in this research want to test the effect implementation of SAKIP to the managerial performance and third, this research also put the individual characteristic factor as a stimulant in the relationship between Implementation of SAKIP and managerial performance that is different with previous research (Sofyani and Akbar, 2013) that placed individual characteristics as determinants of implementation of SAKIP.
Based on the above explanation, the purpose of this study is to test and obtain empirical evidence on the effect of utilization of information technology, training, obedience to regulation on the implementation of SAKIP and the effect of implementation of SAKIP on managerial performance. Furthermore want to test the role of self efficacy and intellect in strengthen the relationship between implementation of SAKIP and managerial performance.

Institutional Theory
Institutional theory is based on the idea that in order to survive, the organization must convince the public or society that the organization is a legitimate entity and deserves to be supported (Meyer and Rowan, 1977). Institutional theory argue that organizations that favor legitimacy will have a tendency to try to adapt to external expectations or social expectations (DiMaggio and Powell 1983) where the organization resides. Adjustment to external expectations or social expectations leads to an organizational tendency to separate their internal activities and focus on systems that are symbolic to external parties (Meyer and Rowan, 1977). Public organizations that tend to gain legitimacy will tend to have similarity or isomorphism with other public organizations (DiMaggio andPowell, 1983 in Sofyani andAkbar, 2013). There are three forms of similarity or isomorphism in institutional theory : first, mimetic isomophism or imitative attempts caused by lack of understanding of organizational technology and ambiguous goal; second, coercive isomorphism or pseudo compliance with existing regulatory demands and third, normative isomorphism with regard to professionalism which is a manifestation of the collective struggle of the members of the organization to determine the conditions and methods of their work, to control "production" and to develop cognition and legitimize their work autonomy (Collins 1979 andAkbar, 2013). In this case, the implementation of SAKIP is expected to reach the normative level or normative isomorphism that emphasizes on professionalism so as to touch the aspect of the substance not just the formality.

Managerial Performance
According to Government Regulation No. 8 of 2006 performance is the output / result of activities / program that would or have been achieved in connection with the use of budgets with quantity and quality measurable. Performance is one important factor that used in measuring the effectiveness and efficiency of an organization. Organizational performance can not be separated from the role or performance of managers in managing an organization to realize the vision, mission and objective of the organization.
Managerial performance is the performance of individuals in managerial functions such as : planning, investigation, coordination, evaluation, supervision, staffing, negotiation, representation (Mahoney et al., 1963 in Murtanto andWinda, 2006). The success of an organization in achieving its goal and fulfilling its social responsibilities depend largely on manager. If the manager is able to perform their duties properly, then the organization will be able to achieve the desired objective and target.

Implementation of SAKIP
Performance measurement as an important part of performance based management especially in local government is very important. According Mardiasmo (2002) argued that the measurement of performance has many goals, in addition to improving the performance of government agencies as well as a form of local government accountability. Further explained, for that the government is required to be able to build good performance and the performance is seted can not only use one size because the complexity of government services require different performance measures for different purposes. Therefore, a comprehensive performance measurement system is needed in accordance with the complexity of government tasks and responsibilities through the implementation of programs and activities to provide the best service to the society.
A comprehensive performance measurement system is designed to provide long-term benefit (sustainable). The first performance measurement system is regulated in Presidential Instruction No. 7 of 1999 on Performance Accountability of Government Institution and currently has been replaced by Presidential Regulation No. 29 of 2014 on Performance Accountability System of Government Institution. Mahsun (2006: 38) stated that the performance measurement system is a system that aims to help public managers assess the achievement of a strategy through financial and nonfinancial measures.
The current condition associated with the implementation of SAKIP as a performance measurement system for program achievements and government activities is still very low. When referring to the mandate in Minister of State Apparatus Utilization and Bureaucratic Reform No. 12 of 2015 about Evaluation Guidelines for Implementation of SAKIP there is no local government that can achieve the highest predicate that is "AA" (Very Satisfactory). Based on Government Regulation No. 8 of 2006 about Financial Reporting and Performance of Government Institutions and see the characteristics of government which public service oriented should be the main priority of government is the acievement of program and activity performance through the implementation of SAKIP oriented on result. It means what is the local government generates by using of budget in the form of services both stuff and service and how much it benefit for society interest. If people are satisfied with government services, the government's performance is good and on the contrary if people are dissatisfied with the services provided by the government, there needs to be improvements to improve performance for the better.

Utilization of information Technology and implementation of SAKIP
The rapid flow of globalization requires the organization in this case the government as a public sector organization to be able to adjust so as not to miss. One of the things that we can see and feel today is the development in the field of information technology.
By leveraging technological sophistication such as computer devices and Internet networks make the job much more effective and efficient than the conventional way. A demand that every organization today is required to make technology information as the main choice in creating a strong information system and able to give birth to a competitive advantage Problems often experienced in the implementation of SAKIP in West Nusa Tenggara provincial government is the collection of achievement data of Performance Annual Plan takes a long time, almost a week and even more. This condition makes implementation of SAKIP less effective and efficient, therefore nowadays SAKIP electronic (e-SAKIP) has been developed which utilizes information technology in implementation of SAKIP. Through e-SAKIP allows information to flow between parts of the organization directly, thereby reducing employees to meet face to face resulting in increased of organizational responsiveness. Razi (2017) found that the utilization of information technology affects the accountability of government performance. This is because the use of good information technology will help complete tasks more quickly and accurately such as storing information, disseminating information so that information can be processed more quickly to determine the next step. This study is line with research of Putri (2015), but Rofika (2014) found different results where the utilization of information technology has no effect on government institutions performance accountability (AKIP) because success in implementation of AKIP does not depend on the technological advances used.
Based on the description, it can be argued that the utilization of good information technology will be able to improve the quality of SAKIP implementation so that it will eventually reach normative isomorphism level. Therefore, in the research can be proposed the first hypothesis as follows: Utilization of information technology has a positive effect on the implementation of SAKIP. Mangkuprawira (2002: 135) stated that training as a process of teaching certain knowledge and skills and attitudes so that employees are more skilled and able to perform responsibilities better, according to the standard. Ideally, training should be designed to realize organizational goals, while simultaneously realizing the goals of individual worker.

Training and implementation of SAKIP
Researches conducted by Cavaluzzo and Ittner (2004) and Sofyani and Akbar (2013) found that training has a positive and significant effect on the implementation of SAKIP. These result indicate that training is a good instument for improving the capacity of employees to perform their duties properly and professionally. Professionalism itself is the emphasis of normative isomorphism in institutional theory to realize the implementation of SAKIP that touches on the substantive aspect rather than as a symbol or administrative demands of a mere formality. Considering the important of this training method, the organization in this case the government constantly pursues a structured and sustainable training related to performance measurement system in order to encourage the successful implementation of SAKIP reach normative level. Based on the description, it can be argued that the training carried out in a structured and sustainable will be able to realize the quality improvement of SAKIP implementation. Therefore, the second hypothesis in this study is that training has a positive effect on the implementation of SAKIP

Obedience to Regulation and implementation of SAKIP
According to Razi (2017) argued that obedience to regulation is the obedience of a person in complying with the regulations that have been established by state institutions or officials who have authority and binding power in order to regulate and discipline every life of the nation and state. In the implementation of SAKIP, various regulations are applicable as a guide starting from the planning, implementation, measurement, evaluation and achieving performance so that synchronization at each stage can work properly. Objective that have been set can be achieved as well as possible in accordance with the provisions as mandated by existing regulations.
The researches of Riantiarno (2011), Zulharman (2015) and Razi (2017) found result that obedience to regulation have a positive and significant impact on government performance accountability. These result indicate that the employees in carrying out official activities prioritize the principle of efficiency and supervision in accordance with applicable regulation so that the higher compliance with regulations, performance accountability is also increasing. The results above are different from research conducted by Putri (2015) that was the obedience to regulation have no effect to government performance accountability because of the possibility of employees do not know and do not understand about the scope of legislation used as a guide in the implementation of government performance accountability. Based on these arguments, then the third hypothesis of this study is the obedience to regulation has a positive effect on the implementation of SAKIP.

Implementation of SAKIP and Managerial Performance
In a strategic management, performance measurement serves as an assessment tool whether a defined strategy has been successfully achieved. From the result of performance measurement conducted feedback so as to create a performance measurement system that can improve organizational performance in a sustainable. Based on the feedback of performance measurement result, public managers can improve performance in the next period, both in planning and implementation.
Research on performance measurement system as a form of performance accountability to managerial performance was done by Simorangkir (2013). The results showed that the performance measurement system has a significant effect on managerial performance, it can be seen based on how the managerial performance in achieving targets set by the organization. So that performance measurement system will be able to improve the performance motivation of manager or official to provide the best for the organization. The results of this study is line with research of Hazmi (2012), Setiyawan (2016) and Putra (2016). Therefore, the quality of good SAKIP implementation as a form of government institution performance measurement system for the implementation of program and activities will be able to improve managerial performance. Based on these arguments, then in the research can be proposed the fourth hypothesis is the implementation of SAKIP has a positive effect on managerial performance.

Implementation of SAKIP with Managerial Performance Reinforced by Self Efficacy
Self efficacy as a person's level of confidence that he or she is capable of performing a broader and more proactive role beyond the traditional technical requirements established (Parker et al., 2003). The achievement of the goal set out in the planning document becomes a necessity for government organization, proving that the government has succeeded in providing the best service to the community so that the community become satisfied with the government's performance. Based on the perspective of institutional theory, the high self efficacy of public officials is expected to improve the quality of SAKIP implementation which will eventually lead to the implementation of SAKIP achieve normative level through managerial functions in organizational management.
Research conducted by Helmizuldi (2015) found the result that self efficacy has a significant effect on managerial performance. This results indicate that a manager is required to be able to take an appropriate decision in accordance with the demands of the organization, with a high self efficacy of a manager able to work effectively in various situations in making the right decision in the process of organizational because the manager knows about what to do with regard to achieving the organization's goals. The results of this study is in line with Aliffudin (2012). Based on these arguments then the fifth hypothesis in this study is self efficacy can strengthen the relationship between the implementation of SAKIP with managerial performance.

Implementation of SAKIP with Managerial Performance Reinforced by Intellect
High intellect will act directed, rational, thorough and prudent and minimize the emergence of problems in implementation through properly planning. Therefore, with high intellect of the public managers will make right and good planning so that the implementation will be on target in accordance with the expected that will ultimately show that the organization has a good performance.
The study of intellect or conscientiousness in the implementation of SAKIP was done by Korzaan and Boswell (2008) and Sofyani and Akbar (2013), the result concluded that individuals with high conscientiousness through rational, disciplined, prudent, tend to use available information make it possible to direct such individual actions to normative behavior patterns. Therefore, public managers who have high intellect will be able to manage the organization with all resources that exist in terms of SAKIP implementation so that it will be able to improve the quality of SAKIP implementation oriented to the substantive aspect in accordance with normative isomorphism in institutional theory. Based on these arguments then the sixth hypothesis in this study is intellect can strengthen the relationship between the implementation of SAKIP with managerial performance

METHOD
This research is a quantitative research with explanatory type. Population in this research are all structural officials of echelon II, echelon III and echelon IV at 46 Regional Apparatus Organizations (OPD) in West Nusa Tenggara Provincial Government which have an important role starting from process of planning, implementation and accountability of management of Regional Income Budget and Expenditure (APBD) through implementation of SAKIP as 138 people.
The sample technique with purposive sampling is referring to the rule of thumb of Cohen (1992) in Hair et al. (2014: 21), the sample size is determined from the number of largest indicators of exogenous variables with a significance level of 0.5% and a minimum R 2 of 0.25. The training variable has the biggest indicator that is six, then the minimum sample size that will be used is 75 respondents. In this study using 100 respondents as a sample.
The variables in this study were measured using Semantic Differencial Scale. Measurement scale is to measure the attitude is not in the form of multiple choice as on the likert scale but is composed of a continuum line with a choice of answers from values 1 to 7. If the respondent gives an assessment in the number 7 means very positive to the object in question, if the number 4 means neutral and if the number 1 means very negative.

Data Analysis
This study tested three exogenous variables and two endogenous variables and two moderating variables. Exogenous variables consist of the utilization of information technology, training and obedience to regulation. Endogenous variables consist of SAKIP implementation and managerial performance and moderating variables consist of self efficacy and intellect. The research model is shown in Figure 1.

Descriptive Statistics
Descriptions of research variables include: Utilization of Information Technology (IT), Training (T), Obedience to Regulation (OR), Implementation of SAKIP (IS), Managerial Performance (MP), Self Efficacy (SE) and Intellect (I) are shown in Table 2.  Table 2 it can be seen that: the variable utilization of information technology (IT) has a standard deviation value of 0.508 is smaller than the mean value. That is, the dominant sample value gathered around the average count of 6.692. From these results are then adjusted with the category table, so it can be said that the utilization of information technology in West Nusa Tenggara Provincial Government tends to be in appropriate condition or category.
The training variable (T) has a standard deviation value of 0.713 smaller than the mean value. That is, the dominant sample values gather around the average count of 6.645. From these results are then adjusted with the category table, so it can be said that the training in West Nusa Tenggara Provincial Government tends to be in appropriate condition or category.
The obedience to regulation variable (OR) has a standard deviation value of 0.525 less than the mean value. That is, the dominant sample values gather around the average count of 5.522. From these results are then adjusted with the category table, so it can be said that the obedience to regulation in West Nusa Tenggara Provincial Government tends to be in appropriate condition or category.
The implementation of SAKIP variable (IS) has a standard deviation value of 0.744 less than the mean value. That is, the dominant sample values gather around the average count of 6.538. From these results, then adjusted to the category table, so it can be said that the implementation of SAKIP in West Nusa Tenggara Provincial Government tends to be in appropriate condition or category.
Managerial performance variable (MP) has a standard deviation value of 0.666 is smaller than the mean value. That is, the dominant sample values gather around the average count of 6.539. From these results then adjusted to the table category, so it can be said that the managerial performance in West Nusa Tenggara Provincial Government tends to be in appropriate condition or category.
Self efficacy variable (SE) has a standard deviation value of 0.589 is smaller than the mean value. That is, the dominant sample values gather around the average count of 6.538. From these results then adjusted to the table category, so it can be said that the self efficacy in West Nusa Tenggara Provincial Government tends to be in appropriate condition or category.
Intellect variable (I) has a standard deviation value of 0.533 is smaller than the mean value. That is, the dominant sample values gather around the average count of 6.642. From these results then adjusted to the table category, so it can be said that the intellect in West Nusa Tenggara Provincial Government tends to be in appropriate condition or category.

Testing Result of Measurement Model, Structural Model and Goodness of Fit
Data analysis method used is Structural Equation Modeling (SEM) based on variant and usually referred to as soft modeling, through Partial Least Square (PLS) analysis tool. In this study testing was done with Smart_PLS 3.0.
This research uses PLS with the following reasons : first, PLS is a data analysis method based on data assumption does not have to be distributed normal multivariate (indicator with category scale ordinal, interval, until ratio can be used on same model), sample not necessarily big, ie the number of samples less than 100 or at least 30 can be analyzed. Second, PLS can be used to confirm the theory, which is still said to be weak, because PLS can be used for prediction, but can also be used to explain the presence or absence of relationships among latent variables. Third, PLS allows algorithm by using series analysis of Ordinary Least Square (OLS) to obtain the efficiency of olgaritma calculation. Fourth, in the PLS approach, it is assumed that all variance measures can be used to explain (Ghozali, 2008: 4).

Measurement Model Testing (Outer Model)
Three measurement criteria are used in data analysis techniques using Smart_PLS to assess the model. The three measurements are convergent validity, composite reability and discriminant validity. Discriminant validity of the measurement model with reflextive indicators is assessed based on cross loading measurements with constructs. Another method of assessing discriminant validity is to compare of the average variance extracted (AVE) value from each construct with the correlation between the construct and the other constructs in the model. If the result of the AVE square root value of each construct is greater than the correlation value between the construct and the other constructs in the model, it is means have a discriminant value of validity that is good (Ghozali, 2008: 25).
The approach to analyze First Order Factor using the repeated indikators approach. This approach has an advantage because this model can be estimated with standard PLS algorithm. Factor loading whose value is below 0.50 will be dropped from the analysis because it has a low convergent validity value. The first stage in smart_PLS is to evaluate the outer model, ie the iteration process of the indicator and the latent variable is treated as a deviation from the mean (mean) with the aim of seeing the relationship between the indicator and its construct. In the first stage, there are 10 indicators that are dropped because it has the value of outer loading below 0.50. These indicators are IT3, T2, OR1, OR5, IS6, MP9, SE1, SE4, SE6, I2. After the dropping, resampling process is done to get the loading factor whose value is above 0.50.

Discriminant ValidityTesting
Discriminant Validity is measured by comparing the square root value of average variance extracted (AVE) of each construct with the correlation between the construct and the other constructs in the model. If the AVE square root value of each construct is greater than the correlation value between the constructs with the other constructs in the model then it has a good discriminant validity value.
The AVE Root discriminant validity result of information technology utilization (IT) is 0.649 higher than correlation between the utilization of information technology (IT) with other constructs in the model. The next variable AVE construct is training (T) of 0.764 higher than the correlation between the training (T) with the other constructs in the model. The AVE Root of obedience to regulation (OR) is 0.695 higher than the correlation between the obedience to regulation (OR) with other constructs in the model. The AVE root of SAKIP implementation (IS) is 0.722 higher than the correlation between the implementation of SAKIP (IS) with other constructs in the model. The AVE root of managerial performance (MP) is 0.648 higher than the correlation between the managerial performance (MP) with the other constructs in the model. The AVE Root of self efficacy (SE) is 0.657 is higher than the correlation between the self efficacy (SE) with other constructs in the model. And the last is AVE Root of intellect (I) is 0.790 higher than the correlation between intellect (I) and the other constructs in the model. Based on the above values show that the AVE square root value of each construct is greater than the correlation value between the construct and the other constructs in the model, it is said to have a good discriminant validity value (Ghozali, 2008: 25).

ReliabilityTesting
Reliability test aims to measure the consistency of an instrument in the form of questionnaires in the research. A questionnaire is said to be reliable when one's answer to a question is consistent over time. Reliability of research instrument in this research is tested by using Composite Reliability and coefficient Cronbach's Alpha. Because Cronbach Alpha is more sensitive to many items on a scale and tends to underestimate, it is more appropriate to use Composite Reliability in measuring internal consistency. The value of Composite Reliablitily should be greater than 0.7 indicating that the data has a high reliability value (Latan and Ghozali, 2012: 81).
The Composite Reliability results showed the Composite Reability value for self efficacy construct < 0.70. Therefore, self efficacy construct is dropped from the model because it is considered unreliable. While the other constructs have Composite Reliability value > 0.70, it show the consistency and stability of instruments used high. In other words after the self-efficacy construct is dropped from the model then all constructs or variables of this study have become a fit measuring tool and all questions used to measure each construct have good reliability. The result of Composite Reliability can be seen in table 3 below :  (2017) Based on the test result using smart_PLS, where in convergent validity 10 indicators with outer loading value below 0,50 dropped, discriminant validity based on the AVE square root value of each construct is greater than the correlation value between the constructs with the other constructs in the model then it has a good discriminant validity value and self efficacy was dropped from the model because composite reliability value of self efficacy < 0.70. Furthermore conducted resampling of the model for getting good model and the result can be seen in Figure 2. Goodness of Fit Model measured using Rsquare of dependent latent variable with the same interpretation as the regression; Q-Square ( Predictive Relevance ) for the structural model, measuring how well the observation value is generated by the model and also its parameter estimation. The Q-square value > 0 indicates the model has predictive relevance; otherwise if the value of Q-Square ≤ 0 indicates the model lacks predictive relevance. Calculation of Q-Square is done by the formula: The value of Q2 is 0.654 or Q-square > 0 so that it can be stated that the structural model is also fit with the data or show the model has predictive relevance.

Evalution of Inner Model
Inner model test is done to see the relationship between construct, significance value and R-square of research model. The Inner Model is evaluated by using R-square for dependent constructs and t-test as well as the significance of the structural path parameter coefficients.
R-square value of SAKIP implementation is 0.451. The R-square value of 0.451 suggests that the construct variability of the SAKIP implementation can be explained by the variability of the construction of information technology utilization, training and obedience to regulation of 45.1 %, whereas 54.9 % are explained by other variables outside of the study. For R-square value of managerial performance is 0.369. The R-square value of 0.369 suggests that the variability of managerial performance constructs that can be explained by the construct variability of SAKIP implementation is 36.9 % whereas 63.1 % are explained by other variables outside of the study. The greater of the R-square value indicates the greater the independent variable can explain the dependent variable so the better the structural equation.

The Result of Hypothesis Testing
Hypothesis testing is proposed by inner model test by looking at R-square value which is Goodness-Fit test model. In addition, by looking at the path coefficients that show the coefficient parameters and the value of statistical significance. The significance of the estimated parameters can provide information on the relationship between research variables. The limit to reject and accept the hypothesis proposed above is that t-statistics must be greater than t-table 1.64, because of a one-tailed hypothesis for p < 0.05. The estimation output for structural model testing is presented in Table 4 .  (2017) Moderation testing by looking at the interaction of intellect as moderating variable and implementation of SAKIP as predictor variable to managerial performance as dependent variable. If the t-statistic value is greater than the t-table value 1.64, then the hypothesis is accepted and otherwise the hypothesis is rejected. The result of moderating effect testing is presented in Table 5.  Table 4 shows that the path coefficient of information technology utilization effect on SAKIP implementation is 0.339 and t-statistic 2,179 > 1.64 (t-table), thus the first hypothesis is accepted. This results prove that the utilization of information technology has a positive and significant impact on the implementation of SAKIP. This results support findings research of Putri (2015) and Razi (2017. This shows with the development of e-SAKIP that is SAKIP based on information technology, where data processing by using application or software then work become more effective and efficient compared to implementation of SAKIP with manual method, so the problem like delay of data collecting of Annual Performance Plan can be overcome Table 4 shows that the path coefficient between the effect of training on the implementation of SAKIP of 0.555 has a t-statistic of 6.144 > 1.64. Thus the second hypothesis is accepted. This results prove that training has a positive and significant impact on the implementation of SAKIP. This result support previous researches of Cavaluzzo and Ittner (2004) and Sofyani and Akbar (2013). Based on the results of this study proves that training is the right method to be applied by organization in improving the quality of human resources. Therefore, government agencies should continue to conduct training in a structured and sustainable thus can direct the SAKIP implementation achieve normative level in accordance institutional theory which emphasize at the work professionalism. Table 4 shows that the path coefficient between the effect of obedience to the regulation on the implementation of SAKIP is 0.049 and has a tstatistic value of 0.558 < 1.64, thus the third hypothesis is rejected. This results prove that obedience to regulation has not affect on the implementation of SAKIP. This result is not consistent with the findings research of Zulharman (2015) and Razi (2017) but support previous research was done by Putri (2015). Based on the result of this study it can be explained that the possibility is caused by the lack of direction or coordination from the managers to their staff related to the understanding of the regulation concerning the implementation of SAKIP and the existing regulations have not been able to accommodate the implementation of SAKIP more comprehensively as an example not yet regulated on the implementation of SAKIP through e-SAKIP which based on information technology. Table 4 shows that the path coefficient between implementation of SAKIP to managerial performance is 0.505 and the value of t-statistics is 4.855 > 1.64, thus the fourth hypothesis is accepted. This results prove that the implementation of SAKIP has positive and significant effect on managerial performance. The results of this study support findings research of Hazmi (2012), Simorangkir (2013), Setiyawan (2016) and Putra (2016). Based on this result can be explained that the performance measurement system in the form of SAKIP which is regulated in Presidential Regulation No. 29 of 2014 as a form of government accountability can improve organizational performance. The performance of government organizations can not be separated from the role of managers in implementing the functions of managerial. Implementation of SAKIP will generates relevant and objective information that help manager to take strategic steps such as program evaluation and revision and budget allocation to produce better programs for improvement organizational performance in the next period. Table 5 shows that the path coefficient of intellect and SAKIP implementation interaction influence on managerial performance is 0.339 and the value of t-statistics is 1.244 < 1.64, thus the sixth hypothesis is rejected. The results of this study indicate that intellect is unable to strengthen the relationship between SAKIP implementation and managerial performance. Individual characteristic like intellect is needed to direct SAKIP implementation become more better in accordance the researches result of Sofyani and Akbar (2013) and Sofyani and Akbar (2015) that individuals with high intellect through the way of thinking rational, disciplined, prudent, likes to analyze and tend to use available information makes it possible to direct those individual actions to normative behavior patterns that emphasize professionalism in accomplishing tasks.

CONCLUSION
Based on the findings of this study indicate a positive and significant impact on the utilization of information technology and training on the implementation of SAKIP, the implementation of SAKIP on managerial performance. Based on these results, the use of e-SAKIP application needs to be developed again in view of e-SAKIP especially in West Nusa Tenggara Provincial Government which was implemented in 2017 so that the quality of SAKIP implementation can be improved. Similarly, training conducted in a structured and sustainable will greatly contribute to improving the quality of human resources in the implementation of SAKIP. Through increasing quality of SAKIP implementation as a government performance measurement system will produce objective and relevant information on government programs and activities that are needed by public managers for further decision making such as program evaluation that has not reached target, budget alocation that has not been accurate, the priority scale for the next period will ultimately result services that satisfied to community as the main stakeholders in governance.