Optimalization Decisions of Local Governmental in Indonesia

The application of the local government finance system currently being used is one of the facilities from the Ministry of Home Affairs to local governments in the field of regional financial management to strengthen perceptions of regional financial management systems and procedures. This study looks at how task and user characteristics will affect the success of accounting information systems and optimal decision quality of all local governments by adopting the theory of Petter, Delone, and McLean (2013). Data was collected through questionnaires distributed to all residents of the provincial/district/ city governments in Indonesia, as many as 111 units of analysis. The data was processed using Structural Equation Modeling (SEM). This is a cross-sectional study viewed from a time horizon. The results showed a positive effect of task and user characteristics on accounting information systems and optimal head decisions decisions.


INTRODUCTION
A chief is someone with chief authority directing subordinates to do part of their work to achieve goals (Hasibuan, 2011). The chief is responsible for taking actions/decisions that benefit the organization, both profit -oriented and nonprofit organizations . Therefore, information is very important thing to consider before making a decision. According to  the decision is the selection of actions from several alternatives, which are the core of planning. High-quality decisions require high-quality information as well. Therefore, if the information system output does not meet good quality criteria, decision-making will not be optimal, reducing the quality of decision-making (Laudon & Laudon, 2014). The head of the organization must have access to accurate information to make decisions. It can be helpful before, during, and following the decision-making process (Socea, 2012). Therefore, the information produced must be of high-quality to be useful for the organization (Atrill & McLaney, 2009). Local governments also faced the problem of decision quality due to the delay in reporting, these findings are an indication that the decisions are still not optimal.
Information systems are crucial for managers in every organization. To be able to survive and succeed, the information system is needed by most organizations (Laudon & Laudon, 2014). According to Mulyani (2016) information systems can improve efficiency, effectiveness, and internal control in decision making. According to Romney and Steinbart (2018), information is good and useful if its relevant, reliable, altogether, on time, can be verified and accessible.
The importance of accounting information in the government sector stems from the government's responsibility to promote good governance (Sukmadilaga et al. 2015). The accounting information systems are needed for financial management improvement in decentralized or autonomous regions on the principles of transparency and accountability following Government Regulation (PP) Number 65 of 2010.
The authors' research on different local governments led them to conclude that the existing accounting information system is not yet completely integrated. At the end of each year, physical labor is required to complete the tasks of recording and computing the depreciation of assets. These tasks apply to tangible and intangible fixed assets, and other assets such as revenue and costs received or paid in advance. The data already available are subsequently sent to an accessible accounting information system, such as SIMDA or IPKD, among others, based on human computation. This approach permits capturing faults during the recording process and sending the data to the accessible information system.
The qualities of those who utilize accounting information systems are another factor that contributes to the effectiveness of these systems. According to the findings of research carried out by DeJesus & Eirado (2012) on the Brazilian public sector, it was discovered that the degree of user education will have an effect on the public sector's utilization of accounting information systems. The Petter et al. (2013) model includes dimensions of user characteristics of the attitude to technology, comfort, trust, and user expectations. With a high level of attitude to technology, comfort, trust, and user expectations, it increasingly supports the accounting information system success.
Based on observations in several accounting sections in the Provincial Government Regions in Indonesia, several problems are often faced, such as employees who are not accustomed to using existing information systems. Hence, they still find it challenging to use it, especially for older employees. Also, not all employees in the accounting department are graduates of the accounting field of study. Some employees are irresponsible, resulting in late submissions of reports from the SKPD to the accounting department.
This study looks at how much task and user characteristics influence the information system's success and the optimal quality of decisions. The model developed by Petter et al. (2013) is the basis for this study model. That model proposes that the performance of information system technology is impacted by the nature of the tasks being performed and the users' attitudes. The authors researched on local governments in Indonesia because there is a diversity in the choice of accounting information systems, the breadth of the area of each local government in Indonesia, and the complexity of its management and application. With these considerations, the authors argue that by taking the object of research in Indonesia, the data can be generalized to reflect conditions in Indonesia as a whole.

Variable
Dimension Indicator Item Task Characteristics (X1) (Hackman & Oldham, 1976;Petter et al., 2013;Kim & Soergel, 2006;Chang et al., 2003) 1 Task Difficulty  1  2 Task Variety  2  3 Task Routineness  3  4 Task Compatibility  4  5 Task Significance  5 User Characteristics (X2) (Petter et al., 2013;Holsapple et al., 2009)  This study was initiated to examine and predict the influence that certain job elements and user characteristics have on the efficiency with which an accounting information system may be deployed and the preferred option. Following the above description of scientific study aims, the authors employed one of the available analytic methods, Structural Equation Modeling (SEM). This research is a cross-sectional study since data collection is conducted just once and is concluded within a few months (Sekaran & Bougie, 2013). In light of this, a tabular summary of the discussed operational variables may be presented in Table  1.
This research use questionnaire and interview methodologies to collect primary data from 119 local governments in Indonesia. The research analysis units are the Regional Financial and Management Agency of Asset (BPKAD) and the provincial and regency/city government Regional Financial Management Agencies (BPKD) in Indonesia. The target respondents from this study were the Heads of the Division and the Sub-Division (Kasubid) of Accounting or at least senior accounting staff at BPKAD and BPKD provinces and regencies/cities in Indonesia.

RESULT AND DISCUSSION
From a total of 119 surveys sent to 119 provincial/regency/city governments, 111 were returned, with the following details: The responses to the questionnaire are organized into five categories, including gender, degree of education, age, length of work, and position, with the following data in Table 3.
To enrich the discussion, a descriptive analysis of data responses was conducted to determine the variable's condition. According to Cooper & Schindler (2014), a descriptive analysis may be carried out by using core symptoms in conjunction with measures of variability. Assessments of main symptoms include the mean, median, and mode, whereas indicators of variability include a range of scores and standard deviations. The average, the middle point, and the most common occurrence are all terms used to refer to the center's symptoms. In order to present a complete view of each variable, the authors use both mean and standard deviation. It is essential to create an overall picture of the features of the task, the characteristics of the user, and the effective implementation of the accounting information system by using the mean value and standard deviation of the answer scores supplied by the respondents and the best decision for the provincial/regional/city in Indonesia.  The quality of the task (X1) was measured using five indicators, and based on respondents' replies, a value of 5.73 was produced as an average score closer to a score of 6 on a scale of 1 to 7, and it could be deemed sufficient. Thus, it may be argued that most local governments in Indonesia have adequate task characteristics. The number of regions with above-average scores is less than those with below-average scores. User characteristics (X2) were assessed using eight different indicators, and based on the responses of those who took the survey, a mean score of 5.48 was reached. This was closer to a score of 5 on a scale from 1 to 7, and it was determined to be adequate.
Thus, it may be stated that most Indonesian local governments have adequate user characteristics. The number of regions with scores higher than the mean was significantly higher than the number of regions with scores lower than the mean. It was determined that the successful implementation of the accounting information system (Y) received a score of 5.61 on a scale ranging from 1 to 7. This was determined by utilizing eight different indicators to evaluate the performance of the implementation. This suggests that Indonesia's provinces, agencies, and cities have effectively adopted an accounting information system.
The data are analyzed with the structural equation model to determine the role that task variables and user characteristics have in the effective deployment of accounting information systems and the influence that this has on making the best decisions. The measurement model, as well as the structural model, are both included in the form. The structural model describes the link between latent variables. In contrast, the measurement model specifies the change in the percentage of each manifest variable (indicator) that latent variables may explain.
It is necessary to do a test of goodness of fit in order to characterize models as excellent or bad depending on how well they describe the variable connection under investigation (Hair et al. 2014). Structural equation modeling may be tested for model adequacy using a variety of model fit testing criteria, as shown in the following table. The adequacy model was evaluated using a Chi-square test, which produced a value of 435.3 and a p-value near zero. Hair et al. (2014) state that structural equation modeling does not favor a p-value on the lower end of the probability scale (smaller than 0.05). Going back to the data shown before, a pvalue lower than 0.05 indicated that the Chi-square test provided significant findings. In light of the Chi-square test results, it became clear that the model produced did not satisfy all of the prerequisites for an effective model. Another metric that continues to connect with the Chi-square test is the Root Mean Square Error of Approximation. However, according to Hair et al.(2014), the model can be accepted if the RMSEA value is less than 0.08.
According to the table, the RMSEA value of 0.078 was still less than 0.08; hence, the model fulfilled the requirement for a good model based on the RMSEA value because it was still less than 0.08. The model fails to meet the standards for a good model, as indicated by the GFI (Goodness of Fit Index) score of 0.759; according to Hair et al. (2014), a good model has a GFI value that is better than 0.90. Since the model satisfies the goodness-of-fit requirements for the size of RMSEA (0.077 0.08) as well as the sizes of NFI, TLI, CFI, RFI, and IFI (> 0.90), the estimated findings of the model may be accepted. This indicates that the resulting empirical model follows the theoretical model.
By utilizing the measurement model, it will be possible to ascertain which indicator represents the hidden variables in the most pronounced fashion. According to Hair et al. (2014), factor loadings that fall between the range of ±.30 to ±.40 are considered adequate for understanding the structure. In this particular research endeavor, there are 25 manifest variables, in addition to five concealed variables. Five manifest variables make up the latent variable of task characteristics. Eight manifest variables make up the latent variable of user characteristics. Eight manifest variables make up the latent variable of successful implementation of the accounting information system. Four manifest variables make up the latent variable of optimal decision. According to goodness-of-fit analysis, it has been concluded that the model can be accepted; this finding indicates that it can be employed to evaluate the study hypotheses that have been proposed. To obtain the effect of task characteristics (X1) and user characteristics (X2) on the successful implementation of accounting information systems (Y) and their influence on optimal decisions (Z).
Based on the weight of the elements it shown that indicator X1.2 (SIMDA/ SIPKD application helps workers perform many types of work) was the factor that had the most impact in terms of how well it reflected the underlying latent variables of the task characteristics variable (X1). In contrast, indicator X1.5 (each employee's job contributes significantly to the organization) was the least effective at capturing the latent variables of task characteristics. Next to the latent variables of user characteristics (X2), indicator X2.2 (SIMDA/SIKPD application can increase employee confidence in the results of their work) was the strongest one in reflecting the latent variables of user characteristics. Conversely, indicator X2. 7 (Employees who use SIMDA/SIPKD have experiences relating to information systems /technology) was the least accurate at capturing user characteristics' latent factors.
The most reliable sign that the accounting information systems were successfully installed is inside the latent variable (Y) is Y7 (SIMDA/ SIPKD application can boost staff productivity). In contrast, indicator Y1 (SIMDA/SIPKD application is freely accessible whenever and whenever needed) was the least effective at representing the latent factors associated with the successful adoption of accounting information systems. Lastly, in the optimal decision latent variable (Z), indicator Z1 (information presented by the SIMDA/SIPKD application helps them make decisions quickly) was the most powerful. Conversely, indicator Z4 (decisions made by can always overcome existing problems) was the weakest. The measurement model specifies the change in each manifest variable's (indicator's) fraction that may be explained by latent variables, whereas the structural model establishes the relationship between latent variables.
Excellent construct dependability is defined by Hair et al. (2014) as better than 0.70 and an expected average variance extracted (AVE) greater than 0.50. The extracted average variance value of 0.676 for the task characteristics latent variable suggests that it may reflect 67.6 percent of the information in each indication. As a result, the construct reliability value of the latent variable task characteristics remained higher than the required threshold of 0.70. Furthermore, the extracted average variance value of 0.579 for the latent variable of user characteristics indicated that the latent variable of user characteristics could, on average, reflect 57.9% of the information in each indication. As a result, the construct dependability value of the hidden variable of user characteristics (0.913) remained greater than the specified threshold of 0.70. The average variance extracted value of 0.717 in the latent variable of successful implementation of accounting information systems indicates that the latent variable of successful implementation of accounting information systems could represent 71.7 percent of the information in each indicator on average. As a result, the value of construct reliability of latent variables related with the success of accounting information system implementation (0.953) was still greater than the indicated threshold (0.70). The average variance extracted value of 0.575 for the ideal choice latent variable revealed that the best decision latent variable could reflect 57.5 percent of the information in each indication. The construct dependability value of the major optimal latent variable (0.839) then remained greater than the target value of 0.70.
A structural model will be developed when the measurement models for each latent variable have been detailed. This stage will look at how the external latent variable affects the endogenous latent variable. According to the conclusions of data processing using the Lisrel 8.70 software, the following are the route coefficients of each independent variable on the effective use of accounting information systems and optimal decisions: The R 2 result indicates that the task characteristics and user characteristics had a combined influence of 58.2% on the effective adoption of the accounting information system in Indonesian local governments. Then the characteristics of the task, user characteristics, and the successful application of the accounting information system simultaneously had an effect of 82.3% on the 's optimal decision. The relevance of the impact of exogenous variables on the effective deployment of the accounting information system and on the 's best choice was then investigated using hypothesis testing. The influence of task characteristics on the successful implementation of the accounting information systems The following statistical hypotheses will be used to test the first hypothesis, which is about the effect of task characteristics on the successful implementation of an accounting information system. H0: 1.1 > Task characteristics do not have a positive effect on the successful implementation of accounting information systems. Ha: 1.1 > Task characteristics have a positive effect on the successful implementation of accounting information systems.
Based on the data in table 4, it can be seen that the t value of the task characteristics variable to the successful implementation of the accounting information system (4.535) was higher than tcritical (1.64), and the probability value (0.000) was lower than 0.05. Because tvalue was higher than tcritical, and the chance of making a mistake was 5%, Ho was thrown out and Ha was accepted. The results of this test show that the task's characteristics had a significant positive effect on how well the accounting information systems were put into place. This means that the application of the accounting information system in the local governments of Indonesia will work better if the task's characteristics are more appropriate.
The results of this study are in accordance with Gelderman's (2002) showing there is an influence of the structure of the task on the satisfaction/success of the information system. Two major impacts on task uncertainty as one part of task characteristics and discovered that task uncertainty altered information characteristics and had a favorable impact on performance. Similar to Elbeltagi et al. (2005) claim that in the creation of Decision Support System (DSS) and Technology Acceptance Model (TAM), task characteristics greatly affect the success of information systems.
According to Kim & Soergel (2006) the characteristics of a task could be a combination of aspects such as the task's analysis, variation, difficulty, routines, complexity, and structure. The characteristics of the work will have an impact on how well the information system performs. According to Petter et al. (2013), the characteristics of the task, in particular its compatibility with other tasks and its level of difficulty, will have an effect on the achievement of the goal of the information system. The influence of user characteristics on the successful implementation of the accounting information systems. The second statistical hypothesis to be tested is the effect of user characteristics on the successful implementation of accounting information systems. H0: 2.2 ≤ User characteristics do not have a positive effect on the optimal decision. Ha: 2.2 ≥ User characteristics have a positive effect on the optimal decision.
According to the data in Table 4, the tvalue of the user characteristics variable based on the effective deployment of the accounting information system (5,764) was more than tcritical (1,64), and the probability value (0.000) was less than 0.05. Because tvalue was larger than tcritical, it was determined to reject Ho and accept Ha at a 5 percent probability of error. The findings of this test suggested that user characteristics significantly contributed to the effective adoption of accounting information systems. This indicates that the use of accounting information systems in Indonesian local governments would be more effective the more appropriate the characteristics of the users. Holsapple et al. (2009) stated that the characteristics of information system users have a substantial impact on the quality of information technology systems, which is supported by the findings of this study. Likewise, research conducted by DeJesus and Eirado (2012) on the public sector in Brazil, revealed that the education of users will influence the accounting information system application in the public sector. Meanwhile, according to Gomes (2020), cultural changes such as education and staff training are needed to ensure the successful implementation of accounting. Sacer and Oluic (2013) demonstrated that one of the factors that influence information technology is employee education, which is one part of the user's characteristics. This is similar to the research of Sacer and Oluic (2013) user engagement and expertise have a substantial impact on the accounting information system's installation success. Meanwhile, according to Mulyani & Kurniawan (2018), the success of SIPKD system implementation is affected by staff competence and organizational culture in the local government. The influence of task characteristics on optimal decision. The third hypothesis to be tested is the effect of task characteristics on the optimal decision with the following statistical hypotheses. H0: 2.1 ≤ Task characteristics do not have a positive effect on th optimal decision. Ha: 2.1 ≥ Task characteristics have a positive effect on the optimal decision.
The tvalue of the job characteristics variable to the optimal choice (1.990) was more than tcritical (1.64), but the probability value (0.047) was less than 0.05. Because tvalue was larger than tcritical, it was determined to reject Ho and accept Ha at a 5 percent probability of error. Thus, it may be stated that the qualities of the work significantly influenced the ideal option. This indicates that the judgments made by the leader of the local governments in Java will be optimum to the degree that the task's qualities are suitable. Following Chong's (1996) research, the findings of this study indicate that, in conditions of high uncertainty, the usage of information systems will grow, leading to more effective managerial decisions and, eventually, a boost in managerial performance.
In the meanwhile, Kim & Soergel (2006) demonstrated that task characteristics may be assessed objectively to quantify the fundamental structural structure of a task or subjectively to measure task participants' impressions. The complexity of the work may be seen easy, yet a certain task implementer may view it as challenging, which in turn will affect the quality of the performance.
The influence of user characteristics on the optimal decision. The fourth hypothesis that will be tested is the influence of user characteristics on the optimal decision with the following statistical hypotheses: H0: 2.2 ≤ User characteristics do not have a positive effect on the optimal decision Ha: 2.2 ≥ User characteristics have a positive effect on the optimal decision According to the data in Table 4, the tvalue of the ideal decision's user characteristics variable (2.080) was more than tcritical (1.64), and the probability value (0.038) was less than 0.05. Because tvalue was less than tcritical, it was decided to reject Ho and accept Ha at a 5 percent probability of error. Thus, it can be inferred that user attributes had a considerable beneficial effect on the best decision. The more pertinent the user's attributes are, the more ideal the decisions of the local government head in Indonesia will be.
This investigation is consistent with the findings of Viswanath et al. (2002) stating that the role of the user is very important in the decision-making process so that the best decision can be generated. Likewise, according to Holsapple et al. (2009), user characteristics will influence the 's decision making. Similarly, Buachoom (2018) reported that user attributes influence the quality of judgments. The older a person is, the worse the quality of their decisions. Similarly, in regard to the amount and types of available employment, the higher the number of jobs and the more types of work will make the quality of the decisions decrease. Conati et al. (2014) said that performance with low tasks where performance is the the final outcome of decision-making is affected by user attributes.
The influence of successful implementation of the accounting information system on optimal decision. The following statistical hypotheses will be used to assess the fifth hypothesis, which examines the impact of effective use of accounting information systems on the best executive decisions. H0: 2.1 ≤ The successful implementation of the accounting information system does not have a positive effect on optimal decisions. Ha: 2.1 ≥ The successful implementation of the accounting information system has a positive effect on optimal decisions.
According to the information in Table 4, the tvalue of the successful implementation of the accounting information system to the optimal choice was higher than tcritical (1.64), and the probability value was lower than 0.05. This was the case despite the fact that the probability value was lower than 0.05. It was found that the hypothesis Ho should be rejected and the hypothesis Ha should be accepted with a chance of error of 5%. This was due to the fact that tvalue was greater than tcritical. Based on the results of this test, it appears that effective utilization of accounting information systems has a significant positive influence on the optimal decisions made by the CEO. This shows that the more efficient implementation of the accounting information system will result in the head of the local government making more judgements that are in the best interests of the community as a whole.
The findings of this study align with DeJesus & Eirado (2012), conclusion that management information systems have a favorable and significant impact on performance. Furthermore, Abumandil & Bin Hasan (2016) assert that the quality of information influences the decision-making process, and that the party's leader is largely held accountable for poor decisions. Similarly, Romney & Steinbart, (2018) claim that the goal of a structured and processed information system is to facilitate decision-making and convey meaning to users. According to the findings of Sukmadilaga et al.'s (2015) research, good leadership has a positive impact on the positive impact of successfully deploying accounting information systems, and the success of accounting information systems can increase the quality of accounting information in businesses. Human resource management expertise, as well as the successful application of the SIPKD system, are important variables that contribute to the overall quality of financial statements. The enhanced quality of budgetary decisions as a result of the efficient implementation of SIPKD and other human resource capabilities (Kurniawan et al., 2020). This study also supports the conclusions of Novianty et al. (2018), who concluded that the quality of decision making is highly influenced by the management accounting information system's quality, dynamic capabilities, user ethics, and top management support Novianty et al. (2018) did this study.

CONCLUSION
The features of the tasks being performed were shown to have a favorable impact on the successful implementation of accounting information systems in municipal administrations. This suggests that the introduction of accounting information systems by local governments in Indonesia will be more successful the more tailored the characteristics of the task are to the system's capabilities.
Local governments' effective deployment of accounting information systems was positively impacted by user characteristics. The more appropriate the characteristics of the user are, the more optimal the decisions of the of the local governments in Indonesia will be. This investigation is consistent with the findings of Viswanath et al. (2002) mentioning that the role of the user is very important in the decision-making process so that the best decision can be generated. The main problem of the local governments in Indonesia is the weakness of accounting information systems because technology infrastructure is lacking. Besides that, the computer facilities are not available to every employee, so this hamper work because they have to take turns using computers.
Task factors positively influenced effective decision making. This test indicates that the features of the work have a considerable beneficial influence on the effective deployment of accounting information systems. This suggests that the introduction of accounting information systems by local governments in Indonesia will be more successful the more tailored the characteristics of the task are to the system's capabilities. The findings of this research agree with those found by Gelderman (2002), who found that the structure of the task has an effect on the degree to which an individual is satisfied with or successful using an information system. Increasing optimal leadership decisions in Indonesia can be further improved by creating good job descriptions and specialization of tasks to user.
User attributes positively influence optimum decision making. This test indicates that user attributes have a strong beneficial influence on the effective adoption of accounting information systems. This indicates that the implementation of accounting information systems by local governments in Indonesia will be more effective the more appropriate the characteristics of the consumers. Users of information systems are said to have a major impact on the overall quality of information technology systems, as stated in the research conducted by Holsapple et al. (2009). The results of this investigation lend credence to the aforementioned assumption. Because this would have an effect on the quality of the financial statements that are produced, which are the result of decisions made by leadership, it is still necessary to improve the quality of human resources in local government in Indonesia by placing individuals in the appropriate positions. This is because this would have an impact on the quality of the statements that are generated.
The efficient implementation of accounting information systems has a large and positive impact on the executive officer responsibilities that are most suitable for the firm. This means that the more effective deployment of the accounting information system will lead to the head of the local government in Indonesia making more sound decisions as a result. The conclusions of this study are congruent with the findings of Abumandil & Bin Hasan (2016), who discovered that the quality of information effects decision-making, with the being the party that is most usually blamed for making the incorrect choice. The findings of this study are congruent with those of Abumandil & Bin Hasan (2016). Similarly, according to Abumandil & Bin Hasan (2016), an ordered and processed information system strives to convey meaning and enhance the decision-making process. Financial report, as the result of the 's choice, will be heavily impacted by the effective application of information systems connected to the report's quality and timeliness.