Does Top Management Support Improving the Quality of Ais?

There is extensive growth in Accounting Information System (AIS) utilization during the COVID-19 pandemic in order to adapt to this situation. Due to the great increase in the use of AIS, it is necessary to develop the quality of AIS itself to produce good quality accounting information for users. The accounting information system can be affected by many factors. This study aims to analyze the effect of top management support on the quality of the AIS. This research was conducted on fifty five private banks in Indonesia. The data was retrieved using a questionnaire distributed directly or online. This type of research is explanative research with a quantitative approach. There are two research results, namely: (1) top management support has a positive effect on the quality of the AIS, (2) the quality of the information system has a positive effect on the quality of accounting information. This is an open access article under the CC BY license


INTRODUCTION
The industrial revolution 4.0 is when technology was increasingly developing and could be used in various aspects of life, including banking.The development of technology in banking is expected to create efficiency for customers when making transactions.One example of technological developments which occurred in banking is the launch of the banking application (mobile banking), which is a form of financial technology that banks can utilize to create convenience for customers when making transactions, from viewing account balances to making transfers which makes it easy for customers through banking applications that can be found on their mobile phones.Financial Service Authority data in July 2019 stated that there are more than 11 million fintech users in Indonesia.This data shows how highly interested Indonesia is in using fintech.These opportunities can be seized by banking institutions, for example, by implementing an AIS employing information technology (Wulandari and Juliarsa, 2017).
High-quality AIS can produce the qualified accounting information (Al-Okaily et al., 2020).The AIS (AIS) is a device that is formed in such a way as to conduct the process of processing and controlling a company's finances (Grande et al., 2011).How well the system can supply the information the business requires could be used to evaluate the quality of the AIS.(Mokodompit & Wuriasih, 2017).
In practice, the AIS in banks is yet to function correctly.As experienced by Bank Mandiri, which encountered a disruption in the information technology system it uses.This error occurred during the data transfer process.The impact of this error was that around 10% of Bank Mandiri customers received inappropriate account balance information.This prevented Bank Mandiri customers from getting accurate information regarding their balances (Rachman, 2019).In banking companies, the quality of accounting information has an essential role in improving performance (Apip, 2015).
The quality of accounting information is affected by a number of things.Dewi et al. (2019) stated that the quality of information is affected by an internal control system and the competence of human resources.Meanwhile, Fitrios (2016) said that the installation of the AIS has an impact on the quality of accounting information as well.Al-Hiyari et al. (2013) offered suggestions for utilizing top management assistance for the adoption of AIS.because, according to the findings of their study, management commitment during the implementation of AIS without any management support will not significantly affect the quality of information.The top management has a close relationship with information systems because the top management uses information generated by the information system to make decisions (Wulandari & Juliarsa, 2017).Top management support can therefore be employed as a key success element for the quality of AIS on the quality of accounting information in Indonesian banking firms.

THEORITICAL FRAMEWORK Top Management Support
Top management support (TMS) is the degree to which the top management is aware of how information systems work and how they participate in closely linked activities.Therefore, TMS is one of the factors that can affect the AIS (Khamis & Ishwara, 2017).This statement aligned with the results of previous research (Limroscharoen, 2018) that stated top management support significantly affects the implementation of AIS.TMS creates a good information system implementation process (Thong & Raman, 1993).Support provided by the top management on information systems will result in a sound system design, usage, and management to improve organizational performance (Khan & Mirchandani, 2013).This is aligned with previous research (Afrizon et al., 2019), which stated that with the support of top management, the performance of the AIS could go well.Based on the above statement, the first hypothesis is: H1: The Top Management Support has a positive effect on the quality of the AIS

AIS and Accounting Information
AIS can be in the form of manual systems or computerized systems (Salehi et al., 2010).Regardless of the type, an AIS is designed to collect data to process the data into information (Salehi et al., 2010).The AIS consists of three subsystems: a transaction processing system, a financial reporting system, and a management reporting system (James Hall, 2008).The AIS is the primary key to producing helpful information for internal and external parties (Kiroğlu, 2020).
The degree to which AIS may increase user performance and increase user happiness is a sign of how well it has been implemented (Salehi et al., 2010).An AIS can be regarded as qualified if it has the criteria that include security, ease of access, reliability, and data integration and can provide accurate and timely information (Rosa & Purfini, 2019).Integration, flexibility, reliability, and efficiency (Napitupulu, 2016) can be used to measure the effectiveness of AIS.The measurement of the effectiveness of AIS in this study refers to the research of Napitupulu (2016), namely, integration, flexibility, reliability, and efficiency.
Information results from processed data (Romney et al., 2018).The degree to which AIS may increase user performance and increase user happiness is a sign of how well it has been implemented.(Romney et al., 2018).Information is needed to adjust to technological changes (Yadav, 2006).Qualified information is relevant, timely, accurate, and concise (Hall, 2008).Relevant means that information has to be able to meet the objectives.Information systems have to present relevant information to avoid any wrong decisions (Hall, 2008).Timeliness means that information has to have a period following the actions supported (Hall, 2008).Accurate means that the information has to be accurate.Complete means that information has to provide a clear message (Hall, 2008).
We use Hall's (2008) assessment of accounting information quality as a benchmark for this study: relevant, timely, accurate, and complete.According to prior research (Kurniawati, 2018), an efficient information system will result in a superior level of financial accountability through the quality of financial statement information.The use of AIS in a company affects the quality of accounting information (Algrari & Ahmed, 2019).According to the given description, the second hypothesis is: H2: The quality of the AIS positively affects the quality of accounting information.

RESEARCH METHOD
This study combined a quantitative approach with an explanatory research methodology.The study's moderating factor was top management support, and the variables investigated were the quality of AIS that affects the quality of accounting information.Primary data are the sort of data used in this investigation.This study collects the data through questionnaires distributed via post, google form, and direct contact.In this study, the population consisted of 83 private banks in Indonesia.The observation unit in this study was employees who use AIS applications.This study used employees from private banks who actively use the AIS.The sampling technique conducted was simple random sampling through Microsoft Excel.The minimum sample that had to be fulfilled was 30 respondents (Cohen et al., 2007).The data analysis technique used was SEM-PLS.

RESULT AND DISCUSSION
The research's hypothesis, as well as the impact of the independent variable on the dependent variable, are both tested using SEM-PLS.SmartPLS 3 was the SEM employed.Testing the outer and inner models is the second step of data processing.

Outer Model Testing (Measurement Model Testing)
The convergent validity, discriminant validity, Cronbach's alpha, and composite reliability must all be assessed at this time (Hair et al., 2014).This is how data processing results are displayed:

Convergent Validity
For each dimension and each indicator, the loading factor number that served as the benchmark in validity testing is visible.If the loading factor is greater than 0.7, then each dimension and indicator is regarded as legitimate (Hair et al., 2014).However, the loading value of 0.5 to 0.6 is still sufficient for the early stages of the investigation (Ghozali & Latan, 2015).The table below displays the results of the tests conducted to determine the loading factor and convergent validity.The amount of the correlation between latent variables and each indicator is indicated by the loading factor number provided in the previous table.The results of the SmartPLS algorithm can directly show the loading factor value in the outer output setting.The findings show that all of the study's indicators are reliable.The average variance extracted (AVE) value is the convergent validity metric that follows.The variables are declared to be legitimate if their AVE value is greater than 0.5 (Hair et al., 2014).It can be justified that the 11 latent variables construct acceptable validities (AVE> 0.5) based on the aforementioned AVE value.Information from each latent variable can be reflected in its visible variables.

Cronbach's Alpha and Composite Reliability
Cronbach's alpha and composite reliability are two factors used in the measurement model reliability test.The output overview of the SmartPLS algorithm findings shows this.The suggested criteria number is greater than 0.700.(Hair et al., 2014).The outcomes of the Cronbach's alpha test and the composite reliability for each study variable are as follows: Because each variable has a value higher than the suggested value, Table 3's findings of the Cronbach's alpha test and composite reliability are stated to be reliable.This shows the high degree of dependability of the assessment model.Using the results of the data processing described above, it can be concluded that the measurement model is accurate and dependable enough to satisfy the criteria for further investigation.

Inner Model Testing (Structural Model)
The SmartPLS application's t-statistical test is used to determine whether the partial hypothesis arrived at utilizing Bootstrap calculations is correct (Ghozali & Latan, 2015) and the R Square (R 2 ) and Q Square (Q 2 ) analysis (Hair et al., 2014) became the test tools in this analysis stage.

R Square (R2) Analysis
The R Square (R 2 ) analysis on each endogenous latent variable showed how much each exogenous variable affected the endogenous latent variables' degree of effect.The impact of endogenous variables increases with increasing R 2 value (Hair et al., 2014).

Q Square (Q2 ) Analysis
The Q square value is calculated to evaluate the performance of the structural model.The guidelines state that a model is predictively relevant if Q2 is more than 0 and is not predictively relevant if Q2 is less than 0. The predictive significance is evaluated using the following formula: The value of Q 2 = 0.932> 0 which indicates that the model has a good predictive relevance.

Hypothesis Test
The hypothesis test is used to test the effect of the independent variables on the dependent variables.In SmartPLS, bootstrap with a significance level of 5% is used to test the significance of the path coefficient.The following figures and tables present the results of the calculations to test the hypothesis.

The Effect of TMS on the Quality of AIS
The original sample value of 0.328, which denotes a positive value, can be seen from the analysis findings from SmartPLS 3, which are displayed in the table above with a significance level of 5%.9.464> T table (1.960) is the statistical number that results, and the p-value is 0.000 0.05.H1 in the study is accepted, which means that top management support positively affects the quality of the AIS.It means that the greater the support from top management, the quality of the AIS will increase.Top management must create a situation that supports employees while the AIS implementation process occurs.

The Effect of the Quality of AIS on the Quality of Accounting Information
The analysis results from SmartPLS 3, which are shown in the table above with a significance level of 5%, reveal the original sample value of 0.393, which showed a positive value.The p-value is 0.000 0.05, and the statistical value is 4,213> T table (1,960).H2 in the study is accepted, which means that the quality of the AIS positively affects the quality of accounting information.The higher the quality of the AIS, the higher the quality of the accounting information produced.AIS that can perform well will produce a piece of qualified accounting information.In addition, AIS must be flexible to changes inside or outside the organization.

CONCLUSION AND RECOMMENDATIONS
First, TMS favours the quality of the AIS, according to the study's findings.This suggests that the level of top management support will directly affect how well the AIS performs.Second, the quality of the AIS has a favourable impact on the quality of financial information.A sound AIS will also generate accurate bookkeeping data for the business.Based on the research that has been conducted, to be able to improve the quality of the AIS, things that need to be a concern for top management are providing support through delivering motivation to employees and also creating a situation that supports the implementation process of the AIS.

Table 2 .
Extracted Average Variance (Convergent Validity)Test Result

Table 3 .
Cronbach's Alpha and Composite Reliability Test Result