Model Satisfaction Users Measurement of Academic Information System Using End-User Computing Satisfaction (EUCS) Method

— One of the concrete actions of the vision and mission of an organization or educational institution is to implement the system of academic information. The one factor that influences the success of the academic information system is the satisfaction of users. The problems in the user satisfaction are also issues that cannot be denied and continued to be studied. In consequently, having measurements related to user satisfaction is required. This research is done to see how the user satisfaction of the existing system has been implemented and to know what factors has influenced the system. The method used in this study is a quantitative method using the End User Computing Satisfaction (EUCS) model whereas 255 respondents are chosen as the sample for the study. The purposive sampling technique is used and the PLS-SEM approach with tools Smart PLS 3.0 is used for analyzing the data. This research consists of 7 hypotheses and 8 variables, namely Content, Accuracy, Format, Timeliness, Ease of Use, System Reliability, System Speed and End-User Satisfaction. The research finds out that the current level of end-user satisfaction is in satisfying level. Inferentially, of the 7 hypotheses tested, 2 of them were rejected and the other 5; Accuracy, System Reliability, Timeliness, Content and System Speed are accepted. Through this research, it is hoped that later it could be used as a practical consideration and theoretical study in the future.


I. INTRODUCTION
long with the development of the times in the modern era as it is today, the advancement of information and communication technology has now developed very rapidly, and provides many influences in all fields.For this reason, in its management, higher education institutions are required to have adequate excellence and competitiveness by improving _______________________________________________________ Received: 6 Febuari 2020; Revised: 20 Januari 2021; Accepted: 22 Januari 2021 R.A. Hidayah, , is a Department of Information System UIN Syarif Hidayatullah Jakarta Indonesia.(nur.aeni@uinjkt.ac.id) the quality of academic services [9].One of the efforts taken is the use of academic information systems, that is now becomes an obligation for higher education institutions to succeed [14] [7].
There are many researches that test several variables in order to see the effect on the users satisfaction.One of those is Ilias et al. [5] that add system speed and system reliability.The result is that these variables significantly influence the user satisfaction.
Another study by Rosalina in Putra et al. [10] evaluates the satisfaction of the academic information system of UIN Jakarta using the EUCS model that is extended with security variables.The results of this study state that the variables of content, timeliness, ease of use, and security in the EUCS model affect user satisfaction, while variables of accuracy and format do not affect user satisfaction.Based on the above, it is then important to know the level of users satisfaction and the factors affect it.In doing so this research was intended to determine the status of end-user satisfaction of the academic system based on the perception of respondents and to examine the factors that influence the satisfaction of users based on the model of academic information system end-User Computing Satisfaction.

A. Definition of Measurement of User Satisfaction
User satisfaction measurement is a measurement of behavior information system users in terms of their response to several related factors in delivering information about products and services [19].This is supported by previous research stated that satisfaction is the main factor in measuring the success of an information system [20].Then it can be inferred that measurement of information system user satisfaction is a measurable assessment conducted hrelated to the user's perception regarding the wearing of an information system by http://journal.uinjkt.ac.id/index.php/aismlooking at user satisfaction as the main indicator.

B. Academic Information System
According to Arifin [18], the Academic Information System is a resource for everything in the form of information that has to do with academic problems on campus.In addition to the information resource on campus, the Academic Information System can also be used as a medium of communication between lecturers and students, students with lecturer, students and related campus officials and anyone in the campus environment.The researcher concluded the Information System Academic is a system that is used to accommodate all matters relating to the academic process in order to improve the quality of the teaching and learning process as well as administrative quality related to the entire academic community at the university.

C. End User Computing Satisfaction
The EUCS evaluation model was developed by Doll & Torkzadeh [1], emphasizes end-user satisfaction with technological aspects by looking at several factors namely content, accuracy, format, ease of use and timeliness.
Content is the dimension to measure user satisfaction in terms of the content of a system.Accuracy is used to measure user satisfaction in terms of data accuracy when the system receives input and then processes it into information.Format dimensions are used to measure the user satisfaction of the appearance and aesthetics of the interface design.
Ease of use is a dimension used to measure satisfaction of the ease of the user in using the system.Timeliness dimension to measure the timeliness of the system presents data [11].

D. System Speed
System speed is one of the features that can affect the web users satisfaction.Based on research conducted by Chin & Lee [17] states that, the operational speed of a system increases user satisfaction.This is supported by the statement of Nawangsari et al. ( 2008) that the access dimension in this case the speed of access to a website, whether downloading or obtaining data or information has a significant impact on the level of user satisfaction.

E. System Reliability
System reliability is also an important variable that has an effect on system user satisfaction.System variable reliability is declared valid and has an effect on system end-user satisfaction [5].This variable measures user satisfaction in terms of information system's resilience towards damage and error. .

III. PROPOSED & HYPOTHESIS MODELS
Based on this opinion, the researchers tried to develop the EUCS model by adjusting the facts of the problems faced by system users in the field.Model development carried out by researchers is also inseparable from the literature review which shows the development of the EUCS model with the addition of other variables that also affect user satisfaction.This does refer to the statement of Ilias et al., [4] which states that many researchers define end-user satisfaction based on objectivity and studies conducted by themselves.So that the researcher developed 7 independent variables namely content, accuracy, format, timeliness, ease of use, system speed and system reliability and the End User Satisfaction as dependent variable in Fig. 1.

IV. RESEARCH METHODS
In general, this study uses a quantitative method (Sugiyono, 2013) as a research procedure that produces data in the form of numerals presented tables and diagrams.In this study, a research model was developed as a source of formulation of a number of hypotheses, where the study consisted of 7 hypotheses.This research itself starts from data collection by conducting library studies and studying related literatures.Furthermore, the researchers also conducted observation studies by visiting the research site and making observations related to the general description of the academic information system which is the main object of research.After making observations, the author conducted an interview with the respondent's representative.As for this study, the authors compose several hypotheses, where the hypothesis then tested using data that has been collected from the questionnaire.
The questionnaire was designed in the form of variable statements based on the EUCS model that was developed and distributed to end-users of the system (students, lecturers and academic staffs).The scale used in this questionnaire is 5 Likert scale.In determining the sample, researchers used a purposive sampling technique where the selected samples are users who already had experience in using academic information systems.Furthermore, quantitative data analysis was carried out through the PLS-SEM approach with Smart PLS [15][2].This is done to analyze measurement and structural models.
In addition, structural analysis models is performed to examine the inner workings model through several stages of testing, ie testing the path coefficient, coefficient of determination, t-test, effect size, predictive relevance, and the relative impact [15] [2].The researcher also interprets the data based on the results of the analysis.Finally, researchers make conclusions and suggestions in accordance with the limitations and research hypotheses that have been made previously.The details of the process are shown in Fig. 2.

A. Results of Demographic Analysis
This stage will display demographic information regarding to the characteristics of the respondents, the role of the system, and the status of the system's user satisfaction.The results show that most of the respondents are female (56%) and college students (86%).And Most of them are from the Faculty of Engineering (16 %) and the Faculty of Teaching Sciences (14 %).From the analysis it can be concluded that about 71% of respondents feel the systems assist them in many ways and about 56% of respondents are satisfied or even very satisfied when using the system.For more details, the results of demographic analysis can be seen in Table I.

B. Inferential Analysis Results a) Model Measurement Analysis Results
The outer model is measured by four stages of testing, namely testing of individual items reliability, internal consistency, average variance extracted, and discriminant validity.The explanation of the results of the model measurement analysis is as follows: • The results of individual testing of reliability items have met the minimum standards which is 3-5 indicators in this study have had a value of loading factors above 0,5 [15].So there is no elimination of the indicators used in this study.See Table 2. http://journal.uinjkt.ac.id/index.php/aism • Results Internal consistency testing shows that all variables in this study already have a Composite Reliability (CR) value above the threshold of 0.7.See Table 3.

•
The average variance extracted test results show that all variables in this study have the value of Average Variance Extracted (AVE) above 0.5.See Table 3 where the correlation between indicators and constructs must be bigger than the correlation with other block constructs.Whereas, for Fornell-Lacekr'scross loading examination, the root value of AVE must be higher than the correlation between constructs and other constructs.
The results of discriminant validity checks as a whole, as well as the value and condition of cross loading can be said to be fulfilling, so the test can be continued at the structural analysis phase of the model.In addition, R2 (coefficient of determination) has a level of 70.6% which means that 7 variables used as hypotheses in this study are considered to have represented the dependent variable of end user satisfaction (end user satisfaction).

b) Model Structural Analysis Results
Structural analysis model ( inner model ) is done through six stages of testing, namely testing path coefficient ( β) , the coefficient of determination (R 2), ttest, effect size (f 2), predictive relevance (Q 2) and the relative impact (q 2). .The results can be seen in the table 7.
Overall, the research model has gone through the structural testing phase of the model to find out about how much the influence of the linkages and the relationship between the variables used in this study.First, for the results of the path coefficient (β) stage, it is known that 2 of the 7 paths have an insignificant effect on the model (i.e., the Format path → End User Satisfaction, Ease Of Use → End User Satisfaction) because the path coefficient value of the 2 paths is below the threshold (ie 0.1).
In addition, referring to the results of the t-test phase, it is known that 5 of the 7 hypotheses proposed in this study were accepted.This is because the value of the t-test from the 5 hypotheses is above the threshold value (ie 1.96).
Furthermore, the results of the effect size (f2 ) stage on all 7 lines indicate that 7 paths as a whole have a small effect.Then, for the results of the predictive relevance stage (Q 2 ), it states that all the variables in this study have a predictive relationship of 46.6%.Finally, for the results of the relative impact stage (q 2 ), the 7 lines have a small effect.See Table 4 below

C. Discussion
Based on the results of the structural analysis model, it is known that H1 (content), H2 (accuracy), H5 (timeliness), H6 (system speed), and H7 (system reliability) in this study were accepted.While FOR (format), EOU (Ease of Use) variables have no effect on system end-user satisfaction.This proves the consistency of the findings of the previous research states that the five variables proved to affect user satisfaction.While the rejected variables namely formatting variables and Ease of Use are also consistent with the previous research conducted [6].Thus the results of this study indicate that there are 5 hypotheses accepted in this study, namely CON → EUS; ACC → EUS; TEAM → EUS;

TABLE I .
PROFILE OF RESPONDENTS