Cloud Computing Adoption Strategies at PT Taspen Indonesia, Tbk

PT. Taspen as Indonesian institution, is responsible for managing social insuranceprograms of civil servants. With branch offices and business partners who are geographicallydispersed throughout Indonesia, information technology is very important to support thebusiness processes. Cloud computing is a model of information technology services that couldpotentially increase the effectiveness and efficiency of PT. Taspen information system. Thisstudy examines the phenomenon exists at PT. Taspen in order to adopt cloud computing inthe information system, by using the framework of Technology-Organization-Environment,Diffusion of Innovation theory, and Partial Least Square method. Organizational factor isthe most dominant for PT. Taspen to adopt cloud computing. Referring to these findings,then a SWOT analysis and TOWS matrix are performed, which in this study recommendsthe implementation of a strategy model of cloud computing services that are private andgradually in process.

IS/IT infrastructure development is one RI WKH GLVFRXUVH VWUDWHJLHV SROLFLHV DQG ZRUNLQJ programs of top level management at PT. Taspen. As state institution, the IS / IT is a critical component in its business processes. Referring to the duties and responsibilities of the IT Division of PT. Taspen, WR DVVHVV WKH LQQRYDWLRQ RI QHZ WHFKQRORJLHV DQG V\VWHPV WKDW DQDO\]H DQG WHFKQRORJ\ E\ ZD\ RI outsourcing, cloud computing services can certainly be the potential and challenges for PT. Taspen in strategic planning of IS/IT based on company needs.  [4]. Research conducted in Korea reveals the fact that the organization, adoption of cloud computing LV LQÀXHQFHG E\ VHYHUDO IDFWRUV VXFK DV RUJDQL]DWLRQ FKDUDFWHU WKH SURFHVV RI GHFLVLRQ PDNLQJ ZLWKLQ DQ organization, implementation phase, competitive effects, services availability, economic factors (price), and managerial support.
In Indonesia, the ministry of communication and informatics count the cloud computing trend on the agenda "Indonesian Communication and Information Technology White Paper 2010" [5]. Mentioned that the driver of cloud computing trend is VDYLQJ WKH FXVWRPHU LQYHVWPHQW QHHG IRU HI¿FLHQF\ increased reliability through the elastic resource availability, reduce the threat of single-point-offailure, and increased utilization.
What factors affect the PT. TASPEN in adopting cloud computing services in the context of technology? 2.
What factors affect the PT. TASPEN in adopting cloud computing services in the context of the organization? 3.
What factors affect the PT. TASPEN in adopting cloud computing services in the context of the environment?

Research Objective
The main purpose of this research is to study ZKDW IDFWRU LV LQÀXHQFLQJ

Diffusion of Innovation Theory
In many previous studies, DOI is one very popular theory to uncover an individual factor in adopting a technological innovation [11] [28].

SWOT/TOWS matrix analysis
In his dissertation, Shimba states to begin DGRSWLQJ FORXG FRPSXWLQJ VHUYLFHV PXVW EHJLQ ZLWK WKH analysis phase [29]. SWOT analysis is a relevant tool to develop strategies in the decision-making process [30]. In general SWOT analysis related to cloud computing can be seen from the characteristics of the technology

LWVHOI DVVRFLDWHG EHQH¿WV DQG GUDZEDFNV DV ZHOO DV WKH SRWHQWLDO EHQH¿WV DQG ULVNV WKDW DIIHFW XVHUV > @
To develop strategies based on existing data, the TOWS matrix analysis is one approach for researchers to produce a strategy for a company or organization. Basically this is the use of TOWS matrix strength to create opportunities, minimize ZHDNQHVVHV DQG DYRLG H[WHUQDO WKUHDWV > @

Research model and hypothesis
Grounded By using SmartPLS program, three options available methods of analysis algorithms that can EH XVHG 3DWK RU VWUXFWXUDO ZHLJKWLQJ 3/6 DQDO\VLV algorithm scheme suggested by Wold is using a path RU VWUXFWXUDO ZHLJKWLQJ PHWKRG > @ 'DWD DQDO\VLV using bootstrap resampling method.    In the model, for each manifest variable should not be correlated. To test this principle it is QHFHVVDU\ WR WHVW GLVFULPLQDQW YDOLGLW\ E\ ¿QGLQJ WKH value of cross loading or square root. Square root of the value of each variable is greater than the value of the correlation [40]. Thus, all the variables in this model is valid.
In addition to testing the validity, reliability test should also be done. This test is done to prove the accuracy, consistency, and precision instrument to measure the construct or variable [39]. Reliability WHVW LV GRQH E\ ¿QGLQJ WKH YDOXH RI &URQEDFK ¶V DOSKD (must be greater than 0,7).
Results of Cronbach's alpha values stated for all reliability indicators have values above 0.7. Thus stated that reliable models.
Effect size or value of the F-Square is interpreted as a large or small effect or latent predictor variables in structural models (table 11).
Predictive relevance is one aspect that can be studied in the inner models. Predictive relevance or Q-Square is also called predictive relevance reuse. This technique is obtained by the procedure in the program bindfolding SmartPLS.  Predictive relevance value is 0,517 or greater than 0. It means model has strong predictive relevance.
Hypothesis can be tested using resampling ERRWVWUDSSLQJ %RRWVWUDSSLQJ ZLOO JLYH WKH YDOXH of T-Statistic from each relation of exogenous to endogenous variable.