An Empirical Study to Understanding Students Continuance Intention Use of Multimedia Online Learning

The purpose of this study was to assess students' ongoing intentions towards online multimedia learning such as perceived usefulness, ease of use, and flow experience. The sample of this study was 523 students who used off-campus/online learning resources and examined the content of online learning resources and their multimedia aspects. The Extended of Technology Acceptance Model (TAM) was used to predict students' continuing intentions. The results showed that students' intentions were positively influenced by their perceived usefulness, ease of use, and flow experience. It is suggested that the designer of multimedia online learning should be more specific in determining the target users to receive and cultivate a more positive sustainable intention.


Introduction
Online learning is one of the new learning paradigms and is becoming the most prospective and growing rapidly today [1][2][3][4].With its rapid growth, online learning providers have produced and designed many competing learning template models such as Moodle, Edmodo, whiteboard system, etc.To proactively achieve learning objectives and design interventions (teacher training, user orientation, student assessment and evaluation, etc.), it is important for designers and managers of online learning to understand how to attract the attention of users who may be less interested in adopting and using online learning systems.
Because students are the end users of the online learning system, learning designers need interesting media to convey the learning content provided by the supervisor or lecturer.Students' attention to the lesson, for example, can be maintained by providing menus and tools for communication channels, discussion forums, message boards, and chat rooms.To be effective and useful for students, the content of the lesson determined by the mentor or lecturer must be interpreted by multimedia specialists in an interesting, economical and communicative way.Thus, multimedia not only serves as a means of communication but also increases student learning motivation, attention and leads to student success [5].
The extent to which a person believes that using a particular system will improve student activity performance is called perceived usefulness [6].In the past few years, many studies have revealed that perceived usefulness as well as ease of use (the degree to which one believes that using a particular system will be effort-free) are important factors in IT/IS adoption.Perceived usefulness in the TAM model initially refers to work-related productivity, performance, and effectiveness [6].Moreover, perceived usefulness is an example of an extrinsic motive.
Another finding confirms that there is also an intrinsic motive called flow experience (Flow Theory).[7] It was noted that flow experience was associated with positive subjective experience and exploratory behavior.This state of flow often occurs in the various types of interactions involved in the context of online learning or Internet suffering.
In relation to the effectiveness of TAM, empirical studies were carried out by several scholars to investigate the relationship between student behavior, multimedia learning content and multimedia learning systems [12][13][14].Their findings suggest that congruence between student characteristics and multimedia technology results in greater enjoyment while experiential influences may moderate behavioral effects.Other empirical studies have confirmed that there are several other key factors in the TAM model such as consumer age, gender, and experience [15][16][17].
It is specifically mentioned that perceived usefulness and perceived ease of use positively affect consumers' intention to use the Internet.This means that the intuitive user interface can make their job easier and therefore do the job more efficiently.Many studies have also shown that perceived ease of use is likely to increase perceived usefulness [18] [15].This relationship has been further supported by extended TAM [8] [6] [19] [20] [21] [10] [22].Thus, the following hypothesis is proposed: H1: Students' perceived ease of use (SPE) has a positive effect on students' continuance intentions (SCI).H2: Student Perceived usefulness (SPU) has a positive effect on student continuance intention (SCI).H3: Students' perception of ease of use (SPE) has a positive effect on students' perceptions of usefulness (SPU).

Flow Experience
The concept of flow experience which was first proposed by [23] is used as a benchmark to determine an experience in accessing information systems.The flow of experience, is also a representation of a person's perception and experience of the use of information systems.This will give encouragement to someone involved in the system to limit themselves to every activity that exists.[24] Says that flow experience is developed and can be measured and analyzed so that flow experience affects customer satisfaction and a person's behavior in the future.
In online activities, internet users will get a different experience.This will have the impact that internet users must master their own environment as well to get a good experience [25].
In another study it was explained that comfort and concentration are the two most important factors in the flow experience [26].In another study [7] explained that the ease of flow of experience is an important positive factor for e-mail users.[25] also showed that the flow experience and perceived comfort for the user have a strong relationship.Therefore, we hypothesize: H4: Student perceived ease of use (SPE) has a positive effect on student flow experience (SFE).

3
In another study, flow experience can be used to analyze and predict a person's sustainability and intention to carry out online activities [25].Therefore, we hypothesize: H5: Student flow experience (SFE) has a positive effect on student continuance intention (SCI).
From the above explanation, we synthesize the related constructs and propose a research model to explain and predict students' continuance intention towards online multimedia learning adoption (as shown in Figure 1).

Instrument Design
Empirical data were collected through a questionnaire.They are divided into two parts of the user's personal information; age, education, occupation, experience of using multimedia online learning and frequency of use, and theoretical construction of the proposed model to measure user perceptions of continued use of multimedia online learning.A questionnaire with a seven-point Likert scale was distributed to 523 students from January 20, 2016 to April 15, 2016.
The construction and source scale items are: Continue intention: four items adapted from [27] [28].Ease of use and perceived usefulness: both of the four items were slightly modified from [29] and [30] Experience flow: four items were adopted from [25].

Sampling Method
The research used a purposive sampling method as there were 523 valid samples taken from online learning students.Respondents must have a qualification that has participated or are participating in multimedia online learning to facilitate improved external validity.(as shown in Table 1).

Outer model
In PLS, the relationship between indicators and latent constructs is referred to as the outer model [31].To test the reliability of each existing construction (construct item) required Cronbach and composite reliability with a construction value of 0.7 or more, which means that the existing construction is acceptable and reliable.Table 2 shows that Factor Loads and reliability tests are in accordance with predetermined indicators.To test the construct validity, in this study two methods were used, namely the convergent validity test and the discriminant validity test.As the theory put forward by [32] which states that a construction can be declared convergent validity if the load factor value on each indicator has a value greater than 0.5, and the average extracted variance (AVE) value is greater than 0.5, and the composite reliability value is greater than 0.7.Table 2 shows that all constructs are in accordance with the theory proposed by [32].Furthermore, to test the discriminant validity, the formula is used that the square root of the AVE must be greater than the correlation coefficient of the construction [33].In Table 3 and Table 4, each of the existing constructs is in accordance with the validity test criteria.