How factors of personal attitudes and learning environments affect gender difference toward mobile distance learning acceptance
Shu-Sheng Liaw a, Hsiu-Mei Huang b,
a General Education Center, China Medical University, 91 Shiuesh Rd., Taichung 404, Taiwan, email: ssliaw@mail.cmu.edu.tw
b Department of Information Management, National Taichung University of Technology
129, Sec. 3, Saming Rd., Taichung 404, Taiwan, email: hmhuang@nutc.edu.tw Abstract
Mobile technology for learning empowers the shift from traditionally pure instructor-centered classroom teaching to fully learner-centered educational settings. Although mobile learning (m-learning) accessing Internet resources anytime and anywhere and it may cause gender difference toward it; thus the issue of the relationship between gender perspectives and m-learning environments has been a controversial topic. Therefore, the main research objective of the research is to explore gender differences toward APP-based m-learning
environments. Subjects of the study were selected university students who had already used an APP-based m-learning system for one month. A total of one hundred fifty nine valid questionnaires were collected, spilt evenly between female and male Taiwan’s university students. The statistical analyses lead to significant conclusions. Both female and male learners have highly positive perceptions toward m-learning. Additionally, the proposed research model is an acceptable conceptual model for investigating female and male learners’ attitudes. Furthermore, factors of personal attitudes and learning environments affect gender difference toward m-learning acceptance.
Keywords: mobile learning, social network interaction, Social Cognitive Theory, gender difference, learner attitudes, APP-based mobile learning environments
1. Introduction
Advanced computer and communication technologies now allow learners to exploit the potential learning effects of mobile devices for educational purposes. Mobile technology could facilitate human social interaction with each other and access to Internet resources anytime and anywhere. Indeed, the growth of mobile devices; such as smart mobile phones, e-books, and tablet PCs have attracted the attention of educators and researchers (Crompton, 2013; Mcconatha, Praul & Lynch, 2008; Motiwalla, 2007). The massive infusion of computing portable devices and rapidly improved Internet capabilities have influenced to diversify the nature of higher education; especially for distance education. In other words, mobile learning environments are considered as important and impressive factors in achieving the learning competencies and enhancing competitive advantages. It means that m-learning (mobile m-learning) is a pedagogical arsenal to support students as they navigate the options available in the expanding world of distance learning.
On the other hand, the issue of the relationship between gender perspectives and m-learning has been a controversial topic within the debate on educational technology culture in university campuses. In a broader context, gender is considered a basis to explain the inequalities and identities in our society. Jung (2012) confirmed that gender differences in the learners’ perceptions toward educational technology; especially distance learning, should be considered in developing, delivering, and applying educational technology. In the background of computer and communication technologies, the literature recognizes that gender is an element to understand the differences in users’ perceptions and attitudes toward technology adoption; especially for learning purposes. Jung (2012, p21) also stated “Asian DE (Distance Education) providers should consider these gender differences when designing their courses and support systems.”
With regard to m-learning environments, does gender affect how learners adopt mobile devices to provide learning effects efficiently and effectively? Indeed, it is necessary to be investigated in this research. Okazaki and Renda dos Santos (2012) stated gender affected some of the causal relationships among learners’ attitudes (i. e. perceived self-efficacy, and perceived anxiety), social network interaction, perceived usefulness, and learners’ behavioral intention toward e-learning. Previous research (Carr, 2005; Okazaki & Renda dos Santos, 2012; Ong & Lai, 2006) proposed that female learners had greater reliance on ease of use due to their lower level of computer literacy or computer experience when male learners had greater emphasis on perceived usefulness in determining behavioral intention toward learning
technology. Therefore, understanding how gender differences influence causal relationship toward m-learning adoption is a critical issue in distance learning field.
The main research objective of this research is to explore gender difference of attitudes toward APP-based m-learning environments. Specifically, the research is to survey the gender perspectives about female and male university students’ perceptions toward using APP-based m-learning environments. Indeed, based on an approach of social cognitive theory, this research is to figure out gender perspectives of behavioral acceptance toward an APP-based m-learning system. We will investigate causal relationship among learner attitudes that include how personal attitudes (such as self-efficacy, self-regulation, and anxiety) affect learning environmental factors (such as perceived ease of use, usefulness, and social network interaction); we will also investigate how learning environmental factors influence behavioral acceptance. Thus, the following section will review existing body of work on effects of learning, gender issue of technology-based learning, and users’ attitudes toward m-learning. Section three will propose a research model and research hypotheses. Next, section four will introduce the APP-based m-learning system for this research and section five describe our methodology that explains participants and survey questionnaire. The research results will be presented in section six and last two sections will discuss our findings, conclusions and suggest future research. 2. Literature review
2.1 The learning effects of m-learning
Nowadays, with mobile technology, learners are able to learn anything, at any place and any time. The available mobile devices could empower the shift from
traditionally pure instructor-centered classroom teaching to fully constructivist learner-centered educational settings away from the classroom.M-learning is independent of the location of the learner, as the mobile device travels literally with its owner. It means that m-learning usually provides interaction functions to allow learners to control (or search) information flow and/or social communication with each other through handheld devices for distance learning.
The term m-learning has different meanings for different communities; it covers a range of use scenarios including e-learning, online learning, and distance education, as well as focusing on learning with mobile devices (Crescente & Lee, 2011). Mobile devices could offer various available educational
alternatives and advantages. Mobile learners have access to the same personal device on a 24/7 basis; and these types of devices are
permissible for use in a broad variety of environments. M-learning can be defined as "learning across multiple contexts, through social and content interactions, using personal electronic devices” (Crompton, 2013, p.4). Although we do not expect
m-learning to fully replace formal classroom learning or other formal learning approaches; however, m-learning can complement and add value to the existing learning styles or learning models; especially for distance education. Meanwhile, m-learning, like other forms of e-learning, is also collaborative platforms to learners and teachers. Content and instructional information being shared in m-learning
environments is almost instantaneous among everyone using the same content, instruction, or platform, which leads to the reception of instant feedback and tips.
Based on the m-learning research findings of Liaw, Hatala, and Huang (2010), the major five applications of m-learning for learning effects are: educational content delivery application, adaptive learning application, interactive application, individual application, and collaborative application. An educational content delivery application refers to creating m-learning platforms on mobile devices to provide various
educational services for learning. An adaptive learning application tends to design learning environments in which content development tools are built to deliver
learning content adaptively to mobile devices. Indeed, an interactive application tends to create mobile learning systems based on mobile devices, sometimes with other technologies such as personal computers, to increase interaction or communication among students and/or teachers. An individual application provides information and services via mobile devices to learners according to learners’ location and/or their needs. Finally, a collaborative application makes use of the available communication and interaction features of mobile devices to encourage joint learning activities. These five applications make m-learning a platform that could improve learning effects significantly.
2.2 Gender difference toward educational technology
The gender gap in educational devices and technologies has interested educational scholars and researchers. Various perceptual factors associated with gender
differences have been explored in previous research (Albert & Johnson, 2011; Chu, 2010; González-Gómez , Guardiola, Rodríguez & Alonso, 2012; Ong & Lai, 2006). The enormous growth of m-learning makes it necessary to re-examine gender issue that has been the object of research in education. In this context, gender difference is a significant factor to take into account in m-learning environments. In other words, it is reasonable to posit that female and male learners have different perceptions toward m-learning environments. Investigating the gender perspectives in learners’ attitudes towards m-learning more thoroughly can assist researchers and educators in understanding how to
consider gender when developing better m-learning platforms for learners. From previous research, Ong and Lai (2006) stated that men had more positive perceived usefulness affecting behavioral intention toward e-learning than women had.
Furthermore, they also explained that men’s rating of perceptions with respect to computer self-efficacy, perceived usefulness, perceived ease of use, and behavioral intention of using learning systems were higher than those of women. On the other side, research of Guardiola, Rodríguez, and Alonso (2012) showed that female students had higher perceived satisfaction resulting from using e-learning than male students had.
Albert and Johnson (2011) argued that female students have more positive attitudes of perceived self-regulation, perceived ease of use, and perceived user friendliness toward learning systems than male students had. It means female learners had more positive perceptions toward learning technology than male learners. In their research Padilla-Meléndez, Aguila-Obra, and Garrido-Moreno (2013) stated that female learners’ positive attitudes toward using learning systems could be affected by perceived playfulness while male learners’ positive attitudes toward using learning systems could be influenced by perceived usefulness.
After examining previous studies (Albert & Johnson, 2011; Chu, 2010; González-Gómez , Guardiola, Rodríguez & Alonso, 2012; Ong & Lai, 2006), we realize that while some studies stated male learners had more positive attitudes toward computer technology for learning purposes; some other studies argued that female learners had higher preference for applying educational technology to assist their learning
activities. Therefore, in this study, we progress to investigate gender differences in using m-learning technology to further and better our understanding. 2.3 Attitudes toward m-learning based on the SCT approach
The Social Cognitive Theory (SCT) represents a broad approach to
understanding human behavior and advanced capacities for learning (Bandurs, 1986). SCT considers the power of human agency in life’s achievements, beliefs and outcomes. Ancuta (2007) stated that SCT provided a concept for understanding, predicting, and intervening in human behavior. According to SCT, learning is affected by individual experience and individuals learn from the consequences of their own and others’ actions as well as develop new cognitive structures for behavioral actions. Therefore, while individuals’ actions produce positive outcomes, they tend to reinforce the cognitive structures, but when individuals past actions produce negative consequences, they will adjust their cognition to produce more positive outcomes. These produce rules of behavior that serve as guides in future situations. Furthermore, past experience serves to shape future actions via cognitive processes they engage when facing new contexts or environments (Wan, Wang, & Haggerty; 2008).
Based on SCT, a model consists of three reciprocal interactive factors (Figure 1): personal, behavioral, and environmental. These
Person
Environment Behavior
Figure 1: Social Cognitive Theory
three factors constantly affect each other (Ancuta, 2007). The interactions that occur between personal and behavioral factors involve the person’s thoughts and actions. Individuals imitate model behaviors learned by personally observing others, the environment or the media (Wang & Lin, 2007). Additionally, the interactions that occur between behavioral and environmental factors involve both learners’ behavioral reaction toward environments and the learners’ ability to change those environments (Wu, Tennyson, & Hsia, 2010).
The
approach of social cognitive perspective holds that successful self-regulated learners possess higher levels of motivation (personal influences), apply more effective learning strategies (behavioral influences) and respond more appropriately to
situational demands (environmental influences) (Pintrich & Schunk, 2002). From the research of Pintrich and Schunk (2002), the expectancy component may be
conceptualized as the students’ belief that they are capable of performing the task. For instance, self-efficacy is to refer that the students’ perceptions of their capability to execute an action required achieve a particular outcome (Bandura, 1986). Bandurs also found that self-efficacy strongly influences the choice of activity, the effort expended, the willingness to persist and task accomplishment. Based on these results, learner personal factors, such as self-regulation and self-efficacy or even anxiety, will influence environmental factors and behavioral factors.
Additionally, based on the Liaw’s (2008) and Liaw’s and Huang’s (2013) research, learner attitudes toward learning technology could be divided into three different tiers: the learner characteristics tier, the affective and cognitive tier, and the behavioral tier (as shown in Figure 2). The learner characteristics tier evaluates how personal factors (such as self-efficacy or anxiety) affect learner affective and
cognitive factors. The affective and cognitive tier (such as perceived usefulness or perceived social communication) examines how affective and cognitive factors
The learner characteristics tier
The affective and cognitive tier
The behavioral tier
influence learner behaviors. And the behavioral tier is to find how the research model could predict learner behaviors to use technology for learning purpose. The results of Liaw’s (2008) and Liaw’s and Huang’s (2013) research provided an acceptable model to investigate learner attitudes toward e-learning or m-learning.
Figure 2: Investigating learner attitudes toward m-learning based on the three-tier research model
Based on the approach of social cognitive theory and three-tire research approach, we propose our research model (as shown in Figure 3).
Personal factors (i.e. perceived anxiety, self-efficacy, self-regulation, etc.)
Learning Environmental factors (i.e. perceived ease of use, perceived usefulness, social network interaction)
Behavioral factors (i.e. behavioral acceptance)
Figure 3: The research model 3. Research hypotheses
Investigating user attitudes toward learning is a crucial issue for improving m-learning effects. The measurement of m-m-learning must incorporate different aspects of learners’ perceptions to build a useful diagnostic instrument. It means investigating learner attitudes toward learning could facilitate the creation of appropriate learning environments for teaching and learning. But evaluating methods of m-learning measurement cannot be performed using a single linear methodology; therefore, collecting data on learner attitudes toward m-learning requires a multidisciplinary approach (Liaw, 2002, 2008). Liaw et al.(2010) also stated that when investigating learner behavioral acceptance toward m-learning, we need to identify the key predictors such as perceived self-efficacy, anxiety, self-regulation, learning interactive activities, and system quality (such as ease to use and usefulness).
Furthermore, progressively understanding gender difference toward m-learning is the major research objective of this study. As stated before, the previous studies (Albert & Johnson, 2011; Chu, 2010; Liaw, 2002; Liaw, 2007; Ong & Lai, 2006) are not conclusive with respect to the level of positive attitudes toward
information technology and gender. Therefore, based on our research model as shown in Figure 3, we will propose our research hypotheses based on personal factors (such
as perceived self-efficacy, anxiety, and self-regulation) will affect learning environmental factors (such as perceived ease of use, perceived usefulness, social network interaction); additionally, learning environmental factors will positively influence behavioral factor (such as behavioral acceptance).
Regarding self-efficacy, Bandura (1997, p3) asserted: “perceived self-efficacy refers to beliefs in one’s capabilities to organize and execute the courses of action required to produce given attainments “. It is concerned not with the number of skills learners have, but with what learners believe they can do with what they have under a variety of circumstances or situation (Bandura, 1997; Rogers, Conner & Murray, 2008). Previous research found perceived self-efficacy to be positively predicted with achievement-related behaviors such as motivation, effectiveness, or positive attitudes (Bandura, 1986; Liaw, 2008). In m-learning environments, a high degree of perceived self-efficacy leads to improved behavioral retention (Liaw, 2008). Previous research had shown that improved perceived self-efficacy was highly correlated with perceived usefulness of using learning technologies and male learners had more positive
attitudes to communicate with other learners on technology-based learning
environments (Chu & Chu, 2010; Liaw, 2008). Previous research found that learners’ self-efficacy influences their attitudes and ability to acquire skills, choice of activities, and willingness to continue in a course of action when learning in m-learning
environments. Therefore, we present the following hypotheses:
H1: Perceived self-efficacy is a positive predictor on perceived usefulness for both female and male university students.
H2: Perceived self-efficacy is a positive predictor on social network interaction for both female and male university students.
H3: Perceived self-efficacy is a positive predictor on perceived ease of use for both female and male university students.
Perceived anxiety refers to an unpleasant emotional state or condition
characterized by apprehension or worry; specifically an emotional fear of potential negative outcomes (Spielberger, Gorsuch, & Lushene, 1970). When individuals have the negative feelings associated with high levels of anxiety, they usually divert cognitive resources from job or task performance. Previous studies had explored the relationship between computer anxiety and Internet use, and found that computer anxiety was negatively related to students’ use of the Internet (Barbeite & Weiss, 2004; Durndell & Haag, 2002; Kanfer & Heggestad, 1999). Moreover, previous research also found that perceived anxiety had negative relationship with learner motivation, and self-regulation in technology-based learning environments (Tsai, 2009). Besides, perceived anxiety is a negative predictor to influence perceived satisfaction (Sun, Tsai, Finger, Chen, and Yeh, 2008) and has negative relationship
with perceived usefulness in technology-based learning environments either (Sun et al., 2008; Tsai, 2009). Thus, we present the following hypotheses:
H4: Perceived anxiety is a negative predictor on perceived usefulness for both female and male university students.
H5: Perceived anxiety is a negative predictor on social network interaction for both female and male university students.
H6: Perceived anxiety is a negative predictor on perceived ease of use for both female and male university students.
Pintrich (2000, p. 453) defined self-regulated learning as “an active,
constructive process whereby learners set goals for their learning and then attempt to monitor, regulate, and control their cognition, motivation, and behavior, guided and constrained by their goals and the contextual features of the environment”. Indeed, m-learning environments provide students with dynamic, interactive nonlinear access to a wide range of information, as well as to self-directed learning in online
environments (Kramarski & Gutman, 2006). And the environments also can provide flexible interaction and communication to enhance social communication; thus potentially making a powerful self-regulated learning environment (Tsai, 2009). Kramarski and Gutman (2006) stated that self-regulated learners are more active participants in the learning process and have more behavioral motivation in their own learning objectives. Tsai (2009) observed that perceived self-regulation is highly correlated with learners’ perceived usefulness in learning environments. Furthermore, both Kramarski and Gutman (2006) and Roca and Gagne (2008) found a high degree of correlation between perceived self-regulation and high system quality, such as ease of use and friendly interactive functions. Therefore, we present the following
hypotheses:
H7: Perceived self-regulation is a positive predictor on perceived usefulness for both female and male university students.
H8: Perceived self-regulation is a positive predictor on social network interaction for both female and male university students.
H9: Perceived self-regulation is a positive predictor on perceived ease of use for both female and male university students.
TAM (Technology Acceptance Model) is a popular research approach for surveying user attitudes toward information technologies (Davis, Bagozzi, & Warsaw, 1989; Venkatesh & Davis, 2000). The concept of TAM suggested that two specific behavioral beliefs, perceived ease of use (EOU) and perceived usefulness (U),
determined an individual’s behavioral intention to use technologies. Sun et al., (2008) and Cheung and Vogel (2013) found that perceived ease of use was a key predictor in affecting learners’ behavioral intention to use technology and positively impacted
H12 H5
Self-efficacy
Perceived ease of use
Behavioral acceptance Perceived usefulness
Social network interaction Anxiety
Self-regulation
perceived usefulness. They also provided evidence that perceived ease of use is a crucial factor on system’s satisfaction. Therefore, we present the following hypotheses:
H10: Perceived ease of use is a positive predictor on perceived usefulness for both female and male university students.
H11: Perceived ease of use is a positive predictor on social network interaction for both female and male university students.
H12: Perceived ease of use is a positive predictor on behavioral acceptance for both female and male university students.
From the research of Liaw and Huang (2014), they stated that perceived usefulness had high correlation with interactive environments in mobile learning environments. They also argued that perceived usefulness was a key factor to predict behavioral intention; this evidence supports TAM concept. Therefore, we present the following hypotheses:
H13: Perceived usefulness is a positive predictor on social network interaction for both female and male university students.
H14: Perceived usefulness is a positive predictor on behavioral acceptance for both female and male university students.
H15: Social network interaction is a positive predictor on behavioral acceptance for both female and male university students.
Based on the research hypotheses, we propose a conceptual research model shown in Figure 4.
4. System development
This research designs an APP-based system as an assisted m-learning environment. The APP-based m-learning system can be operated at personal
computers or mobile devices for distance learning. Students plug and use the system into their smart phones. The major functions in the m-learning system include various applications that Liaw et al. (2010) suggested: educational content delivery function, adaptive learning function, interactive and sharing functions, system brochure, and email to instructor. The educational content delivery function refers to create learning platforms on mobile devices. The adaptive learning function of learning environments mandates building content development tools that support delivery of learning content to mobile devices. The interactive and collaborative function focuses on creating mobile interaction among students and/or teachers for social communication. We integrated our system with the Facebook (www.facebook.com); therefore, users can use the “LIKE” function of the Facebook when they like other learners’ posted resources or posted opinions.
5. Methodology
Regarding the research methodology, we present participants first and then to introduce the research measurement. In this research, 159 university students
answering questionnaire surveys after using an APP-based m-learning system for one month.
5.1. Participants
To progressively understand learners’ attitudes toward the APP-based m-learning system, the questionnaire collected demographic information and posed questions related to experience using computers and Internet; besides, attitudes toward the APP-based m-learning system also were included in this questionnaire. The paper-APP-based questionnaires were distributed with a cover letter. A total of 159 valid responses were collected in a middle Taiwan’s university, 77 participants are female learners and 82 participants are male students. 14 students are 18 year-old, 105 students are 19-20 year-old, 21 students are 21-22 year-old, 5 students are 23-24 year-old, and 14 students are over 25 year-old.
5.2. Measurement
The questionnaire covers seven factors. The five factors of perceived self-efficacy, perceived anxiety, perceived usefulness, social network interaction, and behavioral acceptance were revised from Liaw et al. (2010) (α = 0.96). Questionnaire items for self-regulation were revised from Liaw and Huang (2013) (α = 0.95). Questionnaire items for perceived ease of use were revised from Sun et al., (2008) (α = 0.90). Based on these studies, questionnaire items were designed to focus on investigating learners’
attitudes toward learning systems. The seven-factor questionnaire covered 30 items, each item using a 7-point Likert scale (1 = “strongly disagree” to 7 = “strongly agree”) (shown in the appendix).
6. Results
6.1 Descriptive statistical analyses
The descriptive statistical results are presented in Table 1 and Table 2. The table 1 presents demographic information of female and male students’ experience of computers, smart phones, and the Internet. In the 7-point Likert item questionnaire, from the Table 2, descriptive statistical results showed that female students and male students both had high positive attitudes toward m-learning use in factors of perceived self-efficacy, self-regulation, ease of use, usefulness, social network interaction, and behavioral acceptance. Although female learners had slightly higher mean values than male learners for six of these factors, there were no significant gender differences indicated by the t-Test. Regarding the perceived anxiety, male learners had slightly higher mean value than female learners; there still is no significant gender difference.
Table 1: Students’ computers, smart phones, and the Internet experience Using computers
per week
Using smart phones per week
Using Internet per week
female male female male female male
Less than 1 hour 0 1 2 4 0 0
1-5 hours 11 10 10 12 2 0
6-10 hours 9 11 17 15 14 4
11-15 hours 17 7 15 13 13 11
16-20 hours 7 8 9 9 7 8
Over 21 hours 33 45 24 29 41 59
Table 2: Mean values
Constructs Mean (female) Mean (male) t-Test significance p Perceived self-efficacy 5.45 5.28 No 0.26 Perceived anxiety 2.91 3.01 No 0.66 Perceived self-regulation 5.40 5.28 No 0.16
Perceived ease of use 5.67 5.51 No 0.29
Perceived usefulness 5.60 5.37 No 0.10
Social network interactive 5.32 5.08 No 0.25
6.2 Reliability analyses
Reliability analyses were the procedure used to ascertain the internal consistency of the measures. According to Hair, Black, Babin, and Anderson (2009), Cronbach’s alpha above 0.6 for exploratory study is considered reliable. In this research, Cronbach’s alpha of female students is 0.942 and all constructs of the range of Cronbach’s alpha are between 0.905 and 0.970. Cronbach’s alpha of male students is 0.940 and all constructs of the range of Cronbach’s alpha are between 0.875 and 0.953. Thus all measures in the study are reliable. Table 3 presented the results of Cronbach’s alpha.
Table 3: Cronbach’s alpha of female and male students Constructs Cronbach’salpha
(female) Cronbach’s alpha (male) .942 .940 Perceived self-efficacy .918 .928 Perceived anxiety .957 .953 Perceived self-regulation .944 .875 Perceived ease of use .970 .928
Perceived usefulness .957 .914
Social network interactive .905 .944 Behavioral acceptance .937 .929 6.3 Relationships between factors
The research model is based on partial least square (PLS). PLS technique was applied to detect relationships among constructs. Therefore, we apply PLS to find relationship among seven factors. Kline (2005) states convergent validity is achieved if loadings of the measures to their respective constructs should be at least 0.60. Table 4 shows the range of male students’ loadings are from 0.741 to 0.972; and the range of female students’ loadings are from 0.866 to 0.982. Therefore, convergent validity is
established. Regarding to the composite reliability (CR) and variance extracted (AVE), CR of all constructs was above 0.6 and AVE were above 0.5; thus further supporting convergent validity (Bagozzi &Yi, 1988; Fornell & Larcker, 1981). According to discriminant validity, the inter-variable correlations (Table 5 and Table 6) were examined and the square-root AVE were higher than correlations suggesting that the construct was more closely related to its own measures; therefore, supporting discriminant validity (Fornell & Larcker, 1981).
Table 4: The loading, CR, and AVE
Loading CR AVE Loading CR AVE Perceived self-efficacy Effi01 0.917 0.948 0.858 0.943 0.954 0.874 Effi02 0.947 0.951 Effi03 0.916 0.910 Perceived anxiety Anxi01 0.956 0.973 0.922 0.946 0.969 0.913 Axni02 0.982 0.954 Anxi03 0.956 0.964 Perceived self- regulation Regu01 0.899 0.959 0.796 0.741 0.905 0.614 Regu02 0.886 0.795 Regu03 0.928 0.777 Regu04 0.915 0.775 Regu05 0.886 0.773 Regu06 0.837 0.837 Perceived ease of use Ease01 0.968 0.981 0.945 0.930 0.954 0.874 Ease02 0.980 0.941 Ease03 0.967 0.933 Perceived usefulness Use01 0.866 0.961 0.804 0.868 0.934 0.704 Use02 0.893 0.870 Use03 0.898 0.823 Use04 0.911 0.789 Use05 0.913 0.842 Use06 0.898 0.840 Social network interaction Soci01 0.956 0.976 0.931 0.938 0.965 0.902 Soci02 0.982 0.972 Soci03 0.956 0.939 Behaviora l acceptanc e Acce01 0.868 0.966 0.827 0.859 0.945 0.740 Acce02 0.895 0.801 Acce03 0.930 0.906 Acce04 0.938 0.898 Acce05 0.920 0.827 Acce06 0.905 0.865
Table 5: The inter-variable correlations of female students
BA EU SNI ANX SE SR USE
BA 0.909
EU 0.659 0.972
ANX 0.108 -0.227 0.203 0.950
SE 0.542 0.610 0.323 -0.163 0.925
SR 0.710 0.691 0.528 0.041 0.463 0.783
USE 0.760 0.765 0.603 -0.068 0.468 0.625 0.986 Table 6: The inter-variable correlations of male students
BA EU SNI ANX SE SR USE
BA 0.861 EU 0.524 0.935 SNI 0.668 0.460 0.950 ANX 0.109 -0.114 0.088 0.961 SE 0.432 0.512 0.158 -0.163 0.935 SR 0.639 0.583 0.498 0.041 0.524 0.892 USE 0.738 0.676 0.581 -0.068 0.607 0.823 0.839 6.4 Hypotheses investigation.
In this research, PLS technique was applied to detect relationships among seven factors (perceived self-efficacy, perceived anxiety, perceived self-regulation, perceived ease of use, perceived usefulness, social network interaction, and behavioral acceptance). After PLS analyses, Table 7 showed the results of 15 hypotheses (including female and male learners). Based on the results for female learners, five hypotheses were not supported that including H2, H4, H5, H8, and H11. The H5 was not supported because of positive prediction. This result indicates that female learners with higher anxiety toward m-learning are more willing to socially interact with other learners. For the male learners, three hypotheses were not
supported including H4, H5 and H12. According to the statistical results, female and male students still have gender difference toward m-learning. The hypothesis 4 was not supported because of positive prediction. This result indicates that learners reporting higher level of anxiety toward learning feel more positively towards m-learning being a useful m-learning tool.
Table 7: The results of 15 hypotheses (including female and male students)
Hypothesis Female students Male students
Prediction β p true β p true
H1 Self-efficacyusefulness .135 <.01 Yes .104 <.01 Yes H2 Self-efficacysocial network >.05 No -.22
9
<.01 Yes H3 Self-efficacyease of use .291 <.01 Yes .268 <.01 Yes H4 Anxiety usefulness >.05 No .043 <.05 No
H5 Anxietysocial network .244 <.01 No >.05 No H6 Anxietyease of use -.20 2 <.01 Yes -.12 5 <.01 Yes H7 Self-regulationusefulness .536 <.01 Yes .311 <.01 Yes H8 Self-regulationsocial network >.05 No .245 <.01 Yes H9 Self-regulationease of use .546 <.01 Yes .476 <.01 Yes H10 Ease of useusefulness .313 <.01 Yes .446 <.01 Yes H11 Ease of usesocial network >.05 No .138 <.01 Yes H12 Ease of useacceptance .216 <.01 Yes >.05 No H13 Usefulnesssocial network .604 <.01 Yes .441 <.01 Yes H14 Usefulnessacceptance .391 <.01 Yes .527 <.01 Yes H15 Social networkacceptance .338 <.01 Yes .361 <.01 Yes 7. Discussions
A better understanding of the gender perspectives toward m-learning can assist researchers and educators to realize how to take into consideration the gender factor when developing m-learning platforms for both female and male learners. Based on the results of Table 1, there is no significant gender difference of learner attitudes toward m-learning in seven factor. In other words, although female and male students both have highly positive attitudes toward m-learning environments; there is no significant gender difference. Unlike previous research of Albert and Johnson (2011) as well as Padilla-Meléndez, et al. (2013), our findings provide evidence that both female and male learners have highly positive attitudes toward m-learning
environments.
On the other hand, from the conceptual model in Figure 4, our statistical results overlaid on our conceptual model (as shown in Figure 5 and Figure 6). The results showed that self-efficacy and self-regulation were better predictive factors for male learners while perceived anxiety was a more contributive factor for female students; especially predicting social network interaction. It means when we integrate social network functions into m-learning systems, perceived anxiety is a significant
predictor for female learning; meanwhile, perceived self-efficacy and self-regulation are significant contributors for male learners. Although ease of use is a significant predictor for perceived usefulness; it is not a contributor for male learners’ behavioral acceptance. Furthermore, for female learners, perceived anxiety is a positive predictor on social network interaction but is a negative predictor on perceived ease of use. For male learners, perceived self-efficacy is a negative predictor on social network interaction, and perceived anxiety can positively predict perceived usefulness as well
.216 244
.135 Self-efficacy
Perceived ease of use R2=.601
Behavioral acceptance R2=.655
Perceived usefulness R2=.763
Social network interaction R2=.425
Anxiety
Self-regulation
Figure 5: Female students’ research model
as negatively predict perceived ease of use. From these statistical results, it could be inferred this is evidence of gender difference toward m-learning. And these findings were not detected from previous studies (Albert & Johnson, 2011; Chu, 2010; González-Gómez , Guardiola, Rodríguez & Alonso, 2012; Ong & Lai, 2006; Padilla-Meléndez, et al., 2013).
-.229
.043 .104 Self-efficacy
Perceived ease of use R2=.429
Behavioral acceptance R2=.631
Perceived usefulness R2=.546
Social network interaction R2=.407
Anxiety
Self-regulation
Figure 6: Male students’ research model
Indeed, based on the research model in Figure 3, we realize personal factors (i.e. anxiety, self-efficacy, and self-regulation) are significant contributors for
environmental factors (i.e. ease of use, usefulness, and social network interaction). However, we found that gender difference occurs here because perceived self-efficacy and self-regulation contributed more for male learners while perceived anxiety was a better predictor for female learners. Furthermore, environmental factors, such as perceived ease of use, usefulness, and social network interaction, were significant predictors for learners’ behavioral acceptance toward m-learning environments, but perceived ease of use is not a predictor on male learners. It seems that male students are more comfortable with using m-learning systems than female students.
8. Conclusions
In this research, we proposed a 7-factor conceptual model for investigating the gender differences of attitudes towards the m-learning system. The factors include perceived self-efficacy, anxiety, self-regulation, ease of use, usefulness, social network
interaction, and behavioral acceptance. We have designed an APP-based m-learning system that was used by undergraduate students in the context of three courses. After the period of one month, we collected questionnaires examining students’ attitudes and performed the analysis. The results lead to following conclusions:
8.1. The SCT is an acceptable theory to understand attitudes toward m-learning. The research is to investigate learner attitudes toward m-learning based on SCT. Indeed, the study focuses on the predictive relationships among personal factors,
environment factors, and behavioral factor. After statistical analyses, we found that personal factors (such as perceived self-efficacy, anxiety, and self-regulation) were predictors on learning environment factors (such as perceived usefulness, ease of use, and social network interaction) as well as learning environment factors were positive predictors on behavioral acceptance. On other words, the SCT is acceptable
theoretical concept to understand learners’ attitudes, both female and male learners, toward m-learning.
8.2. The proposed model is an acceptable conceptual model for investigating female and male learners’ attitudes.
The proposed research conceptual model of Figure 4 is an adaptive conceptual model to survey learners’ (including female and male learners) attitudes toward m-learning. In Figure 3 and Figure 4, learners’ personal factors could predict learning
environmental factors. At the same time, learning environmental factors could also predict behavioral factors. Therefore, our research results support previous research of Liaw (2008) and Liaw and Huang (2013).
8.3. Gender difference is an issue in attitudes toward m-learning acceptance.
After investigating our 15 hypotheses as presented in Figure 4, we may conclude that gender influences attitudes towards m-learning acceptance. Perceived anxiety is a positive significant predictor for social network interaction for female learners while perceived self-efficacy and perceived self-regulation have more positive contribution for male learners. Perceived usefulness has the highest contribution on behavioral acceptance toward m-learning for both female and male learners. These results are aligned with previous studies (Carr, 2005; Okazaki & Renda dos Santos, 2012; Ong & Lai, 2006) claiming that female and male learners have different cause relationship among attitudes toward m-learning.
8.4. Social network interaction is a significant factor to enhance behavioral acceptance.
Social network interaction is a significant predictor on behavioral acceptance toward m-learning for both female and male learners. This statistical evidence supports that social interaction-based m-learning systems are adaptive mobile learning platforms for both female and male learners. The result could support social network interaction as a key success factor to enhance learners’ acceptance toward m-learning.
8.5. Both female and male learners have highly positive attitudes toward m-learning. Both female and male learners have highly positive perception toward applying APP-based m-learning systems for learning purposes. With some previous studies (Chu, 2010; Ong & Lai, 2006) found higher positive attitudes toward educational
technology for male students; other studies (Albert & Johnson, 2011; González-Gómez, et al., 2012) found that female learners prefer applying educational
technology to assist their learning activities more. This research shows that female and male learners have both highly positive perceptions toward applying APP-based m-learning environments.
8.6. Perceived anxiety is a significant predictor for female learners.
Based on the statistical results, perceived anxiety is the most significantly positive predictor on social network interaction for female learners. These results can be explained when female learners have higher perceived anxiety toward m-learning; then they also have higher intention to share their learning experience and ideas. On the other hand, perceived anxiety is a positive predictor of perceived usefulness for male learners. These results indicate that for male learners’ higher perceived anxiety toward m-learning increases their positive feeling toward m-learning as a useful learning tool. Besides, the results show that both for female and male learners perceived anxiety is a negative predictor on perceived ease of use.
8.7. Perceived self-efficacy is a negative predictor on social network interaction for male students.
The statistical results show that perceived self-efficacy is a negative predictor on social network interaction for male students; but perceived self-efficacy is not a predictor on social network interaction for female learners at the same time. The information indicates when male learners have more self-efficacy using APP-based m-learning, then they have less interested in social interaction with other learners. 8.8. Perceived self-regulation has gender difference,
The statistical results provide evidence that perceived self-regulation has gender difference. Based on the Table 7, perceived self-regulation has more contribution on perceived ease of use and usefulness for female university students; while it has more significant impact on social network interaction for male university learners.
In the future research, when investigating gender difference toward educational technology, especially APP-based mobile technology, we may not only focus on learners’ computer and Internet experience or students’ perception toward educational technology, we should also do more efforts to figure out relationship and prediction among factors of learner attitudes that based on gender difference or perspectives. Indeed, based on the social cognitive theory approach, future research could investigate progressively more gender difference on relationship and prediction among personal attitude factors and learning environment factors to understand better about learner behavioral acceptance toward mobile technology.
Acknowledgements:
We appreciate Professor Marek Hatala revised the paper. The research was partial supported by national projects of MOST101-2511-S-039-004-MY2, and
References
Albert, L. J., & Johnson, C. S. (2011). Socioeconomic Status– and Gender-Based Differences in Students’ Perceptions of E-Learning Systems. Decision Sciences
Journal of Innovative Education, 9(3), 421-436.
Ancuta, O. (2007). Social cognitive theory (SCT). Journal of Interation Recipes, Idea,
1(1), 110.
Bagozzi, R. P., &Yi, Y. (1988). “On the evaluation of structural equation models”, Journal of the Academy of Marketing Science, 16 , pp. 74-94.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive
theory. Englewood Cliffs, NJ, Prentice Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman. Carr, D. (2005). Contexts, gaming pleasures, and gendered preferences. Simulation &
Gaming, 36(4), 464-482.
Cheung, R., & Vogel, D. (2013). Predicting user acceptance of collaborative technologies: An extension of the technology acceptance model for e-learning.
Computer & Education, 63, 160-`175.
Chu, R. J. C. (2010). How family support and Internet self-efficacy influence the effects of e-learning among higher aged adults – Analyses of gender and age differences. Computers & Education, 55, 255-264.
Chu, R. J., & Chu, A. Z. (2010). Multi-level analysis of peer support, Internet self-efficacy and e-learning outcomes – the contextual effects of collectivism and group potency. Computers &Education, 55(1), 145–154.
Crescente, M. L., and Lee, D. (2011). Critical issues of m-learning: design models, adoption processes, and future trends. Journal of the Chinese Institute of Industrial
Engineers, 28(2), 111–123.
Crompton, H. (2013). A historical overview of mobile learning: Toward learner-centered education. In Z. L. Berge & L. Y. Muilenburg (Eds.), Handbook of mobile learning (pp. 3–14). Florence, KY: Routledge.
Davis, F. D., Bagozzi, R. P., & Warsaw, P. R. (1989). User acceptance of computer technology: a comparison of two theoretical models. Management Science, 35(8), 983–1003.
Fornell, C. R., & Larcker, D. F. (1981). Evaluating structural equation models with unobserved variables and measurement error. Journal of Marketing Research, XVIII, pp. 39-50.
González-Gómez , F., Guardiola, J., Rodríguez, O. M., & Alonso, M. A. M. (2012). Gender differences in e-learning satisfaction. Computers & Education, 58, 283-290. Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2009). Multivariate Data Analysis, Prentice Hall, NJ.
Jung, I. S. (2012). Asian Learners’ Perception of Quality in Distance Education and Gender Differences. International Review of Research in Open and Distance
Learning, 13(2), 1-24.
Kline, K. B. (2005). Principles and Practice of Structural Equation Modeling. Guilford Press, NY.
Kramarski, B., & Gutman, M. (2006). How can self-regulated learning be supported in mathematical E-learning environments? Journal of Computer Assisted Learning,
22, 24–33.
Liaw, S. S. (2002). An Internet Survey for Perceptions of Computer and World Wide Web: Relationship, Prediction, and Difference. Computers in Human Behavior, 18, 17-35.
Liaw, S. S. (2007). Investigating sex differences and learners’ autonomy toward e-learning: based on UK and Taiwan surveys, Psychological Reports, 100, 949-954. Liaw, S. S. (2008). Investigating students’ perceived satisfaction, behavioral intention, and effectiveness of e-learning: A case study of the Blackboard system. Computers & Education, 51(2), 864-873.
Liaw, S. S., Hatala, M., & Huang, H. M. (2010). Investigating acceptance toward mobile learning to assist individual knowledge management: based on activity theory approach. Computers & Education, 54(2), 446–454.
Liaw, S. S., & Huang, H. M. (2014). Investigating learner attitudes toward e-books as learning tools: Based on the activity theory approach, Interactive Learning
Environments.
McConatha, D., Praul, M., & Lynch, M. J. (2008). Mobile learning in higher education: An empirical assessment of a new educational tool. The Turkish Online
Journal of Educational Technology, 7 (3), 15-21.
Motiwalla, L. F. (2007). Mobile learning: A framework and evaluation. Computers &
Education, 49 (3), 581-596.
Okazaki, S., & Renda dos Santos, I. M. (2012). Understanding E-Learning Adoption in Brazil: Major Determinants and Gender Effects, International Review of Research
in Open and Distance Learning, 13(4), 91-106.
Ong, C. S., & Lai, J. Y. (2006). Gender differences in perceptions and relationships among dominants of e-learning acceptance. Computers in Human Behavior, 22, 816– 829.
Padilla-Meléndez, A., Aguila-Obra, A. R., and Garrido-Moreno, A. (2013). Perceived playfulness, gender differences and tech nology acceptance model in a blended learning scenario. Computers & Education, 63, 306-317.
Pintrich, P. R. (2000). A motivational science perspective on the role of student motivation in learning and teaching contexts. Journal of Educational Psychology, 95,
667–686.
Roca, J. C., & Gagne, M. (2008). Understanding e-learning continuance intention in the workplace: A self-determination theory perspective. Computers in Human
Behavior, 24, 1585–1604.
Rodgers, W. M., Conner, M., & Murray, T. C. (2008). Distinguishing among
perceived control, perceived difficulty, and self‐efficacy as determinants of intentions and behaviours. British Journal of Social Psychology, 47(4), 607-630.
Sun, P. C., Tsai, R. J., Finger, G., Chen, Y. Y., &Yeh, D. (2008). What drives a successful e-Learning? An empirical investigation of the critical factors influencing learner satisfaction. Computers & Education, 50, 1183-1202.
Tsai, M. J. (2009). The model of strategic e-learning: Understanding and evaluating student e-learning from metacognitive perspective. Educational Technology &
Society, 12(1), 34–48.
Venkatesh, V., & Davis, F. D. (2000). A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies. Management Science, 26(2), 186-204.
Wan, Z., Wang, Y., & Haggerty, N. (2008).Why people benefit from e-learning differently: The effects of psychological processes on e-learning outcomes.
Information & management, 45, 513-521.
Wang, S. L., & Lin, S. S. J. (2007). The application of social cognitive theory to web-based learning through NetPorts. British Journal of Educational Technology, 38(4), 600-612.
Wu, J. H., Tennyson, R. D., & Hsia, T. L. (2010). A study of student satisfaction in a blended e-learning system environment. Computers & Education, 55(1), 155–164.
Appendix: Questionnaire: Factor Questionnaire item Questionnaire content Perceived self-efficacy
Effi01 I feel confident using the APP.
Effi02 I feel confident using resources of the APP. Effi03 I feel confident operating APP functions. Perceived
anxiety
Anxi01 I feel it is difficult using the APP. Anxi02 I feel nervous using the APP. Anxi03 I feel using the APP is complicated. Perceived
self-regulation
Regu01 The APP is an individual learning tool. Regu02 The APP is a self-regulated learning tool. Regu03 The APP is an active learning tool. Regu04 The APP is a self-paced learning tool for
individual use.
Regu05 The learning steps are easy to regulate by myself.
Regu06 The learning contents are easy to regulate by myself.
Perceived ease of use
Ease01 The APP functions are easy to use. Ease02 The APP is easy to operate.
Ease03 The content is easy to read. Perceived
usefulness
Use01 The interactive function is useful. Use02 The sharing function is useful. Use03 The social function is useful.
Use05 I will use the APP to find learning resources. Use06 I will use the APP to interact with other
learners. Social
network interaction
Soci01 The APP can enhance social learning effects. Soci02 The APP can improve social sharing.
Soci03 The APP can assist social interaction. Behavioral
acceptance
Acce01 The APP can enhance learning efficacy.
Acce02 The APP can enhance learning performance. Acce03 The APP can enhance learning motivation. Acce04 The APP can enhance learning satisfaction. Acce05 The APP can enhance learning enjoyument. Acce06 The APP can enhance learning fun.
