Please download to get full document.

View again

of 10

Study on a Structural Equation Model of the Effects of E-learning Perceived by Elementary School Students

Study on a Structural Equation Model of the Effects of E-learning Perceived by Elementary School Students Innwoo Park (Korea University) Jin Ho Lim (Korea University) Taewoong Kim (Korea University) Abstract
0 views10 pages
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Documenttranscript
Study on a Structural Equation Model of the Effects of E-learning Perceived by Elementary School Students Innwoo Park (Korea University) Jin Ho Lim (Korea University) Taewoong Kim (Korea University) Abstract The purposes of this study are to create a Structural Equation Model for explaining the factors which influence the effects of e-learning perceived by elementary school students, to examine the model fit of the Structural Equation Model, and to explain the structural relationship among the factors. To this end, the theoretical framework was developed to explain elementary students' perceived effects of e-learning, based on critical analysis of the previous studies. Second, the model fit of the measurement model based on the theoretical framework was tested, using the confirmatory factor analysis and structural equation modeling(sem) analysis. Third, the hypothesized structural equation model was revised and refined as the final model for the most appropriate one. Finally, the paths and effect size of the factors which influence the students' perceived effects of e-learning were analyzed. The result of the confirmatory factor analysis for measurement model pointed out that the fit of the AGFI did not satisfy the statistical criterion. Accordingly, Squared Multiple Correlations(SMC) was conducted, and 6 factors were eliminated to revise the measurement model. The second test for the revised measurement model was conducted. The result indicated that the fit of indices in the SEM were higher than the previous model with χ2=2,401.83, df=979, GFI=0.913, AGFI=0.900, TLI=0.938, CFI=0.943, and RMSEA= Both credibility and validity were examined and turned out to be appropriate. Next, the SEM analysis was also conducted to verify the appropriateness of the hypothesized structural equation model based on the revised measurement model. The result indicated that the entire model was appropriate, but GFI and AGFI still did not satisfy the desired statistical criterion. The result led to finding the more appropriate model than the present model. The hypothesized structural equation model, therefore, was revised according to both the previous studies and the modification indices, using AMOS 5.0. The revised model was eventually decided as final, indicating a more appropriate fit than the previous model. The model fit of the final structural equation model presented χ2=2, (p=0.000), df=999, χ2/df=2.779, GFI=0.900, AGFI=0.887, TLI=0.924, CFI=0.929, and RMSEA= In conclusion, this study presented that the theoretical framework and the final structural equation model were found to be appropriate for explaining the factors which influence elementary students' perceived effects of e- learning. This study also added new findings to the relevant studies in that the characteristics of e-learning environments caused the effects of e-learning in the characteristics of the learners as a parameter. Finally, the findings about the paths and effect size of the factors which influence the effects of e-learning in this study can be used to design e-learning service which is most appropriate to the students' abilities and to understand the processes of the elementary students' learning and perceived effects during e-learning. Introduction The term e-learning started to be widespread in Korean elementary and secondary education after it was used in the Education Ministry s announced measures in 2004 to reduce private tutoring expenses by normalizing public education (The Ministry of Education and Human Resources, 2004). E-learning, therefore, has a relatively short history in the Korean education field. Now there is a move to broaden the concept of e-learning in Korea as the government has been actively working to use e-learning in teaching and learning in elementary and secondary schools (Jang Sang-hyeon et al. 2004; Han Geon-woo, Song Gi-sang, Lee Yeoung-jun, 2005). This has caused varied terms such as cyber learning, online learning, and internet learning, commonly used in elementary and secondary schools, to be all replaced by the term e-learning (Byeon Yeoung-gye, Kim Gyeong-hyeon, 2003). The trend has made e-learning a familiar teaching and learning method for the elementary and secondary education. As part of its policies, the Education Ministry is providing e-learning services nationwide including EBS Lectures for the Scholastic Ability Test and Cyber Home Learning for elementary and secondary school students. These services are aimed at expanding education beyond the classroom into cyberspace; enabling students to learn according to their own levels and have greater right to learning choices; and enhancing the quality of public 241 education by providing students with self-directed learning opportunities that enable them to study whenever necessary. Despite the enthusiasm surrounding e-learning, few efforts are being made to prove the effectiveness of e- learning in elementary and secondary education. Some are even voicing concern about its effectiveness (Han Geonwoo, Song Gi-sang, Lee Yeoung-jun, 2005). Park Innwoo (2004) said that e-learning, by nature, is self-directed learning. This means that a more careful and effective approach is necessary when providing e-learning service to elementary and secondary schools than to universities and companies as schoolchildren are less capable of directing their own learning than adults (Song Sang-ho et al., 2005). Han Geon-woo, Song Gi-sang, and Lee Yeoung-jun (2005) stressed the need to make constant efforts to enhance the effectiveness of e-learning in their research on Ways to improve EBS lectures for the scholastic ability test for the good of the public education, where they argued that the EBS lectures failed to take advantage of e-learning by conducting one-way lectures, and pointed out the lack of lecture information and evaluation. These points indicate the urgency of taking a systematic and comprehensive look at the effects of e- learning used in elementary and secondary education. Theoretical Background Every study has different yardsticks for measuring the effectiveness of e-learning. Many studies focusing on its academic effects adopted learners academic achievement as the major yardstick (Kang Ok-mi, 2001; Kim Seeun, 2002; Lee In-sook, 2002; Moore & Kearsley, 2005; Neuhauser, 2002; Hiltz, 1990), while others used learners satisfaction (Park Seong-ik, Yoon Soon-gyeong, 2000; Thurnond et al., 2002). Some studies took learners academic achievement, participation, and satisfaction into account to explain the effectiveness of e-learning (Kang Yeounghwan, 2004; Seo Hye-jeon, 2001; Lim Jeong-hoon, Lee Hang-nyeong, 2003; Yoo Pyeong-jun, 2003b). This study defines the effectiveness of e-learning as individual learner s comprehensive and subjective perception of the educational experience and benefits gained through e-learning, including their academic achievement, participation and satisfaction. Based on this definition, it examines existing studies and relevant documents for the theoretical background to find that most studies on the effectiveness of e-learning dealt with individual learner s characteristics and e-learning environmental characteristics as factors influencing effectiveness. The findings will be explained in detail below. Kang Myeong-hee (2004) argued that computer-based learning is not suitable to all learners, and that benefits and level of satisfaction can vary according to learners characteristics. This means that learners characteristics determine whether e-learning is effective. Relevant studies focused the individual characteristics that affect the effectiveness of e-learning. The characteristics identified include learning motivation, achievement motivation, academic self-efficacy, computer self-efficacy, learners familiarity with technology, learners desire and ability to direct their own study, learners educational background or preliminary education, learners personality, learner s interest, time management strategies, positive attitude, test anxiety, cognitive strategies, and meta-cognitive strategies (Kang Ok-mi, 2001; Kim Se-eun, 2002; Song Sang-ho, 2000; Yang Yeon-sook, Yoo Pyeong-jun, 2003; Eom Woo-yong, Choi Eun-hee, 2001; Lee In-wook, 2002, 2003; Lee Ji-yeon, 2004; Lim Jeonghoon, Lee Hyang-nyeong, 2003; Yoo Pyeong-jun, 2003a; Cho Il-hyeon, Lim Gyu-yeon, 2002; Ju Yeoung-ju, Moon Ja-yeoung, 2004; Compeau & Higgins, 1995; Hasan, 2003; Houle, 1996; Kelsey, Lindner & Dooley, 2002; Moore & Kearsley, 2005; Oliver & Shapiro, 1993; Torkzadeh & Van Dyke, 2002). Other studies found that the effectiveness of e-learning is dependent on e-learning environmental characteristics (Song Sang-ho, 2004). They focus mainly on design and operational elements of e-learning, and identify elements including composition of learning contents, learner support, convenience or ease of use, interaction such as teachers feedback and operators social role, user interface, teachers capability, various evaluation methods, preliminary education, interaction between learners, accessibility, and physical and psychological support (Park Seong-ik, Yoon Soon-Gyeong, 2000; Seo Shin-seok, 2003; Seo Hye-jeon, 2001, Lee Jiyeon, 2004; Cho Il-hyeon, Lim Gyu-yeon, 2002; Choi Jeong-im, 1999; Harasim, 1986; Kelsey, Lindner & Dooley, 2002; Thurmond et al., 2002). Purpose of the Paper Although there are a number of theories and studies on the effectiveness of e-learning, few studies take a comprehensive look at various e-learning factors such as learners profile, general learning-related elements, webbased learning-related elements, learners satisfaction, learners academic achievement, and learners participation. As a result, Lee In-sook (2003) and Yoo Pyeong-jun (2003b) strongly raised the need to conduct comprehensive research taking all e-learning factors into account. This study aims to design a comprehensive theoretical model to explain the effectiveness of e-learning based on existing studies, which revealed the effectiveness of e-learning 242 according to learners characteristics and environmental characteristics, and to verify the model by measuring paths between factors and effect size to identify factors contributing to the perceived effects of e-learning. This will help us understand the elements that must be considered to enhance the effectiveness of e-learning. Research Questions The purposes of this study are to design a model that can explain factors affecting the perceived effects of e-learning; to identify a structural relationship between those factors; and to assess the model fit of the structural equation model. To that end, it has set up the following research tasks. Research Task Ⅰ: To design a theoretical structural model of factors influencing the effects of e-learning perceived by elementary school students based on existing studies Research Task Ⅱ: To devise a provisional structural equation model based on the theoretical model, and test the model fit of the equation model Research Task Ⅱ-1: To decide whether the provisional structural equation model based on the theoretical model is the optimum model or not, and, if not, to identify the optimum model Research Task Ⅱ-2: To discover whether hypothetical paths between factors affecting the perceived effects of e-learning are significant The Model This study used the Input-Environment-Outcome (I-E-O) model by Astin (1993) as the conceptual basis for a structural model of factors influencing the perceived effects of e-learning. The I-E-O model has been used by many researchers to verify the relationship between learner s characteristics, e-learning environmental characteristics, and e-learning effectiveness (Astin & Sax, 1998; Campbell & Blakey, 1996; House, 1999; Kelly, 1996; Knight, 1994a, 1994b; Long, 1993; Pace, 1976). Adopting the ideas of Astin (1993), the study defined input as the characteristics of individual learners participating in educational programs and environment as the practical experiences and activities educational programs provide for students. Based on existing studies, outcome was defined as the learners academic achievement and satisfaction through participation in learning (Kang Yeoung-hwan, 2004; Koo Gyo-jeong, 2005; Seo Hye-jeon, 2001; Yoo Pyeong-jun, 2003b; Lim Jeong-hoon, Lee Hang-nyeong, 2003). 243 Figure 1 shows the theoretical structural model indicating the elements of each factor and hypothetical paths between them. Figure 1. Theoretical Structural Model Measurement Issues Table 1 shows the composition and sources of a survey tool to analyze a structural model. The questionnaire has three factor categories e-learning environmental characteristics, learners characteristics, and e- learning effectiveness. The e-learning environmental characteristics category consists of four parts, which as are the same as the four elements composition of learning contents, learner support, usability, and interaction. The learners characteristics category also consists of four parts intrinsic motivation, extrinsic motivation, academic self-efficacy, and computer self-efficacy and the e-learning effectiveness category three parts learners academic achievement, learners participation, and learners satisfaction. 244 Table 1. Composition and Sources of Survey Tool Category Elements Sources E-learning environmental characteristics Learners characteristics E-learning effectiveness Composition of learning contents Seo Hye-jeon (2001) Usability Yoo Il, Hwang Jun-ha (2002) Learner support Interaction Stewart, Hong & Strudler (2004) Intrinsic motivation Velayo (1993); Stein (1997); Extrinsic motivation Chung Jae-sam, Lim Gyu-yeon (2000) Academic self-efficacy Yang Myeong-hee (2000), Lee Jae-gyeong (2000) Computer self-efficacy Al-Khaldi & Al-Jabri (1998); Oh Yoon-jin (1999) Learners academic achievement Lim Jeong-hoon, Chung In-seong (1999); Seo Hyejeon (2001) Learners participation Yoo Pyeong-jun (2003b); Chung Jae-sam, Lim Gyuyeon (2000) Learners satisfaction Lim Jeong-hoon, Chung In-seong (1999); Seo Hyejeon (2001) The questionnaire is composed of 53 questions measured 1 to 5 according to a Likert-type scale, with 1 meaning strongly disagree ; 2 disagree ; 3 neither agree nor disagree; 4 agree ; and 5 strongly agree. The questions were created based on existing studies to verify a provisional structural model devised on the basis of a theoretical structural model. In relation to the question items, a validity test by experts and a reliability test were conducted. For the reliability test, a preliminary survey was also carried out on 130 sixth-graders in elementary schools in Seoul from Apr. 5-7, 2006, to compute Cronbach s α of the questionnaire. The reliability of the questionnaire was.94, which is fairly high. Data Analysis The survey for the study was conducted on 1,500 sixth-graders in seven elementary schools in Seoul and Gyeonggi Province Apr , Some 82.7 percent, or 1,241, out of the 1,500 questionnaires distributed were collected, and 1,154 of them were found valid for statistical analysis. With the collected data, a confirmatory factor analysis of a measurement model based on the theoretical structural model was carried out using AMOS 5.0. The analysis included the assessments of the model fit index and construct reliability, and calculation of the variance extracted index to evaluate if the scales of the measurement model are representative of the factors concerned. A discriminant validity test was also conducted using a correlation matrix analysis of constructs. An analysis of a measurement structural equation model based on the measurement model showed that AGFI, a model fit index, was not high enough to be valid, and as a result, six measurement variables were removed using squared multiple correlations, modifying the measurement model. A model fit test of the modified measurement model found that most of the model fit indices improved in the measurement structural equation model with the value of χ 2 =2,401.83, df=979, GFI=0.913, AGFI=0.900, TLI=0.938, CFI=0.943, and RMSEA= The reliability and validity of the measurement model were also assessed to be valid. All the results of the confirmatory factor analysis were valid enough to next conduct an analysis of a structural equation model, which was carried out for a provisional model based on the measurement model modified. A model fit test of the provisional model showed a fairly good fit, but the values of GFI and AGFI were lower than acceptable, raising the need for a better model than the provisional model. After checking the modification index of the provisional model and the theoretical background of existing studies, two paths indicating a direct relationship between interaction and learners participation, and between composition of learning contents and learners participation, were added. 245 Table 2. Results of Model fit Test of Provisional Model and Modified Model Model x 2 d.f P x 2 /df GFI AGFI TLI CFI RMSEA Provisional model (1) 2, , Modified model (2) * 2, Model 1: Model * Modified model = Provisional model + (Interaction Learners participation, Composition of learning contents Learners participation) A model fit test of the modified provisional model showed an improvement of model fit with the value of χ 2 =2, (p=0.000), df= 999, χ 2 /df=2.779, GFI=0.900, AGFI=0.887, TLI=0.924, CFI=0.929, and RMSEA= Thus, the modified model was chosen as the final model. Figure 2 shows the final modified model and path coefficients. Figure 2. Final modified model and path coefficients Discussion Efforts to raise the effectiveness of e-learning are needed not only in elementary and secondary education but also in university and corporate education. Many studies have therefore been conducted to assess the effectiveness of e-learning and thus provide an effective e-learning service. This study aims to design a model explaining the effects of e-learning perceived by elementary school students and assess the paths and effect size of various factors influencing the effectiveness of e-learning. This is expected to promote in-depth understanding of how elementary school students learn in e-learning and enable creation of an e-learning service that suits the needs of child learners. The following conclusions can be reached based on the study results. First, the final structural equation model was proved valid to explain factors influencing the effects of e- learning perceived by elementary school students through the confirmatory factor analysis and structural equation model. The model fit test of the final model showed a fairly good fit with the value of χ 2 =2,776.39(p=0.000), df=999, χ 2 /df=2.779, GFI=0.900, TLI=0.924, CFI=0.929, and RMSEA= This means that the final structural equation model is a proper structural model to explain the effects of e-learning perceived by elementary school students. 246 Second, unlike models in past studies, the model in this study showed that e-learning environmental characteristics influence the effectiveness of e-learning through the mediation of learners characteristics. For example, in the composition of learning contents, an element of the e-learning environmental characteristics that affect academic achievement, the effect size was significant at.447, even when taking into account indirect effects alone that take place through the mediation of intrinsic motivation, computer self-efficacy, and learner s participation. Learner support also showed a significant effect size of.228 toward academic achievement, even taking into account indirect effects alone that took place through the mediation of intrinsic motivation and computer self-efficacy. These results indicated that e-learning environmental characteristics can have indirectly influence the effectiveness of e-learning through the mediation of learners characteristics. This study is distinguished from past studies by its interest in the mediating effects of e-learning environmental char
Advertisement
MostRelated
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x