Web-based learning programs: Use by learners with various cognitive styles

Web-based learning programs: Use by learners with various cognitive styles Ling-Hsiu Chen Department of Information Management, Chaoyang University of Technology, 168, Jifong East Road, Wufong, Taichung County 413, Taiwan


http://ac.els-cdn.com.cmich.idm.oclc.org/S0360131509002930/1-s2.0-S0360131509002930-main.pdf?_tid=1b4c91da-8fff-11e6-bc81-00000aab0f27&acdnat=1476223831_6f2e8e0fc78d8f716b57d3854a3688db

abstract

To consider how Web-based learning program is utilized by learners with different cognitive styles, this study presents a Web-based learning system (WBLS) and analyzes learners’ browsing data recorded in the log file to identify how learners’ cognitive styles and learning behavior are related. In order to develop an adapted WBLS, this study also proposes a design model for system designers to tailor the preferences linked with each cognitive style. The samples comprise 105 third-grade Accounting Information System course students from a technology university in central Taiwan. Analytical results demonstrate that learners with different cognitive styles have similar but linear learning approaches, and learners with different cognitive styles adopt different navigation tools to process learning. 2009 Elsevier Ltd. All rights reserved.

article info Article history: Received 13 June 2009 Received in revised form 7 October 2009 Accepted 13 October 2009 Keywords: Human–computer interface Navigation Cognitive style

1. Introduction Technological advances in information and network technology have made the transition from conventional classroom-based to virtual learning space. On-line learning is a new learning domain, while Web-based learning has been predicted as the main distance learning approach in the future (Harasim, 2000). Many higher education institutions have invested heavily in on-line learning and offer Web-based learning programs. Additionally, a growing number of companies also adopt Web-based learning system (WBLS) to train their employees. Traditional class-based learning typically guide the learners to follow one fixed learning process it is less flexible to reflect learns’ different needs. Unlike traditional class-based learning Web-based learning programs can offer many powerful personalized/adaptive mechanisms to suit learners’ requirements. Multiple navigation tools are the most popular adaptive mechanism provided in WBLS. The most common navigation tools designed in Web-based learning programs are hierarchical maps, alphabetical indexes, back–forward buttons, main menus, find topic and ‘‘where I am” functions (Chen & Macredie, 2004). These navigation tools provide various functions to help different learners to construct their knowledge in their own way. On the other hand, learners using WBLS may have diverse backgrounds, preferences and skills. To aid more efficient learning, WBLS must understand and identify different requirements and provide personalized services that can accommodate the learners’ needs. The essential element of developing personalized environments is learner modeling in which empirically collecting and evaluating the influence of individual differences on learners’ navigation behavior (Zukerman, Albrecht, & Nicholson, 1999). Among all individual differences cognitive style has been recognized as particularly relevant to learners’ interaction with the WBLS (Chen & Macredie, 2004). Cognitive style, particularly field-dependence/field-independence (FD/FI), is one of the most widely investigated factors of individual differences in research of on-line learning. FD/FI refers to the perceptual dependence and independence on the framework of the prevailing visual field (Witkin, Moore, Goodenough, & Cox, 1977). The tendencies to rely primarily on external or internal references play a key role in the learning progress and restructuring information (Messick, 1976). Previous researches argued that learners with different cognitive styles present different properties in their learning approach (Chen & Macredie, 2004; Chou, 2001; Davis & Cochran, 1989; Ford & Chen, 2001; Frank & Keane, 1993; Goodenough, 1976; Jonassen & Grabowski, 1993). Several studies also indicated that matching learners’ cognitive styles with the design of Web-based learning programs is an important thing for learning performance. While, previous studies employed questionnaires to identify the relationships between cognitive styles and navigation behavior in the Web-based context. Those studies lack direct information (real usage behavior) to understand the real navigation behavior and learning strategies of learners. Accordingly, this study aims to uncover how cognitive styles influence learners’ learning approach in WBLS by analyzing learners’ browsing records and propose a design model to provide personalized services for different learners. 0360-1315/$ - see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.compedu.2009.10.008 E-mail address: ling@cyut.edu.tw Computers & Education 54 (2010) 1028–1035 Contents lists available at ScienceDirect Computers & Education journal homepage: www.elsevier.com/locate/compedu 2. Theoretical rationale and hypotheses 2.1. Web-based learning system The IEEE Learning Technology Standard Committee define a Web-based learning system as a learning technology system that utilizes Web-browsers as the primary approach of interaction with learners, and the Internet as the primary strategy of communication among its subsystems and with other systems. Web-based learning offers the following benefits over conventional classroom-based learning including: it can be used at any time and place; the learning material is easy to update; it fosters the interaction between the learner and the teacher in several ways; it can incorporate multiple media such as text, audio, graphics, video and animation; it enables learners to form learning communities; facilitators can easily check learners progress, and it allows for a learner-centered approach that can address the many differences between learners (Jolliffe, Ritter, & Stevens, 2001). To add value, a WBLS must provide personal service to meet the different users who use it. Previous research has indicated that human factors have a significant influence on users’ interaction with Web-based instruction in the WBLS. Among all human factors cognitive style (Chen & Macredie, 2004) has been recognized as particularly relevant to users’ interaction with the WBLS. 2.2. Cognitive style Cognitive style is an individual’s preferred and habitual approach to organizing and representing information (Riding & Rayner, 1998). These preferred modes of processing information are individually different, but consistent (Witkin et al., 1977). Therefore, cognitive style is one of the commonly studied measures of individual differences (Oughton & Reed, 1999). Among the different measuring approach to identifying cognitive style, ‘‘field-dependence (FD)/field-independence (FI)” is one of the widely studied cognitive style employed in education research (Messick, 1976; Witkin et al., 1977). Individuals with different cognitive styles present different properties in their learning strategies. Field-dependent individuals typically rely more on external references, meaning that they are more influenced by format-structure (Jonassen & Grabowski, 1993), and prefer to be guided in their learning processes (Chou, 2001). Additionally, FD individuals tend to adopt a Breadth-first path to process their information. In other words, FD individual concentrate first on building an overall picture of the subject area, then consider detail (Ford & Chen, 2001; Goodenough, 1976). Therefore, FD individuals have a tendency to undertake global and passive learning strategies, since they are influenced by format-structure, and need salient cues in learning (Chen & Macredie, 2004; Chou, 2001; Davis & Cochran, 1989; Frank & Keane, 1993; Goodenough, 1976). Conversely, FI individuals rely more on internal references; are less affected by format-structure, and show ability to think for themselves (Goodenough, 1976; Jonassen & Grabowski, 1993). FI individuals tend to apply a Depth-first path to process their information, focusing first on individual parts of the object and on the detail of one topic at a time, and building up procedural understanding step by step. That is, FI individuals prefer to employ analytical and active learning approaches (Chou, 2001; Frank & Keane, 1993). Previous investigations in cognitive styles have also indicated that learning was significantly better in matched conditions than in mismatched conditions. Ford (1995) conducted an empirical study on the effect of matching and mismatching on students’ learning performance, and found that students in conditions that matched their cognitive styles obtained higher test score than in conditions mismatched with their cognitive style. Ford and Chen (2001) designed two hypermedia learning systems (Breadth-first and Depth-first) to explore the relationship between matching and mismatching instructional presentation style and students’ cognitive styles in a computer-based learning context. Their results showed that performance was significantly better in matched conditions than in mismatched conditions. In both these studies (Ford, 1995; Ford & Chen, 2001), FD students performed better in the Breadth-first version (Holist condition), and FI students obtained higher test scores in the Depth-first version (Serialist condition). Fullerton (2000) concluded that FD learners obtained lower score than FI and intermediate learners in a condition mismatched with their preferred manipulation. In examining the relationship between reference strategy and cognitive styles, Lee (2000) discovered that FD learners’ performance deteriorated when they received an instructional strategy that contradicted with their cognitive style. Therefore, matching learners’ cognitive styles with the design of Web-based learning programs is an important factor for learning outcome. 2.3. Developing a rationale for the hypotheses Not all learners learn in the same way: different learners with different cognitive styles utilize different learning and navigation strategies. Liu and Reed (1995) found that students with different cognitive style have significant differences in media choices. Ford and Chen (2000) examined the relationships between learners’ cognitive styles and navigation strategies in a hypermedia system, and discovered that learners’ cognitive styles affect their usage of navigation tools. Non-linear learning is the major dimension that determines learners’ cognitive styles (Lee, Cheng, Rai, & Depickere, 2005). FI learners are better able than FD learners to set their own learning paths. These individuals enjoy independent learning, and like to jump freely from one point to another to understand the learning content and to construct information (Chen & Macredie, 2004; Lee et al., 2005). Accordingly, FI learners tend to adopt non-linear approaches in Web-based instructional programs (Chang, 1995; Durfresne & Turcotte, 1997; Reed & Oughton, 1997). In contrast, FD learners prefer to be guided in their learning, and to apply linear approaches to learn (Chou & Lin, 1997). These individuals tend to follow the sequence provided by the program, and do not like to jump from one subject to another subject while using a WBLS (Chang, 1995; Durfresne & Turcotte, 1997; Reed & Oughton, 1997). Hence, the first hypothesis of this research is proposed as: Hypothesis 1. Learners with different cognitive styles adopt different Web-based learning approaches. In addition to the difference in non-linear/linear learning approach, FD/FI learners also have different preferences for navigation tool usage. Chen and Macredie (2004) undertook a paper-based questionnaire to investigate the relationships between learners’ cognitive styles and their perceptions and attitudes toward the Web-based instructional program. The results revealed that students with different cognitive styles significantly favored different navigation tool types. Chen and Ford’s (1998) results showed that FD students used the main L.-H. Chen / Computers & Education 54 (2010) 1028–1035 1029 menu more often than FI students in a Web-based learning program. FI learners are likely to conduct an active, analytical approach to learning, and prefer to focus on understanding local detail of each individual element. Thus FI learners may prefer freedom of navigation, and like to use navigation tools that can support them to reach and engage specific information faster. For instance, keyword searching and indexing are navigation tools that match FI learners’ learning strategies. FD learners are easily lost in hyperspace; they prefer to use passive, global learning approaches, and like to take an overview of all material before introducing detail. These learners tend to adopt navigation tools that can show the whole picture of context as an anchor for them to organize information (e.g. ‘‘hierarchical map” and main menu) (Chen & Macredie, 2002; Chou & Lin, 1997). Thus, the following hypothesis is proposed: Hypothesis 2. Learners with different cognitive styles utilize different navigation tools provided by the Web-based learning program. 3. Methodology First, a Web-based course was created to test the proposed hypotheses using quantitative data obtained from the log file of the Webbased learning system. This Web-based course was implemented on a Windows XP system, with PHP 4.3 as the front-end script language, and MySQL as the database server. The Web-based learning system architecture is shown in Fig. 1. In Fig. 1 learners log in the Web-based learning system and then check with their accounts stored in learner account database. After learners log in the Web-based learning system successfully, learners can use proper navigation tools to search for course materials that they need or are interested in. The number of times for each navigation tool usage and non-linear learning scores are computed and recorded in the log file at the same time. Fig. 2 shows the content of the log file. 3.1. Participants This study was performed in the Department of Information Management at Chaoyang University of Technology. A total of 105 undergraduate students participated in this study. All participants had the fundamental computing and Internet skills necessary to operate a Web-based instructional program. All participants were initially inexperienced in the content domain of Accounting Information System, and all volunteered to take part in the experiment. 3.2. Research instruments 3.2.1. Learning materials The course unit was a third-year Accounting Information System (AIS) course about the role of the accounting subsystems in a business information system. The topics included an overview of accounting data processing, system development overview, file processing and the database approach, revenue cycle, expenditure cycle, conversion cycle and financial cycle. Each topic also included many sub-topics. Fig. 3 illustrates the entire layout of the user interface. The top frame shows the course unit and course materials. When a learner clicks on a course material, the content of the selected course material is presented in the middle window. The top-right frame contains four navigation tools, namely ‘‘find topic”, ‘‘alphabetical index”, ‘‘hierarchical map” and ‘‘where I am” to enable learners can learn according to their preferred learning styles. The ‘‘find topic” and ‘‘alphabetical index” functions allow learners to reach specific information quickly. The ‘‘hierarchical map” and ‘‘where I am” show the user’s position in the context of the entire learning process, and serve as anchors for learners. Additionally, learners can use the ‘next page’ and ‘previous page’ buttons, at the bottom frame of each learning material page, to process learning. The teaching–learning approach in this program is based on the concept of self-organized learning. Namely, students were given the freedom to select their own routes and tools for navigating through the subject matter. Students could study the topics and sub-topics in any order. The program started by introducing its learning objectives and explaining the available navigation tools; it was presented in 235 pages, each coded with a page number. 3.2.2. Cognitive Styles Analysis Riding’s Cognitive Style Analysis (CSA) was applied to identify the learners’ cognitive styles. The CSA classifies individuals as fielddependent or field-independent according to their Wholist/Analytic (WA) scores (Riding & Sadler Smith, 1992). The test includes two subtests, each with different purposes. The first subtest presents items containing pairs of complex geometrical figures that the individual is Learner login Learner account Database Learner Learning Material Database Learning material browsing Log file Fig. 1. System architecture of the Web-based learning system. 1030 L.-H. Chen / Computers & Education 54 (2010) 1028–1035 required to judge as either the same or different. This task requires field-dependent capacity. The second subtest involves judging whether a simple shape is contained in the complex one. This task requires the disembedding capacity associated with field independence. This study followed the recommendations of Riding (1991), where WA scores of 1.36 and above denote field independence; scores below 1.03 indicate field-dependence individuals, while scores between 1.03 and 1.35 are intermediate. Table 1 presents the overall range of Serial account module_id start_date start_time end_date end_time action page_no. Fig. 2. The illustrated content of log file. Fig. 3. Main screen of the Web-based learning program. L.-H. Chen / Computers & Education 54 (2010) 1028–1035 1031 the WA scores in this study. According to Table 1, most of the learners (n = 41) are intermediate (WA score: Mean = 1.170, SD = 0.102), onethird of learners (n = 35) are FIs (WA score: Mean = 1.993, SD = 0.628), and 29 learners are FDs (WA score: Mean = 0.792, SD = 0.232). 3.2.3. Learners’ learning behavior The data gathered from the log files of learners were analyzed to measure learning behavior (learning approach and navigation tools usage) in the Web-based learning system. Learning approach: This study represents learners’ learning approaches in terms of degree of non-linearity. Non-linear learning (NL) is the degree of learners jumping among each learning material. A higher NL value indicates a learner with higher degree of jumping (more non-linear) among learning materials, and a lower NL value signifies that the learner applies a linear approach to navigate learning material. Each learner’s NL score was computed as: NL ¼ Xm k¼1 Jk=m where m is the total number of navigation time, Jk represents the jumping ration of time kth, and Jk is illustrated as follows: Jk ¼ Xn x¼2 jpx px1j=n 1 where n is the number of navigation pages of time kth, and px is the page number in navigation xth. Navigation tools usage: Learners’ navigation tools usage was represented by the following four variables (Table 2). Table 1 The WA scores in this study. Mean SD Minimum Maximum Field-dependent (FD) (N = 29) 0.792 0.232 0.1 1.0 Field-independent (FI) (N = 35) 1.993 0.628 1.36 3.92 Intermediate (IM) (N = 41) 1.170 0.102 1.04 1.34 Overall 1.340 0.621 0.1 3.92 Table 2 The formulas of learners’ navigation tools usage. Variable Definition Formula Find topic usage (AFT) Average number of times individual user used ‘‘find topic” tool AFT ¼ Pm i¼1FTi=m ‘‘Alphabetical index” usage (AAI) Average number of times individual user used ‘‘alphabetical index” tool AAI ¼ Pm i¼1AIi=m ‘‘Hierarchical map” usage (AHM) Average number of times individual user used ‘‘hierarchical map” tool AHM ¼ Pm i¼1HMi=m ‘‘Where I am usage” (AWIM) Average number of times individual user used ‘‘Where I am” tool AWIM ¼ Pm i¼1WIMi=m FTi is the total using numbers of ‘‘find topic” tool in time i. AIi is the total using numbers of ‘‘alphabetical index” tool in time i. HMi is the total using numbers of ‘‘hierarchical map” tool in time i. WIMi is the total using numbers of ‘‘Where I am” tool in time i. m is the total number of time login this learning system. Table 3 The summary of descriptive statistics. Navigation behavior CS Mean SD Minimum Maximum Non-linear interaction FD 2.17 3.24 0 32.33 FI 3.12 6.04 2.16 95.5 IM 2.54 3.47 1.39 53.2 ‘‘Find topic” usage FD 0.95 0.56 0 3.33 FI 1.48 5.67 0 10 IM 0.81 0.52 0 3.33 ‘‘Alphabetical index” usage FD 0.38 1.08 0 3 FI 4.25 11.71 0 6 IM 0.73 6.50 0 3.3 ‘‘Hierarchical map” usage FD 2.98 6.75 0 10.35 FI 0.95 2.50 0 3.33 IM 0.81 2.77 0 4.21 ‘‘Where I am” usage FD 2.99 8.08 0 22.33 FI 0.94 2.37 0 2.68 IM 0.41 1.53 0 10.56 FD: field-dependent; FI: field-independent; IM: intermediate. 1032 L.-H. Chen / Computers & Education 54 (2010) 1028–1035 4. Results Table 3 lists the descriptive statistics for the non-linear learning approach and navigation tools usage. According to Table 3, FI learners tend to adopt non-linear approaches to understand the learning materials (Mean = 3.12), while FD learners prefer to learn by using a linear way. Furthermore, FI learners like using ‘‘alphabetical index” (Mean = 4.25) and ‘‘find topic” (Mean = 1.48) on local specific learning materials. FD learners prefer to use ‘‘where I am” (Mean = 2.99) and ‘‘hierarchical map” (Mean = 2.98) to help them to learn. 4.1. Interaction pattern in different cognitive style The differences in learners’ cognitive styles and their pattern of interaction with the Web-based learning system were analyzed. ANOVA and Scheffe were applied to measure differences among three groups of non-linear learning. Table 4 shows the analytical results, which reveal that learners with different cognitive styles had no difference in their pattern of non-linear learning (F = 1.577, p = 0.212). Accordingly, H1 is not supported. 4.2. Navigation tools usage in different cognitive style Table 5 presents the average frequency that learners with different cognitive styles employed navigation tools. The table data show that FI learners adopted ‘‘find topic” and ‘‘alphabetical index” most frequently, while FD learners preferred to use ‘‘hierarchical map” and ‘‘where I am”, and Intermediate learners were least likely to use the ‘‘find topic” and ‘‘hierarchical map” functions than FI and FD learners. Two different groups of navigation tools were applied to explore the influence of learners’ cognitive styles on learners’ navigation behavior. The MANOVA revealed a significant difference in the use of ‘‘alphabetical index” (F = 3.557, p < 0.1), ‘‘hierarchical map” (F = 2.671, p < 0.1) and ‘‘where I am” (F = 2.887, p < 0.1) on learners with different cognitive styles, according to Table 5. FI learners utilized ‘‘alphabetical index” significant more frequently than FD and intermediate learner. The FD learners applied the ‘‘hierarchical map” and ‘‘where I am” functions significantly more frequently than FI and intermediate learners. Accordingly, H2 is supported. 5. Discussion The findings of this study provide details and direct information of the navigation behaviors of learners with different cognitive style within the Web-based learning program. Analytical results demonstrate that most learners adopt linear learning approaches to understand the learning materials, regardless of cognitive style. Several previous investigations have found that FI students are more likely than FD students to enjoy non-linear approaches to exploring topics of interest and restructure idea, while FD students prefer to follow linear programs (Chang, 1995; Chen & Macredie, 2002, 2004; Lee et al., 2005; Reed & Oughton, 1997). Their findings imply non-linear learning is the primary factor in indicating students’ cognitive styles. In contrast to those findings, this study demonstrates that learners with different cognitive styles do not present any different in their learning approach for non-linear navigation. One possible reason for this inconsistency is the application of different measurements to discover students’ approaches to interaction in Web-based instruction programs. Whereas this study analyzed learners’ non-linear learning behavior based on their browsing behavior recorded in log file directly, previous studies discovered students’ learning approaches using a questionnaire. The indirect data applied in those studies might reflect students’ perceptions or attitudes toward Web-based learning program, rather than their real behavior. The second reason for this inconsistency might be that this study neglected other factors that significantly affect on learners’ interaction with Web-based learning program. Sex (Ford & Miller, 1996), levels of expertise in domain knowledge (Chen, Fan, & Macredie, 2006; Mitchell, Chen, & Macredie, 2005) and levels of experience Table 4 Results of ANOVA analysis of non-linear learning in various cognitive styles. Dependent variables Independent variables Cognitive style F value p-value Scheffe FD (N = 29) FI (N = 35) IM (N = 41) Mean (SD) Mean (SD) Mean (SD) Non-linear learning 4.34 (3.24) 6.24 (6.04) 5.07 (3.47) 1.577 0.212 ns FD: field-dependent; FI: field-independent; IM: intermediate; ns: not significant. Table 5 Results of MANOVA analysis of navigation tools usage in various cognitive styles. Dependent variables Independent variables Cognitive style F value p-value Scheffe FD (N = 29) FI (N = 35) IM (N = 41) Mean (SD) Mean (SD) Mean (SD) ‘‘Find topic” 0.95 (0.56) 1.48 (5.67) 0.81 (0.52) 2.236 0.112 ns ‘‘Alphabetical index” 0.38 (1.08) 4.25 (11.71) 0.73 (6.50) 3.557 0.032a FI > FD, IM ‘‘Hierarchical map” 2.98 (6.75) 0.95 (2.50) 0.81 (2.77) 2.671 0.074a FD > IM, FI ‘‘Where I am” 2.99 (8.08) 0.94 (2.37) 0.41 (1.53) 2.887 0.060a FD > IM, FI FD: field-dependent; FI: field-independent; IM: intermediate; ns: not significant. a p < 0.1. L.-H. Chen / Computers & Education 54 (2010) 1028–1035 1033 with system (Reed & Geissler, 1995; Reed & Oughton, 1997; Reed, Oughton, Ayersman, Ervin, & Giessler, 2000; Reed, Oughton, Ayersman, Giessler, & Ervin, 1995; Shih, Munoz, & Sanchez, 2006) have been recognized as particularly relevant to students’ interaction with Webbased learning system. The preferred learning approach of students with different cognitive styles may vary due to those factors. The final reason for inconsistency might be differences in culture. The participants of previous investigations were rooted in western culture (United Kingdom, USA or Australia), while this study’s participants are from eastern culture (Taiwan). In contrast to west, learners in Taiwan are accustomed to learning in an instructor-centered environment (following a sequence provided by instructor) with little emphasis on individual learning. Consequently, learners in Taiwan tend to be more FD, they learn in a linear way and follow the sequence provided by the program within Web-based learning program regardless of their real cognitive styles. This study also found that cognitive styles have significant effect on learners’ navigation tools usage. FI students prefer to apply the ‘‘alphabetical index” function to locate specific learning material. These behaviors indicate that FI students tend to utilize an active and analytic learning approach directed by their own learning path. Compared with FI students, FD students prefer to use ‘‘hierarchical maps” to help them to understand the structure of the overall learning content. FD students’ navigation behaviors indicate they adopt a global approach to learning, and prefer guidance from the program. Additionally, FD students use the ‘‘where I am” function frequently, since they are very easily lost in a Web-based learning program, and therefore need this function to discover their absolute position. These findings are similar to those of Chen and Macredie’s (2004) and Ford and Chen’s (2000) study. 6. Conclusion The overall findings of this study reveal that cognitive style plays an important role in learners’ interaction with Web-based learning program. Therefore, instructors and Web-based learning system designers should consider providing different support facilities to suit the requirements of learners from each cognitive style. Results from this study indicate that learners prefer to apply a linear approach to navigating learning material, but learners with different cognitive styles have different preferences with regard to navigation tools in the Web-based learning program. Hence, an adaptive Web-based learning program should accommodate the different cognitive styles of learners, rather than focusing on one learning style. To achieve this goal, this study proposed a design model, presented in Fig. 4, to provide suggestions for Web-based learning program designers to tailor the preferences associated with each cognitive style. The proposed model presents the necessary features of a Web-based learning program. According to this model, the system designer should consider the approach to content presentation, and the design of navigation tools, to support the requirements of different learners’. The content presentation style in a Web-based learning program should provide learners with a guided learning path, for instance by presenting an additional link button (next page or continue) to direct students to the next learning item. In term of navigation tools to support students, designer must provide two different features to accommodate learners with each cognitive style. For FI learners, Web-based instructional programs should provide a flexible but with guidance tools that can help FI students reach specific information effectively. For example, an ‘‘alphabetical index” might be a good choice for FI students. Useful tools for FD learners present the complete picture of the content, and also can show learners’ current positions to match FD students’ learning needs. This study has demonstrated the importance of cognitive styles in developing Web-based learning programs, but needs to be undertaken with a larger sample to provide additional evidence, especially with a sample of learners from eastern culture. As this study focuses on discovering the relationship between learners’ cognitive styles and real usage behavior as well as proposing a design model to provide suggestions for Web-based learning program designer. Further research should develop and conduct a test for verifying the adapting Webbased learning program is more helpful to the student than the existing model. The other limitation is that this study focused on cognitive style as a differentiating factor. Further studies could consider other factors that may affect learners’ learning pattern, such as the levels of system experience, gender and the levels of domain knowledge. The impact of different factors on learning pattern of interaction within Web-based instructional programs is an interesting topic for future research. Therefore, future studies could help to construct robust user models for developing customized Web-based learning programs that can accommodate the preferences associated with various individual characteristics. Acknowledgment The authors would like to thank the National Science Council of the Republic of China, Taiwan for financially supporting this research under Contract No. NSC 97-2511-S-324-001. FI learner Navigation support Content presentation Providing guidance Flexible with guidance Global with anchor FD learner Fig. 4. Design model of adapting Web-based learning system for learners with eastern culture. 1034 L.-H. Chen / Computers & Education 54 (2010) 1028–1035 References Chang, C. T. (1995). A study of hypertext document structure and individual differences: Effects on learning performance. PhD. dissertation, University of Illinois at Urbana – Champaign. Chen, S. Y., Fan, J., & Macredie, R. D. (2006). Navigation in hypermedia learning systems: Experts vs. novices. Computers in Human Behavior, 22(2), 251–266. Chen, S. Y., & Ford, N. J. (1998). Modeling user navigation behaviours in a hypermedia-based learning system: An individual differences approach. International Journal of Knowledge Organization, 25(3), 67–78. Chen, S. Y., & Macredie, R. D. (2002). Cognitive styles and hypermedia navigation: Development of a learning model. Journal of the American Society for Information Science and Technology, 53(1), 3–15. Chen, S. Y., & Macredie, R. D. (2004). Cognitive modeling of student learning in Web-based instructional program. International Journal of Human–Computer Interaction, 17(3), 375–402. Chou, H. W. (2001). Influences of cognitive style and training method on training effectiveness. Computers and Education, 37, 11–25. Chou, C., & Lin, H. (1997). Navigation maps in a computer-networked hypertext learning system. In Paper presented at the annual meeting of the association for educational communications and technology. Albuquerque, NM, February 12–16. Davis, J. K., & Cochran, K. F. (1989). An information processing view of field dependence–independence. Early Child Development and Care, 51, 31–47. Durfresne, A., & Turcotte, S. (1997). Cognitive style and its implications for navigation strategies. In B. Boulay & R. Mizoguchi (Eds.), Artificial intelligence in education knowledge and media learning system (pp. 287–293). Kobe, Japan: Amsterdam IOS Press. Ford, N. (1995). Levels and types of mediation in instructional system: An individual difference approach. International. Journal of Educational Multimedia and Hypermedia, 19(4), 281–312. Ford, N., & Chen, S. Y. (2000). Individual differences, hypermedia navigation and learning: An empirical study. Journal of Educational Multimedia and Hypermedia, 9(4), 281–312. Ford, N., & Chen, S. Y. (2001). Matching/mismatching revisited: An empirical study of learning and teaching styles. British journal of Educational Technology, 32, 5–22. Ford, N., & Miller, D. (1996). Gender differences in Internet perceptions and use. Aslib Proceedings, 48, 183–192. Frank, B. M., & Keane, D. (1993). The effect of learner’s field independence, cognitive strategy instruction, and inherent word-list organization on free-recall memory and strategy use. Journal of Experimental Education, 62(1), 14–25. Fullerton, K. (2000). The interactive effects of field dependence–independence and internet document manipulation style on student achievement from computer-based instruction. EdD. dissertation, University of Pittsburg. Goodenough, D. (1976). The role of individual differences in field dependence as a factor in learning and memory. Psychological Bulletin, 83, 675–694. Harasim, L. (2000). Shift happens online education as a new paradigm in learning. The Internet and Higher Education, 3, 41–61. Jolliffe, A., Ritter, J., & Stevens, D. (2001). The online learning handbook – developing and using Web-based learning. London: Kogan Page Limited. Jonassen, D. H., & Grabowski, B. (1993). Individual differences and instruction. New York: Allyn and Bacon. Lee, J. (2000). The effects of information conveying approaches and cognitive styles on learners’ structural knowledge and perceived disorientation in a hypermedia environment. PhD. dissertation, Indiana University. Lee, H. M., Cheng, Y. W., Rai, S., & Depickere, A. (2005). What affect student cognitive style in the development of hypermedia learning system? Computers and Education, 45, 1–19. Liu, M., & Reed, W. M. (1995). The effect of hypermedia assisted instruction on second-language learning through a sematic-network-based approach. Journal of Educational Computing Research, 12(2), 159–175. Messick, S. (1976). Individuality in learning. San Francisco: Jossey-Bass. Mitchell, T. J. F., Chen, S. Y., & Macredie, R. D. (2005). Hypermedia learning and prior knowledge: Domain expertise vs. system expertise. Journal of Computer Assisted Learning, 21(1), 53–64. Oughton, J. M., & Reed, W. M. (1999). The influence of learner differences on the construction of hypermedia concepts. A case study. Computers in Human Behavior, 15, 11–50. Reed, W. M., & Geissler, S. F. (1995). Prior computer-related experiences and hypermedia metacognition. Journal of Research on Computing in Education, 30(1), 581–600. Reed, W. M., & Oughton, J. M. (1997). Computer experience and interval-based hypermedia navigation. Journal of Research on Computing in Education, 30, 38–52. Reed, W. M., Oughton, J. M., Ayersman, D. J., Ervin, J. R., & Giessler, S. F. (2000). Computer experience, learning style and hypermedia navigation. Computer in Human Behavior, 16, 609–628. Reed, W. M., Oughton, J. M., Ayersman, D. J., Giessler, S. F., & Ervin, J. R. Jr., (1995). Computer experience, learning style, and hypermedia navigation. Computers in Human Behavior, 16, 619–628. Riding, R. J. (1991). Cognitive styles analysis. Birmingham, UK: Learning and Training Technology. Riding, R., & Rayner, S. G. (1998). Cognitive styles and learning strategies. London: David Fulton Publisher. Riding, R. J., & Sadler Smith, E. (1992). Type of instructional material, cognitive style and learning performance. Educational Studies, 18, 323–340. Shih, P., Munoz, D., & Sanchez, F. (2006). The effect of previous experience with information and communication technologies on performance in a Web-based learning program. Computer and Human Behavior, 22(6), 962–970. Witkin, H. A., Moore, C. A., Goodenough, D. R., & Cox, R. W. (1977). Field-dependent and field independent cognitive styles and their educational implications. Review of Educational Research, 47(1), 1–64. Zukerman, I., Albrecht, D. W., & Nicholson, A. E. (1999). Predicting users request on the WWW. In Proceedings of the 7th international conference on user modeling, UM99 (pp. 275–284). Banff, Canada.