KNOWLEDGE CONSTRUCTION USING WEB-BASED CONSTRUCTIVIST APPROACH: A CRITICAL EVALUATION OF STUDENTS PERFORMANCE M.

Hafiz Yusoff Department of Computer Science Uni. Malaysia Terengganu hafiz744@gmail.com Abstract At present, researches on improvement of web-based e-learning education are more focused on introducing new technology and exploration of software standard. This study focused on the content structure towards a more appropriate to users cognitive style. In this study, empirical research technique was adapted to test the effectiveness of two approaches; constructivist approach (Cons. A) and objectivist approach (Obj.A) as the basis for the development of educational websites. Both types of websites are developed using these approaches and have been tested based on 3 parameters; i) the students knowledge level, ii) the level of students motivation, and iii) the level of students retention. Three tests were conducted on 154 students of Business Management subject from 2 secondary schools in Malaysia. A 2x2 (web-based learning approach x cognitive style) factorial design was employed in this quasi-experimental study. Two independent variables were investigated; the approaches in the web-based learning environment (constructivist approach and objectivist approach), and the cognitive styles (field Independent; FI, and field Dependent, FD). The control group received the objectivist approach while the experimental group received the constructivist approach treatment. They were randomly assigned to each of the two treatment groups. Meanwhile, the students performances were based on their pre-test and post-test knowledge test scores which identified as the dependent variable. It was discovered that the students who received knowledge using the web-based constructivist approach performed significantly higher than those who received objectivist approach in their knowledge test results. All the data that were obtained from experimental process have been analyzed for statistical descriptive and analysis covariate (ANCOVA) using statistical software. Keywords: E-learning, Knowledge construction, web-based learning environment, constructivist approach, cognitive style 1. Introduction

In education, Internet has also changed the conventional teaching method to a teaching method based on webs or web-based learning. Nevertheless, the effectiveness of a web-based learning largely depends on the ability of the students using it, and its compatibility with their level of thinking. Therefore, the content compilation factor in educational websites, the combination of learning theories and attractive displays are important factors to ensure maximum effects on students.

Proceedings of Regional Conference on Knowledge Integration in ICT 2010

533

1.1. Theoretical Background 1.1.1 Learning Theory: Constructivist vs Objectivist Approach The constructivism learning approach emphasized that human gained knowledge from his mental efforts. Constructivism involved students in a meaningful experience where students interpret the information themselves without following what was mapped by other people [1]. This approach focuses on how to study, generate and test the knowledge gained. Among the constructivist approach in learning is the inquiry based instruction theory by Collins and Stevens [2] and learning through discovery like learning cycle model by Lawson [3]. The approach suggested by this constructivist group is focused on the student where learning take place as a result of the student effort or through the experience of the student himself. Constructivist believed that an active, spontaneous and self-guided learning approach is necessary [3]. Objectivist approach believed that knowledge is in a cumulative form and learning will be the outcome based on prerequisites fulfilled, such as past knowledge required for further learning. Among the learning theories following this objectivist approach is the theory brought forward by Gagne[4], Landa[5] and Scandura[6]. Gagne[4] listed five learning products, which are verbal information, intellectual skill, cognitive strategy, motor skill and attitude. 1.1.2 Cognitive Style: Field Dependent (F.D) vs Field Independent (F.I) Different cognitive style explains that each learner have different ability, faith, interest and learning style [7]. Field dependent (FD) learner is a student who cannot be free from environment elements or distracting background when trying to isolate an aspect in a certain situation. It was also found to be difficult to use a structure on a non-structural situation [8]. In this research, students achieving the GEFT score of lower than the GEFT mean score for the sample group were categorised as FD students. Field Independent (FI) learner is a student who can overcome distracting effects of background elements when trying to isolate an aspect in a certain situation. They are also found to have the skill to develop structure on a non-structural situation [8]. In this research, students achieving the GEFT score of more than the GEFT mean score for the sample group will be categorised as FI students. 2. Research Methodology 2.1 Research Design and Procedures This research was 2x2 quasi-experimentally conducted using the pre-test - post-test control group design as suggested by Cook and Campbell [9]. This research design consisted of treatment group and control group. Subjects were randomly selected into these two groups. Independent variable treatment was conducted on subjects in the treatment group. The following are steps in making this research project a success; Step 1: Prepared the research proposal. Step 2: Designed and developed educational websites. Step 3: Conducted formative test on the suitability of the websites produced.
Proceedings of Regional Conference on Knowledge Integration in ICT 2010 534

Step 4: Conducted a pilot study to test the validity and reliability of items in Pre-test and Post-test, as well as both websites developed. Step 5: Conducted orientations on learning through educational websites for the teachers involved. Step 6: Conducted orientations on computer and educational websites usage among the participating students. Step 7: Conducted GEFT test. Step 8: Selecting samples for both control and treatment groups. Step 9: Conducted orientations on computer usage and web browsing using program and other websites for students. Step 10: Conducted a Pre-test on both control and treatment groups. Step 11: Conducted learning sessions for both groups. Step 12: Conducted Post-test 1. Step 13: Conducted attitude measurement questionnaire. Step 14: Conducted Post-test 2. Step 15: Carry out data analysis. 2.2 Research Instruments 2.2.1 Knowledge Test (Pre-test and Post-test) These two tests were given to determine the existing knowledge level of students participating in the research before and after being treated with Business Management web-based learning. Pre-test was given to students prior to participating in the Business Management webbased learning. The said test contained 25 multiple choice questions related to Fundamental of Business Management " topic and was given to all students participating in this research. The time given to answer all those questions was 35 minutes. The construction weight for the questions given covered informative, comprehensive, application and analytical questions. On the difficulty level, 5 questions were for low level, 10 medium level questions and 10 high level questions. The student's score was determined according to item-by-item where a correct answer will be awarded 1 point and a wrong answer will not be awarded any point. Post Test was given right after the students have participated the Business Management web-based learning. This test also contained the same 25 questions as Pre-test, but the order of the questions was changed. The test was also given to all students participating in this research. The time given to answer the questions was 35 minutes. Post Test2 is to observe the retention effect between the two web sites developed using different approach. The test was given to all research group members after two weeks Post-test1 was conducted. This test also contained the same 25 questions, but the order of the questions was changed so that it varies from Pre-test and Post-test1. The time for answering the questions was maintained at 35 minutes. Because of the test instrument was developed by researcher, a pilot test was conducted to test its validity and reliability. The question contents validity was evaluated by two state Business Management main coaching teachers and a lecturer with expertise in test items development field. This is to ensure that the said test is in accordance with the topics and concepts required in the said subject's curriculum. The said questions' reliability as well as validity was also determined through a pilot test conducted on a group of students taking
Proceedings of Regional Conference on Knowledge Integration in ICT 2010 535

Business Management subject in Form 4 from two schools that are not participating in the actual research. Data obtained was analyzed using Item and Test Analysis Program - ITEMAN Version 3.5. 2.2.3 Group Embedded Figure Test (GEFT) GEFT measuring instrument was used to classify cognitive style for field-dependent (FD) learner and field-independent (FI) learner [7]. The GEFT test used in this research is the GEFT Test published by Mind Garden, Inc. This test requires the subject to seek and fit simple figures in a number of complex shapes. Generally, this test comes in three parts and based on the allocated time. For part I, the time given was 2 minutes. To complete part II and III, the time given was 10 minutes. Based on the said test, subjects who scored above average in GEFT are known as FI learners, while subjects who scored lower than average are known as FD learners. This test was conducted after a sample of 154 students were identified and selected randomly. 3.0 Findings and results In order to identify whether there is any significant difference in the pretest scores between the control and experimental groups, an analysis was first being carried out. A ttest analysis revealed that both groups differed significantly prior to the treatment (t: 2.25; p: 0.03), with the constructivist approach group performed significantly better than the objectivist approach group (Cons. A. mean: 47.9, Obj. A mean: 44.08). Thus, due to this significant difference prior to the treatment, Analysis of Covariate (ANCOVA) was used for data analysis in which the pretest score was identified as the covariate. Also, due to the application of ANCOVA, the posttest mean scores reported in the following sections were based on the adjusted means instead of the observed means. 3.1 Results The descriptive statistics of the adjusted posttest scores for the two treatment groups and the two cognitive style groups involved in the study are shown in Table 1. Table 2 indicates the ANCOVA analysis for the WBL approaches and cognitive style as the two main effects of this study, while Table 3 indicates the Pairwise Comparisons between the four groups (Cons. A. FI, Cons. A.-FD, Obj.A-FI, and Obj.A-FD). Table 1. Descriptive statistics matrixes of the adjusted posttest mean scores for the four groups (2 x 2) Cons. A Obj. A Total n 29 41 70 FD M 51.32 54.32 54.49 SD 1.30 1.15 1.01 n 39 32 71 FI M 63.54 54.71 58.02 SD 1.18 1.37 1.01 n 68 73 141 Total M 59.81 52.97 SD 0.98 0.94

Proceedings of Regional Conference on Knowledge Integration in ICT 2010

536

Table 2. ANCOVA for cognitive styles and WBL approach (Adjusted Posttest) Source Cognitive styles WBL 1587.12 * p < 0.05 Table 3. Pairwise Comparisons (Adjusted Post-test) Mean Diff. Signif. Obj. A - FI Cons. A FI Cons. A - FD Cons. A- FD Obj. A - FD 8.827 12.217 -3.00 0.00 0.00 0.513 SS df MS 437.81 1587.12 F-value 6.11 25.04 p-value 0.02* 0.00*

437.81 1 approaches 1

i) Research Question 1: Is there any significant difference in terms of knowledge level between the constructivist approach and objectivist approach group? Based on the ANCOVA result (Table 2), there is a significant difference in terms of knowledge level between the constructivist and objectivist approach groups (F1, 138 : 25.04, p<0.05). This result clearly indicates that the constructivist approach group performed significantly better than the objectivist approach group (MCA: 59.81; MOA: 52.97). ii) Research Question 2: Is there any significant difference in term of knowledge level between FI and FD students? The result in Table 2 indicates a significant main effect for the two levels of field dependence- independence, (F1, 138 = 6.11, p: 0.02). The FD and FI students exhibited significant differences in knowledge level, with the latter group of students performed better than the former group in the knowledge test (MFI: 58.02; MFD: 54.49). iii) Research Question 3: Is there any significant difference in term of knowledge level between FI students who received constructivist approach and FI students who received objectivist approach? A pair-wise comparison analysis (Table 3) between the FI students in the constructivist approach group and the FI students in the objectivist approach group revealed p<0.05 with the mean difference of 8.83 (MFI-CA: 63.54; MFI-OA: 54.71).This indicated that the FI students in the constructivist approach group performed significantly higher than the FI students did in the objectivist approach group. iv) Research Question 4: Is there any significant difference in term of knowledge level between FD and FI students who received the constructivist approach?

Proceedings of Regional Conference on Knowledge Integration in ICT 2010

537

A pair-wise comparison analysis (Table 3) between the FI and FD students in the constructivist approach group revealed a p-value of <0.05 with the mean difference of 12.22 (MFD-CA: 51.32; MFI-CA: 63.54). This indicated that the FI students performed significantly higher than the FD students did in the constructivist approach group. v) Research Question 5: Is there any significant difference in term of knowledge level between FD students who received constructivist approach and FD students who received objectivist approach? A pair-wise comparison analysis (Table 3) between the FD students in the constructivist approach group and the FD students in the objectivist approach group indicated an insignificant difference in knowledge level (p:0.513) with the mean difference of 3.00 (MFD-CA: 51.32; MFD-O.A: 54.32). However, although not significant, the FD students who received the objectivist approach group had performed slightly better compared to those in the constructivist approach group. vi) Research Question 6: Is there any significant difference in term of retention between the students groups who leaned through a website that used constructivist approach compared to the retention of the group who are learning through an educational website developed using objectivist approach. The retention was measured using Post-Test 2, which was conducted two weeks after Post-Test 1. The Repeated Measures of the ANOVA analysis was used to see the difference. Referring to Table 4, the student group who learned using the website designed based on constructivist approach obtained a higher Post-Test 1 score mean (Mean = 61.35) compared to the Post-Test 2 score mean (Mean = 57.03). There is a difference of approximately 4% between the Post-Test 1 and Post-Test 2 for student group who learned using the website developed based on constructivist. In the meanwhile, the student group who are learning using the website developed based on objectivist approach have also obtained a higher PostTest 1 score mean (Mean = 51.53) compared to Post-Test 2 score mean (Mean = 48.27). The difference between Post-Test 1 and Post-Test 2 is approximately 3% for the student group who are learning using the website developed based on objectivist approach. Table 4: Post-Test 1 and Post-Test 2 mean for constructivist and objectivist groups Learning Std. Approach Mean Deviation N Constructivis TIME 1 t 61.35 11.680 68 (Post-Test 1) Objectivist 51.53 10.669 73 Total 56.27 12.168 141 Constructivis TIME 2 t 57.03 11.847 68 (Post-Test 2)
Proceedings of Regional Conference on Knowledge Integration in ICT 2010 538

Objectivist 48.27 Total 52.50 12.145 141 10.895 73

The result of the ANOVA Repeated-Measures anaylsis on this study data, shown in Table 5 had produced Wilks' =0.99, F1,139 = 0.59 and p=0.45. This difference is not significant. This research suggested that the learning retention of the student group who received constructivist approach is equal to the retention of the student group who received objectivist approach. Table 5: The Post-Test 1 and Post-Test 2 ANOVA Repeated-Measures ( for constructivist and objectivis approach website design) Multivariate Test (c) Noncent Observa Eta . tion Hypoth Expone esis Square Parame- nt Erro Value F df r df Sig. ter (a) Pillai's Trace 139.0 TIME .176 29.786(b) 1.000 00 .000 .176 29.786 1.000 Wilks' Effect Lambda 139.0 .824 29.786(b) 1.000 00 .000 .176 29.786 1.000 Hotelling's Trace 139.0 .214 29.786(b) 1.000 00 .000 .176 29.786 1.000 Roy's Largest Root 139.0 .214 29.786(b) 1.000 00 .000 .176 29.786 1.000 TIME * Pillai's 139.0 .004 .585(b) 1.000 00 .445 .004 .585 .118 DESIG Trace N Wilks' Lambda 139.0 .996 .585(b) 1.000 00 .445 .004 .585 .118 Hotelling's Trace 139.0 .004 .585(b) 1.000 00 .445 .004 .585 .118 Roy's Largest Root 139.0 .004 .585(b) 1.000 00 .445 .004 .585 .118 a Generated using alpha = .05
539

Proceedings of Regional Conference on Knowledge Integration in ICT 2010

b c

Accurate Statistic Design: Intercept+METHOD With Subject Design: TIME

vii) Research Question 7: Is there any significant difference in term of students motivation level toward using educational website that used constructivist approach compared to learning through educational website developed using objectivist approach. Referring to Table 6, the student group who are learning using the website designed based on constructivist obtained a Motivation Test score mean of 3.69 compared to the student group who are using the objectivist approach (Mean = 3.60). As shown in Table 7, the one-way ANOVA analysis result on this study had produced the values of F1,139 =2.392, p=0.124. This difference is not significant. This research finding suggested that there is no difference in the motivational level of students who are receiving constructivist approach compared to the motivational level of the student group who are receiving objectivist approach. Table 6: Mean and standard deviation for the Motivation Test score Motivation Test Standar Standa 95% Confidence d rd Interval Deviati N Mean on Error For Mean Lower Upper Limit Limit Constructivi st 68 3.6998 .36161 .04385 3.6122 3.7873 Objectivist 73 3.6092 .33350 Total 141 3.6529 .34906 .02940 3.5948 3.7110 .03903 3.5314 3.6870

Table 7: One-way ANOVA for Motivation Test score mean with website design (constructivist and objectivist approach) Motivation Test Sum of Square Mean Square Df F Sig. Among .289 1 .289 2.392 .124 Groups With 16.769 139 .121 Groups Total 17.058 140 4.0 Discussion
540

Proceedings of Regional Conference on Knowledge Integration in ICT 2010

This research involved developing two websites related to the Business Management subject that were designed using two different approaches; one using objectivist approach and another using constructivist approach. Next, this research compared the effectiveness of the two designs to construct new knowledge, motivation and knowledge retention of students after learning through these two websites. This research also looked at the effect of using two different website design with different cognitive style on their achievement. This study found that the students who learned using the website developed based on constructivist approach had obtained a significantly higher achievement on their knowledge test compared to the students who learned using the website developed based on objectivist approach. Questions given at the beginning of the lesson through the website developed based on constructivist approach will assist the students to think to find the answers for the said situation. A strong curiousity had impelled the students to explore, observe, predict, test hypotheses, and build concepts concerning the questions given through the suggested links. Students will also use their past knowledge to apply to this new situation. When learners are actively involved in organizing their own understanding on certain matters, it will give them a more meaningful understanding. On the other hand, for the control group who used the website developed based on objectivist approach, they were only being passive information receivers through the information display activities which were linearly arranged. This group had undergone a deductive and passive learning process where all the learning objectives and directions were provided for them. Learners are only required to follow the lesson prepared from one step to the next. Finally, they will know whether their learning objectives were achieved or not depending on the result of the quiz prepared after completing the tutorial. This study had found that learners in a field independent (FI) cognitive group obtained a significantly higher achievement compared to learners with field dependent (FD) cognitive style. FI learners are individuals with individualistic attitude, and do not need others in processing information. They have their own abilities in structuring their knowledge. On the other hand, FD learners prefer social interactions, depending on others in structuring information for them, and more ready to accept input given to them by those with authorities. They are also more capable in choosing necessary information in their learning. However, FD learners are facing difficulties in freeing themselves from disruptive background elements and often faced with problems in choosing important and less important information in the said website environment. This study found that there was no significant difference in terms of retention between the group who learned through an educational website using constructivist approach and group of objectivist approach. However, on the overall, both the control and treatment groups still showed a high achievement with only a small decline in the achievement after getting their lessons using the provided materials two weeks later. Among the factors which may have caused this insignificant difference in terms of retention between the control and treatment group may have been the time and the too short test frequency factors. The difference may be more pronounced if the delayed post test were to be conducted more than once and in a longer period. Learning through a website developed based on constructivist approach should give a better retention effect than learning through a website developed based on objectivist approach. In constructivist, learners are involved in building and structuring their knowledge themselves. Situations such as this should have reinforced learners memory of things they have learned. On the other hand, memory is easily lost when learners only received and agreed with the knowledge passed to them through others explanation as displayed in the website developed based on objectivist approach. There was no significant difference in terms of motivational, for learners who learned using a objectivist approach website compared to learners who learned using constructivist
Proceedings of Regional Conference on Knowledge Integration in ICT 2010 541

approach website. Eventhough this finding was insignificant, based on the score mean difference between the two groups, it was shown that the motivation level of the treatment group was higher than the control group. Among the factors that may have caused this insignificant findings would be both groups were using materials developed based on computers and websites. Web-based learning materials are still considered new today. As usual, a novel and not much used material will give a higher motivation to learners to use it. This findings confirmed Hackbart [10] assumption, stating that new technologies are able to increase motivation and gain learners interest to follow the provided lessons. In addition, each group only used the materials prepared for their group only, without looking at materials used by the other group. In other words, the treatment group cannot compare the advantages of the approach gained by the control group, and vice versa. They only evaluate the materials used by their own group. 5.0 Conclusion This research had proven that website design that was developed based on constructivist approach was able to improve learners knowledge level in their learning compared to objectivist approach. On the whole, field independent (FI) cognitive style learners benefitted more than field dependent (FD) learners in learning through web-based learning materials. This study also found that FI learners obtained a higher performance than FD learners through learning using website developed based on constructivist approach. Also, the FD and FI learners achievement did not show any difference in learning using website developed based on objectivist approach. For FI cognitive learners, their performance was more outstanding in learning using website developed with constructivist approach than website developed based on objectivist approach. While FD learners did not show any difference in terms of achievement, either through website developed based on constructivist or objectivist approach. Overall, FD learners also succeeded in improving their achievement in learning through website, but their achievement was still low compared to FI learners in the said learning. Among the factors identified causing the FD learners inability to improve their achievement as well as the FI learners in their learning was that the said learning was conducted individually. Apart from that, this study finding also showed no significant difference in terms of retention and motivation among learner groups who used the constructivist and objectivist approach websites. However, both groups showed high learning retention and motivation in their learning through both websites assigned to them. It is hoped that this study will be able to give a picture of the effectiveness of the website developed based on constructivist approach in improving learners achievement and motivation in their learning. It is also hoped that this research findings can serve as a guide to those who are involved in developing educational website in this country. Other than that, to ensure the web-based instruction effectiveness, cognitive style differences is among the factors that should be considered and applied in the said learning material environment. References [1] Heinich, R. Molenda, M. Russell,J.D. & Smaldino, S. (1999). Instructional Media and Technologies for Learning. Englewood Cliffs. NJ: Merrill [2] Collins, A. & Stevens, A.L. (1983) . A cognitive theory of inquiry teaching. Di dalam Reigeluth (Ed.) Instructional Design Theories and Models: An Overview of Their Current Status, Hillsdale: Lawrence Earlbaum Associates. m.s. 250-276. [3] Lawson, A. E. (1995). Science Teaching and The Development of Thinking. California: Wadsworth Publishing Company
Proceedings of Regional Conference on Knowledge Integration in ICT 2010 542

[4] Gagne, R.M. (1985). The Condition of Learning (4th ed.). New York: Holt Rine Hart & Winston. [5] Landa, L. N. (1983). The algoheuristic theory of instruction. Dalam Reigeluth (Ed.) Instructional Design Theories and Models: An Overview of Their Current Status, Hillsdale: Lawrance Earlbaum Associates. m.s. 257-776. [6] Scandura, J.M. (1983). The structural learning theory. Dalam Reigeluth, C.M (Ed) Intructional-design Theories and Models : An Overview of Their Current Status. Hilldale, NJ: Lawrence Erlbaum. [7] Witkin, H.A., Oltman, P., Raskin, E. & Karp, S. (1971). A Manual for the Embedded Figures Test. Palo Alto CA: Consulting Psychologists Press. [8] Lourdusamy, A. (1994).Different cognitive style and implication to education. In: Professorship Talk (1995/3), 5 Feb 1994, Penang: Universiti Sains Malaysia [9] Cook, T.D. & Campbell, D.F. (1979). Quasi-Experimentation Design & Analyisis Issues for Field Settings. Chicago: Rand Mc Nally. [10] Hackbart, S. (1997), Web-based learning in the context of K-12 school curriculums. Educational Technology, 37 (3), m.s. 59-71.

Proceedings of Regional Conference on Knowledge Integration in ICT 2010

543