Academic journal article Educational Technology & Society

A Fuzzy-Based Prior Knowledge Diagnostic Model with Multiple Attribute Evaluation

Academic journal article Educational Technology & Society

A Fuzzy-Based Prior Knowledge Diagnostic Model with Multiple Attribute Evaluation

Article excerpt

Background and objectives

Evaluating and strengthening the prior knowledge of individual students is an important task before teaching and learning new knowledge or skills, since prior knowledge affects how instructors and students interact with the learning materials they encounter (Chieu, 2007; Moos & Azevedo, 2008; Ozuru, Dempsey, & McNamara, 2009). From the perspective of instructors, gaps in the students' prior knowledge often confound their best efforts to deliver effective instructions (Roschelle, 1995). Moreover, they can also affect how instructors plan their teaching strategies for new material in order to enhance students' learning motivation and performance (Biswas, 2007; Tseng, Chu, Hwang, & Tsai, 2008).

If the students do not have the necessary prior knowledge, then there is a strong risk that they may build new knowledge on faulty foundations (Dochy, Moerkerke, & Marten, 1996). It can thus be seen that inadequate or fragmented prior knowledge is an important issue, and if the instructors' expectations of the students' knowledge are very different from their actual knowledge, then both teaching and learning are likely to adversely affected (Hailikari, Katajavouri, & Lindblom-Ylanne, 2008).

To avoid this risk, tests are usually adopted to assess how well students understand a concept or piece of knowledge (Panjaburee, Hwang, Triampo, & Shih, 2010; Tao, Wu, & Chang, 2008; Treagust, 1988). Nevertheless, conventional testing systems usually assign only an overall score or grade to students, and thus instructors and students may be unable to identify which specific concepts or pieces of knowledge are misunderstood, making it difficult to improve the learning performance of students (Gerber, Grund, & Grote, 2008; Gogoulou, Gouli, Grigoriadou, Samarakou, & Chinou, 2007; Hwang, Tseng, & Hwang, 2008). To work around this issue, instructors can further analyze the testing results to determine the students' learning deficiencies. However, this is a time-consuming task that presents a heavy workload for instructors, since there are often many students on a course, especially in higher education or e-learning contexts. Hence, previous work has led to the development of a prior knowledge diagnosis (PKD) model to assist instructors and students in diagnosing and strengthening prior knowledge before new instruction is undertaken (Lin, Lin, Huang, 2011).

Nevertheless, one of the major problems when applying the PKD model is that it only uses correctness rates answered by students to determine their level of understanding with regard to particular concepts, and diagnoses based on a single attribute lead to inaccurate results (Hwang, Tseng, & Hwang, 2008). Therefore, it is necessary to develop a more effective approach to assist instructors in identifying the specific learning problems of individual students in the context of multiple attributes, and this issue actually matches a traditional computer science problem, called Multiple Attribute Decision Making (MADM) (Chen & Hwang, 1992). The MAMD problem is to select the best choice among the previously specified finite number of alternatives (Seel, & Dinter, 1995), with the alternatives evaluated based on their attributes.

Therefore, this study proposes a Fuzzy Prior Knowledge Diagnostic (FPKD) model by applying the Efficient Fuzzy Weighted Average (EFWA) technique (Lee & Park, 1997) to assist instructors in diagnosing the level of students' understanding of prior knowledge, and to provide appropriate feedback to individual students. Based on this model, a testing and diagnostic system has been implemented, and an experiment on an interdisciplinary bioinformatics course was conducted to demonstrate the efficacy of the proposed approach.

Fuzzy Prior Knowledge Diagnostic Model

The aim of the FPKD model is to assist instructors in diagnosing students' prior knowledge with multiple attributes. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.