Asia-Pacific Forum on Science Learning and Teaching, Volume 8, Issue 1, Article 5 (June, 2007)
Teaching argumentation through the visual models in a resource-based learning environment

Previous Contents Next

The existing visual models of argumentation

More and more researchers think science education ought to equally emphasize cognitive psychology and philosophy of science for the sake of developing better instruction in science education (Duschl & Hamilton, 1992; Nersessian, 1989; Perkins, 1985; Perkins & Salomon, 1989; Resnick, 1987). Therefore, I tried to adopt theoretical models regarding argumentation from the domains of psychology and philosophy for the instructional design. Three visual models of argumentation were considered for assisting students' argumentation, which are Toulmin's model (Toulmin, 1958), Means and Voss's model (Means & Voss, 1996), and Lakatos' scientific research programmes (Chang & Chiu, 2005b).

Toulmin's model (Toulmin, 1958) has been applied by many researchers for analyzing argumentation (Erduran, Simon, & Osborne, 2004; Fimenez-Aleixandre, 2002; Kelly & Chen, 1999; Russell, 1983; Simonneaux, 2001; Tirri & Pehkonen, 2002). Since the early periods, some philosophers like Plato, Socrates and Aristotle concerned themselves with thinking, and all of them thought that reasoned arguments constructed the core of thinking (Kuhn, 1991). From that time, formal logic was taken as the best model of thinking until the middle of the twenty century. In 1958, Toulmin published the important book entitled "The uses of argument", where he pointed out the limits of using formal logic to think and brought up his own model of a reasoned argument and emphasized the advantages of using argumentation as a kind of thinking (Toulmin, 1958). Composing Toulmin's model are data, claim, backing, warrant, qualifier and rebuttals (Figure 1), where special features of the elements need to be pointed out, e.g. warrant is looked upon as a hypothetical and bridge-like statement, backing is regarded as the categorical statements of fact, and qualifier is taken as a tentatively qualified conclusion (Toulmin, 1958). The example is also presented in Figure 1.

Figure 1: Toulmin's model of argumentation (Toulmin, 1958)

In 1996, Means and Voss provided three models generated from an empirical study regarding informal reasoning process (Means & Voss, 1996), but these models could be traced back to 1991. Based upon the Aristotle's enthymeme, Voss and Means brought up their first model of argumentation (Figure 2), in which there is a supporting reason and a claim (Figure 2A), and also pointed out the existence of contradiction to the claim (Figure 2B) (Voss & Means, 1991). Moreover, they generated three advanced models from their empirical study regarding informal reasoning (Figure 3), which include the skeletal model, the enhanced model, and the elaborated model (Means & Voss, 1996). In the skeletal model, people only generate a reason (R) to support the conclusion (C) (Figure 3A), and the enhanced model is with qualifiers (Q) which provides various solutions of various conditions (Figure 3B). In terms of the elaborated model, it includes more reasons and qualifiers, and the presence of the counterargument (using "=" to show the contradiction status) (Figure 3C).

Figure 2: Voss and Means’s model of argumentation (Voss & Means, 1991)

Figure 3: Means and Voss's three advanced models (Means & Voss, 1996)

Due to some limitation found while analyzing students' informal argumentation regarding socioscientific issues, Lakatos' scientific research programmes, a famous theory in the field of philosophy of science, was presented as an alternative theoretical model for analyzing argumentation (Chang & Chiu, 2005b). Lakatos's programmes embraces the five indicators regarding the skills of informal argumentation, which include (Chang & Chiu, 2005b):
Making claims: individuals could make the claims for issues.

The main essence of the Lakatos' programmes stresses the concept of a series of theories instead of the concept of theory, the basic concept from logic of discovery (Lakatos, 1978). There are four connected components embedded in the Lakatos' model which are hard-core (HC) located in the core of the model, protective belt (PB) surrounding HC, negative heuristic (NH) and positive heuristic (PH) holding in the model (Lakatos, 1970). HC is the core and foundation of the theory and it possesses the tough and unchangeable features. PB is composed by auxiliary hypotheses for preventing HC from being attacked. NH and PH are both strategies embedded in the model with separate functions to forbid rebuttals and to expand theory. Moreover, there is an important function of PB is to adjust the auxiliary hypotheses when the theory meets the anomaly and could absorb the anomaly via NH and PH. The graphical representation of the Lakatos' scientific research programmes is shown in Figure 4. The basic idea is to use the middle circle symbolizing HC, and PB is another bigger circle surrounding HC. Furthermore, two arrows used to show NH and PH in PB. The same direction with the running minute hand represents PH, and reversed one stands for NH.

Figure 4: Lakatos's scientific research programmes model

The example of using Lakatos' programmes to construct arguments regarding genetically modified food has been implemented and found it is a workable model by Chang and Chiu (2005b). The quotation from the past study is as below.

(HC) I would buy the genetically modified food (claim), because I learned about genes from a biology course, which makes me know our genes will not change after we eat the genetically modified food (reason). Besides, genetically modified food has its exploitable value, if our government did not make an effort, then our country will fall behind other countries (PH). Generally, we have not heard anything harmful to humans, but it is necessary to conduct clinical trials like the process before marketing the medicine (NH).

In accordance with the theoretical background, the purpose of this study is to teach argumentation through the presentation of visual models in a resource-based learning environment. The research questions include: (1) what kind of visual model do most participants choose to construct their arguments? (2) what are the reasons for choosing this visual model? (3) what kind of questions do participants feel interesting to explore concerning genetically modified food (GMF)?

Copyright (C) 2007 HKIEd APFSLT. Volume 8, Issue 1, Article 5 (June, 2007). All Rights Reserved.