The learning effect of modeling ability instruction

In the past decades, model and modeling have been recognized as an important medium of scientists’ inquiry, teachers’ teaching, and students’ learning about the sciences. In the process of learning and teaching science, models are important representation and tools (e.g., Gilbert & Boulter, 1998a, 1998b; Grosslight, Unger, & Smith, 1991; Harrison & Treagust, 2000; Justi & Gilbert, 2002; Justi & van Driel, 2006). Scientific model and modeling processes could also make students develop higher order scientific thinking and working, such as developing meta-cognition to understand the inquiry process in science community, getting familiar with the development and construction of knowledge and individually reflecting on the understanding of scientific knowledge (Clement, 1989; Coll, France, & Taylor, 2005). It is also revealed that students can transfer the modeling ability to their science learning and inquiry, through modeling instruction (Schwarz & White, 2005), and they are able to distinguish the relationship between macro-phenomenological level and micro-phenomenological level (Ergazaki, Komis, & Zogza, 2005). Model and modeling competence is deemed an important part of scientific literacy as well (Gilbert, Boulter, & Elmer, 2000).

The goal of scientific literacy is to help individuals to live better in the current science and technology-oriented society, and to achieve the goal of scientific literacy to promote students’ and citizens’ learning interests, understanding and respect for sciences; in order to do so, it needs to be more emphasized on the instructional design and the evaluation of science learning (Aikenhead & Ryan, 1992; Champagne & Newell, 1992; Jenkins, 1992; Laugksch, 2000; Laugksch & Spargo, 1996). How to make students understand the scientific concepts, and meanwhile, also possess modeling ability? A good learning effect on engaging scientific concepts with daily life phenomenon and hands-on activity have been disclosed (Chang & Chiu, 2005). However, in-service teachers very often mention the difficulty of teaching with limited time for each lesson, and this is the reason why they could not add too much life related content to the scientific concepts or spend time in inducing students’ learning interests and/or modeling ability via hands-on activities. Therefore, how to help teachers teach in a more efficient and fruitful way becomes increasingly important. The basic idea of this study is to develop a "modeling ability instruction" to promote students’ science learning and modeling ability.

Based upon the statements above, the purpose of this study is to investigate the learning outcome of the instructional design embedded the theories in regard to "modeling" and "cognitive apprenticeship" to promote students’ modeling ability and understanding of the scientific concepts about the battery. The importance of model and modeling in science education is going to be presented, and also the framework regarding modeling ability is delineated. Moreover, the effect of embedding the theory of cognitive apprenticeship in teaching from the former research is described.

Model and modeling in science learning

It has been revealed that students can perform better when they perceive more knowledge about models (Harrison & Treagust, 2000; Hodson, 1993). Besides helping science learning, visual models are mentioned to help students’ ability to create dynamic mental models (Velázquez-Marcano, Williamson, Ashkenazi, Tasker, & Williamson, 2004). In addition, modeling instruction has been revealed to correct students’ misconceptions (Metcalf & Tinker, 2004). According to the advantages found from the literature, in this scienceand technology-developing century, cultivating students’ modeling ability needs to be taught. Therefore, in this study, developing modeling ability instruction for promoting learners’ modeling ability is a goal, and the instruction is anticipated to enhance learners’ scientific concepts. While teaching the challenging topics of electricity, light, and so on, Slotta and Chi adopted ontology training before teaching scientific concepts, and the results revealed that students could learn about electric current far better after ontology training (Slotta & Chi, 2006). In this study, a similar hypothesis is developed; students could learn better about scientific concepts after modeling ability instruction. Both modeling ability framework and the theory of cognitive apprenticeship are considered in developing modeling ability instruction.

The framework of modeling ability

This study was a sub-project of a two-year project jointly funded by the National Science Council (NSC 95-2511-S-156 -001-MY2). One of the main targets of the join project is to develop a questionnaire for evaluating students’ modeling ability. Three dimensions to the modeling ability include ontology (Chou, 2007), epistemology (Wu, 2007) and methodology (Chiu, 2007). The essence of these three dimensions are that individual needs to know (1) Ontology: what is the model and modeling? (2) Epistemology: how to know the model and modeling? (3) Methodology: how to manipulate the model or create models to learn concepts or solve problems? Figure 1 shows the detailed description. For ontology, the learners need to know that the model could have a corresponding relationship and present the format and relation of variation in regard to scientific concepts. In terms of epistemology, learners learn models and modeling through individual representation, the process of developing knowledge, and the specific context in which to present scientific phenomenon. Through methodology, learners can build up the modeling ability to solve problems, understand the phenomenon, and connect and develop the mental models. We could say modeling is a complex process (Halloun, 1996). A total of 46 Likert scale items were developed to evaluate students’ "general" modeling ability (meaning without any specific scientific concept involved into each item) based upon the framework of modeling ability, and adapted to investigate students’ improvement of modeling ability in this study. The three dimensions of modeling ability are also embedded into the instructional design of this study. The items on the questionnaire are presented by Chiu (2007), Chou (2007) and Wu (2007).

Figure 1. The framework of modeling ability (Chiu, 2007).

Besides the foundation regarding the general modeling ability questionnaires, mentioned above, it is also important to know whether students could apply their modeling ability in a specific context. Hence, in this study, a context-based modeling ability test was developed. Halloun points out that modeling is a complex process and it includes identifying the problem, model selection, model construction, model validation, model analysis, model deployment, model application and model re-development or re-construction (1996). In this study, these eight dimensions are considered into the design of the context-based modeling ability test, mainly focus on investigating whether students could model their own models. The detail ideas of developing this test are presented in the latter section of methods.

The learning effect on cognitive apprenticeship

The idea of apprenticeship has been taken as a kind of teaching method for a long time, and cognitive apprenticeship has been accepted as a well-structured and comprehensive teaching theory. The theory of cognitive apprenticeship was developed by Collins, Brown and Newman (1989). It represents an ideal teaching environment which hopes to guide students learning through experiencing the process of experts dealing with complex tasks. Its purpose is to promote students’ self-learning and application ability. The main feature of adopting cognitive apprenticeship is to explicitly understand experts’ problem solving process and to let students learn through observation. It is revealed that students could develop more learning strategies via experts’ demonstration (Graeber, Neumann, & Tergan, 2005). Cognitive apprenticeship emphasizes learning in an authentic environment for achieving learners’ deep understanding about concepts (Collins, Brown, & Newman, 1989). Additionally, cognitive apprenticeship provides novices the opportunity to observe what experts’ doing and thinking and also makes learners obtain guidance from the experts. Cognitive apprenticeship focuses on making students actively construct their own knowledge through experts’ modeling process and teaching behavior; meanwhile, it also promotes students’ internal motivation via the authentic environment and collaboration experiences (Graeber et al., 2005). In sum, cognitive apprenticeship fits the environment for teaching and learning about the complex skills and concepts (Chiu, Chou, & Liu, 2002; Tilley, 2001).

In terms of the framework of cognitive apprenticeship, there are four dimensions considered in this learning environment: content, pedagogical methods, the sociology of learning, and the sequencing of learning activities. Namely, a good learning environment needs to consider these four dynamically interactive dimensions . The detailed points for each dimension are described in Table 1.

Table 1. The characteristics for each dimension (Collins et al., 1989).

Dimensions		Characteristics
Content		Domain knowledge Heuristic strategies Control strategies Learning strategies
Pedagogical methods		Modeling Coac hing Scaffolding and fading Articulation Reflection Exploration
Sociology of learning		Situated learning Culture of expert practice Intrinsic motivation Exploiting cooperation Exploiting competition
Sequencing of learning activities		Increasing complexity Increasing diversity Global before local skills

After cognitive apprenticeship theory was considered by Collins et al. in 1989, more recently many researchers have adopted this theoretical framework to conduct studies with good results. Based upon the significant effect on the past research results, cognitive apprenticeship theory is adopted in modeling ability instruction to teach and promote students’ modeling ability and their understanding of scientific concepts.

Research purpose and research questions

According to the theoretical background depicted above, the purpose of this study is to investigate the learning outcome of modeling ability instruction based upon the theories regarding modeling ability and cognitive apprenticeship (CA). Three main research questions are investigated in this study, which are: (1) whether the general modeling ability could be enhanced by CA and modeling instruction; (2) whether the learning outcome of scientific concepts could be improved by CA and modeling instruction; (3) whether the context-based modeling ability could be increased by CA and modeling instruction.