Rockhampton, Queensland

Questions of how people use analogies to acquire and create knowledge have occupied scientists, educators and psychologists for more than 40 years (e.g., Oppenheimer, 1956; Glynn, 1991; Gentner & Medina, 1998). The popular view of science is that it is a logical and rational process; therefore, it is understandable that scientific, empirical and cognitive methods should dominate science analogy research and discussions about the importance of analogy in discovery and learning (e.g., Clement, 1993, Dupin & Johsua, 1989; Gick & Holyoak, 1983). The rational approach (sometimes called the cold-rational view) has yielded significant benefits for teachers in the form of Glynn's Teaching-With-Analogies model and the FAR guide (Treagust, Harrison & Venville, 1998); however, rational research studies have little to say about the reasons why teachers and students are attracted to analogy in the first place. Thus, the affective dimension of analogy use is the subject of this paper.

The interest-generating power of analogies often is anecdotally reported by teachers but when interest is mentioned in analogy research, it is incidental and taken as a given rather than a focus of the study. Certain concepts like refraction can only be explained by analogy (Harrison, 1994) and analogies are widely accepted as conceptual change tools (Dagher, 1995a). The specific use of analogy to raise learners' interest levels is rarely reported. Indeed, Pintrich, Marx and Boyle's (1993) review of the motivational literature on conceptual change showed that affective factors are largely ignored in conceptual learning studies. They argued that affective or "hot-irrational" issues are just as important in concept learning as cognitive issues because interest determines the level of student engagement with cognitive activities and explanations. I argue that the interest-dimension of analogy and model use should be scrutinised alongside their concept learning potential.

Two detailed reviews of analogy use in school science (Dagher, 1995b; Duit, 1991) concentrate on the conceptual growth potential of analogies and devote limited attention to their interest-generating power. In his review, Duit shows that analogies are effective conceptual change agents because they enhance understanding by making connections between scientific concepts and the students' life-world experiences, and by helping students visualise abstract ideas. He points out that analogies "provoke students' interest and may therefore motivate them" (1991, p.666) but interest is the last factor in his list of analogy's advantages. Duit explains in detail the constructivist benefits of analogy, but does not explore the motivational power of analogies and models. The absence of detail pertaining to interest and motivation is easily explained: few studies of analogy discuss this factor.

Analogical thinking extracts useful structural and relational information from a learner's repertoire of familiar instances or events (the analog or source) and maps it onto the unfamiliar science concept (which is called the target). The familiar teaches the student about the unfamiliar. Analogies are interesting and motivating for students when the teacher's analog can be enriched from the students' own experience. If, however, the analog is unknown to or poorly visualised by the student, then s/he will feel marginalised or frustrated and this will lower his or her interest in the analogical discussion. Interest and engagement are, therefore, crucial to learning - it is "important to begin to build the connections between the motivational and the cognitive components of student learning" (Pintrich et al., 1993, p.168).

The depth or rigour of an analogy also is an important determiner of learning outcomes. Gentner and Markman (1997) show that relational learning (as opposed to instrumental knowledge) is the desideratum of effective analogies; thus, the analogies discussed later in this paper are limited to ones that enhance relational understanding. For example, in the wheels analogy for the refraction of light (Figure 1), a pair of wheels rolls obliquely across the boundary from a hard surface (where they roll fast) onto a softer surface (where they roll slowly). The wheels' direction bends towards the normal because the wheel that first crossed onto the soft carpet slows down and covers less distance than the other wheel (still on the hard surface). A ray of light passing from air to glass also slows down on entering the glass. If the ray strikes the boundary at an angle, one side of the ray enters the glass first and slows before the other side of the ray. Because the slowed side is covering less distance than the other side, the light ray bends towards the normal. Students can explore this analogy by rolling wheels from card to carpet and carpet to card. If the wheels are coated with paint, the tracks are recorded and students see the similarity between the behaviour of the wheels and the refracted light ray. In our experience (Harrison & Treagust, 1993), students are fascinated by this phenomenon and enthusiastically apply it to other refraction problems.

Investigating the affective dimension of such an analogy is facilitated by Pintrich et al.'s (1993) motivational and self-efficacy theories. This research program helps us understand the benefit of examples like the wheels-refraction analogy and Thagard (1989) also claims that motivational factors enhance conceptual change. For these reasons, I decided to review my own and joint studies to find instances of analogical affect on concept learning. The reviewed studies comprise Harrison (1992, 1997, 2001), Harrison & Treagust (1993, 2000, 2001), Treagust, Harrison, Venville & Dagher (1996), and Treagust et al. (1998).

The method involved scrutinising published papers and unpublished theses to identify potential instances where interest and motivation influenced the learning outcomes of planned taught analogies. The concepts and their supporting analogies were taught by a range of experienced teachers including the author. The student ages cover Years 8-11 and the teaching took place in middle school science, chemistry and physics classes. As the cases already are reported in detail, their choice was a purposeful on the ground that they contained instances of analogical affect (Patton, 1990). Each case was chosen because the evidence showed that students found some part of it interesting. Nevertheless, for each case, the original data were revisited to clarify events and some new data are added. Analyses like this often are more subjective that the original article because of the time lapse; however, in a converse sense, new interpretive perspectives can add to the data's meaning and impact.

Issues of space also mean that the following vignettes are compressed and the reader is referred to the original articles should s/he desire more information regarding theory, method and interpretation. The intent of this paper is to revisit the cases to draw new sense and to generate research questions that will guide future study into the affective dimensions of science analogies and models.

The wheels analogy was first reported by Harrison (1992) and published by Harrison and Treagust (1993). The setting was a Year 10 physics class at an all-girls independent school and the girls were taught by Mrs Kay (pseudonym). Mrs Kay was identified by her colleagues as an expert physical science teacher and she was interviewed to identify the analogies she used in her lessons. In the interview she expressed a desire to trial Paul Hewitt's (1992) wheels analogy for the refraction of light (p.437). At the time, she was teaching optics to two Year 10 physics classes and one class had already studied refraction; thus, the analogy could only be used with the second class. The situation was fortuitous because both classes were of comparable ability and achievement enabling Treagust et al. (1998) to later study the conceptual change differences between the class taught with the analogy and the one without the analogy. A motivational event associated with the second study will be discussed later in this paper.

Teaching the wheels analogy. The lesson in which the wheels analogy was demonstrated was audio-taped. Mrs Kay was interviewed pre- and post-lesson and nine girls were individually interviewed post-lesson. The lesson and the interviews were transcribed and provide the comments that illustrate the students' interest.

The lesson. Mrs Kay, led her students through the analogy with considerable panache, First, she demonstrated how light bends towards the normal as it enters a perspex block and then bends away from the normal when it exits the perspex. The top diagram in Figure 1 depicts the set-up: a light box and a large rectangular perspex prism were blue-tacked to the white board. The room was darkened and the ray bending was demonstrated, and questions like "why does the ray bend towards the normal when it goes from air into perspex" were asked.

Mrs Kay introduced the analogy by showing the hard card butting up to the soft carpet, the wheels and some bright fluorescent paint. The students were attracted to the paint and wanted to know how it was to be used: "what if we spill it ...". Mrs Kay first conducted a 'dry-run': she rolled the wheels so that they obliquely crossed from the hard card to the soft carpet. She drew attention to the change in the wheels' direction as they slowed down. Then it was time for the paint. She invited a student to liberally coat a pair of wheels with paint - "this is what all the paint's for, ... coat the pair of wheels with this nice fluoro paint ...". This was motivating because the bright paint was already a point of interest. Mrs Kay then asked another student to push the paint coated wheels so that they rolled obliquely from the hard card onto the soft carpet. As the wheels rolled across the join, the wheel tracks bent towards the normal. The students were impressed and were talking their way through the analogy with the teacher when a student interjected with

Mrs Kay took this up and added

In the interviews that followed, several students related the 'wheels slowing down and turning' to experiences they had of driving farm vehicles on sandy tracks and skateboards running off concrete paths and onto grass.

Discussion. This exchange is significant because the students' concept learning was enhanced by the spontaneous introduction of familiar experiences. This helps explain why some analogies are effective and others are not. When the students are able to make connections between the analog-target and their experiences, they are motivated to explore the analogy in deeper and more meaningful ways. In this case, I suggest that the student interest that began with a superficial attribute (the fluoro paint) grew as they added everyday examples and this helped them 'analogise the analogy' to their life experiences. Personalising the analogy enhanced relational concept mappings of the type advocated by Gentner (1983). With the conversation focused on everyday matters, the students were excited and willingly sought connections and like examples. The opportunity for them to tell their stories strengthened the conceptual links that they were making by adding personal relevance. Successful connections like these enrich the analogical mappings because the students see how the science concept explains their everyday happenings.

To verify that the students had developed a relational understanding of the wheels-refraction analogy, they were asked if they could explain why light bends away from the normal on leaving the perspex. Cara explained that light bends

Note. These vignettes are part of an extensive case. Should the reader need more theory, method and data, s/he is directed to Harrison and Treagust (1993). The point of the vignettes is to show that the wheels analogy was cognitively effective and affective because it captured the students' interest.

The knowledge that one Year 10 class studied refraction with the analogy and the other without enabled a second study (of conceptual change effects of the analogy) to be conducted three months later by Treagust, Harrison, Venville and Dagher (1996). In this interview-about-instances study, 25 of the 29 students who were taught refraction with the wheels analogy were interviewed and 14 of the 25 who were taught by the same teacher but without an analogy also were interviewed (all with the same protocol). The explanation provided for the second class stated that the side of the light ray that enters the perspex first slows down before the other side of the ray, thus the trailing side travels further than the preceding side forcing the ray to bend. End of the school year constraints made it difficult to integrate student and researcher availability, thus the maximum length of each interview was 15-20 minutes. Six students from the analogy class and 11 from the second class were absent or unavailable for interview.

This second study was an empirical study of cognitive change; however, motivation and interest appear to be important conceptual change learning factors. While this study was not searching for hot-emotive, interest-based events, one such case emerged during an interview with Dana. On her own admission, Dana was unable to explain refraction, insisted that she was "no good at science" and told how she had failed the optics test. I will argue that her lack of success in the optics topic is related to the type of questions that are asked of students like Dana

WATER

Sketch 2 Sketch 3

Figure 2: The three interview-about-instances diagrams used in the interviews.

Dana's sketch

cardboard

1 1

AIR

WATER

Figure 3: Dana's early (1) and later (2) responses to the torch in water questions

Discussion. Treagust et al. (1998) discuss in detail the nature of Dana's conceptual change once the analogy entered her thinking; however, the salient issue in this paper is the role of the analogy in arousing her interest and motivating her to become involved with the problems and talk about the problem. I suggest that there is a signal lesson here. Dana is not alone, she belongs to a large group of students for whom physics (and other sciences) lacks relevance, interest and opportunity for them to grow in knowledge. Students like Dana do not fail science, science fails them because it does not excite their interest. School science no longer has an elitist mandate to select those students who are capable of tertiary science studies. Disinterested students cannot be ignored as "lacking in knowledge and/or ability". School science has a warrant to productively engage students in exploring the science knowledge that affects their lives. In their Beyond 2000 report, Millar and Osborne (1999) ask "Who is school science education for?" They answer by insisting that "teaching science is to enable young people to become scientifically literate - able to engage with the ideas and views which form such a central part of our common culture" (all p.2006).

Ian's desire to interest his students highlights a problem familiar to every classroom teacher: despite the best intentions of the teacher and the provision of excellent learning resources, many students who should learn, do not. The provision of ideal cognitive conditions cannot overcome student reluctance to learn when the student lacks motivation and therefore does not want to engage in the learning activities. "Three aspects of an individual's behaviour - choice of task, level of engagement, and willingness to persist at the task" determine the student's level of engagement and his or her learning outcomes (Pintrich et al., 1993, p. 168). The popular constructivist principle that it is the student who decides whether or not to join in and learn can insulate us from the importance of interest and excitement. This is the task of the teacher and interesting analogies and models make excellent motivators.

Theory states that the forces between nuclei and electrons in covalent bonds are purely electrostatic. The valence shell electron pair repulsion (VSEPR) theory uses the repulsion between electron pairs around a central nucleus to explain linear, triangular planar and tetrahedral molecules (e.g., ethyne, ethene and methane respectively). Students have difficulty visualising abstract repulsion between adjacent covalent bonds. They are, however, familiar with the elasticity of inflated balloons.

3. Two balloons forming a linear shape

The model's sudden change in shape as one balloon was unexpectedly burst was "attention-grabbing". This promoted interest and when later asked to explain their understanding of molecular shapes three students used the balloon "game" to explain the forces between four, three and two bonds respectively and the resultant shapes (Harrison, 1997).

During the class presentation, the teacher pays careful attention to the students' familiarity with the analog and identifies the Like attributes (i.e., those features held in common) and the Unlike attributes (i. e., those features not held in common) of the analog and the target. To achieve this, the features of the analog and target are socially negotiated with students, similarities are drawn between them, and ways that the analog and target are not alike are explicitly identified. This is the Action phase of analogical teaching and usually involves no more than three cognitive steps: (a) familiarity with the analog, (b) mapping of the shared attributes, and (c) discussing with the students where the analogy breaks down.

3. Reflection. Following the presentation of the analogy, the teacher reflects on the analogy's interest, clarity, usefulness and conclusions. Were the shared and unshared attributes clear and adequately discussed? How can the balloons analogy be improved to make it more interesting (do the students want to repeat it themselves?) and more cognitively effective? The Reflection may take place within the lesson itself or after the lesson when preparing for the next lesson. In practice, the three phases are not distinct but merge into one another. Because Reflection is a characteristic of all good teaching, competent teachers will likely implement this step as a matter of course. Indeed, the FAR guide emerged from many pedagogical discussions with experienced teachers.

The examples featured in this paper - the wheels analogy, the interviews-about-instances study, the two teacher interviews and the balloons analogy show that familiar analogies have the capacity to interest and excite students. Not all analogies, however, are this effective. The Melbourne Cricket Ground analogy was used by Harrison (1997) to analogically model the spaciousness of an hydrogen atom; but too few students in the class were interested in football and cricket to enable them to access to the benefits of this analogy. Hence, a familiar alternative was needed. An effective analogy, based on the proximity of each student's home to the school and different sized sport balls was used to model the nucleus, electron and the space between them in a hydrogen atom.