 |
Asia-Pacific Forum on Science Learning and Teaching, Volume 19, Issue 1, Article 1 (Jun., 2018) |
One group pre-test post-test experimental design using quantitative and qualitative research methods was carried out for this study (Krathwohl, 1997) to assess the effects of model-based inquiry on pre-service physics teachers’ epistemologies of models and on the models they constructed. Moreover, correlational research design was used to determine any relation between participants’ epistemologies of nature and function of models and their model creation. Changes in participants’ epistemologies as well as in their models and a possible relationship between the two were examined statistically for the quantitative aspect of the study. On the other hand, the purposes of the qualitative part were to evaluate participants’ models and to provide justification for the quantitative research.
Participants of this study were 11 senior pre-service physics teachers, six of whom were females. Ages of the participants ranged from 22 to 24 years. The instructional strategy in the class was model-based inquiry (MBI). Anonymity was preserved by using codes for the participants; therefore, P1 through P11 represented the pre-service teachers. The study took place in an elective course called Conceptual Physics in the physics teacher education program at a state university. The course aimed to develop participants’ conceptual knowledge of dynamics and help them reformulate their views of effective science teaching. The pre-service physics teachers took the course for 2 h/week and worked as groups when it was necessary. They chose their peers themselves. Hence, P1 worked with P2, P3 worked with P4, P5 worked with P6, P7 worked with P8, and P9 worked with P10 and P11. The students constructed their models and conducted experiments as groups; however, they wrote their inquiry reports individually.
Treatment and Instructional Context
According to Chinn and Samarapungavan (2008), instruction should engage students in thinking about their beliefs and epistemology, their general understanding of causal and non-causal models; the specific entities and activities of the models they are learning, together with how these models link to phenomena; and a variety of relations with other models. Model-based inquiry is an instructional strategy whereby learners are engaged in inquiry in an effort to explore phenomena and construct and reconstruct models in light of the results of scientific investigations (Campbell, Oh & Neilson, 2012). Therefore, the pre-service teachers studied some dynamics concepts during the model-based inquiry process. Meyer-Smith and Mitchell (1991) present some reasons for the difficulty in changing pre-service teachers’ beliefs such as short duration of course and critical timing of university-based experiences. Hence, the duration of the model-based inquiry instruction was one semester. Since the participants had not had any experience with modelling in inquiry before, the instruction in the first and second weeks of the semester focused on the implementation of model-based inquiry as an instructional method. The participants were requested to generate initial models, develop inquiry questions, propose hypotheses, do investigations and conduct experiments to test their models. They constructed three dimensional models, revised their models and used them for explanations. The instructional context is demonstrated week by week in Table 1.
Table 1. Model-based inquiry instruction.
Weeks
Model-Based Inquiry
1-2
Pre-administration of the model epistemology questionnaire. Model-Based Inquiry (MBI) was explained as a teaching strategy and some cases were given as examples.
4-8
MBI was implemented for the first activity. Worksheet for the first activity was distributed. The students were started to work as groups. They generated initial models suggested processes or structures potentially explanatory of the phenomenon. They developed inquiry questions in tandem with their models. The students stated potential relationships between variables and used their models to propose hypotheses. They conducted experiments and took measurements to test the models. Experiments were related to an inclined plane. The students also used models to collect data and evaluated the hypotheses. They modified their models if it was necessary. They started to write their inquiry reports. The students used patterns in the data and models to explain the phenomenon. They assessed and revised their models by taking into account additional evidence or aspects of the phenomenon. The students presented their final models and discussed how their models could generate different hypothesis. They argued about if their models could apply to other phenomena. They handed their inquiry reports in.
9-13
MBI was implemented for the second activity. The students followed the same procedure they did for the first activity. However, this time experiments were related to free fall of various objects with different features in various mediums.
14
Post-administration of the model epistemology questionnaire.
The participants worked on two inquiry activities during the course period. Each activity lasted five weeks. The first activity was about a half pipe in a skate park. The question was as follows: The person sitting on the deck of the half pipe as shown in the picture wants to send a ball to his friend sitting on the opposite deck by dragging it in the pipe instead of throwing the ball. If amplitude, transition and flat bottom of the half-pipe can be changed, how can the person send the ball as fast as possible? The second activity was based on the difference between Galileo’s and Aristotle’s ideas about falling objects. The participants were given a discussion in an unfinished dialogue among three people and asked to explain who was right and who was not right by providing evidence. They were also demanded to complete the unfinished dialogue. The dialogue was taken from the book written by Galileo Galilei, which was translated by Crew and Salvio in 1914. The model-based inquiry instruction has explained in more detailed in Arslan-Buyruk and Ogan-Bekiroglu (2018).
The authors of this paper are physics educators. The first author was the instructor of the course; therefore, she had two roles. One was as a teacher and the other one was as a researcher. Two researchers prepared the lesson plans and worksheets together. The first author observed and guided the groups, started and led discussions, and prevented irrelevant talk during the activities. Both authors had roles in planning the activities, conducting the research, and analyzing the data.
The pre-service teachers’ epistemologies of nature and function of models were assessed with the help of the epistemology questionnaire used by Gobert and Discenna (1997) adapted from Grosslight et al (1991). The purpose of the questionnaire is to describe students' understanding of what a model is and what it is used for. The questionnaire has nine open-ended questions. The first three questions aim to find out what students' understanding of a model is and how it is used (What comes to mind when you hear the word model? How would you describe a model to someone who didn't know what a model is? What are models used for?). The following two questions seek to assess students' understanding of how the word model might relate to the water cycle model presented (Students are given a diagrammatic model of the water cycle and asked: Can this be considered a model? Why?, Could you use this as a model? If so, how?). Finally, the last four questions ask about how models are designed and created, whether a model could be changed, and if there could be multiple models of the same phenomena (How close does a model have to be to the real thing? How do you know what to include in a model? Can scientists have more than one model for the same thing? Are there instances that would require this model or any model to be changed? If yes, what are they?). The questionnaire was administered to the students before and after the model-based inquiry instruction.
The participants’ initial and final models were evaluated by observing and asking questions to them. Their models were examined from three perspectives: the nature of models, the function of models, and the role of models in inquiry based on the rubric developed by Windschitl and his colleagues (2008b).
Analysis of Model Epistemologies: The participants’ model epistemologies were analyzed based on a conceptualization of how experts perceive models determined in the literature (Crawford & Cullin, 2004; Justi & Gilbert, 2002; Schwartz & Lederman, 2008; Windschitl, Thompson & Braaten, 2008b). According to the expert view, models can represent a system of ideas with explanatory power for some process or event, models can be created in different representational modes for different purposes (e.g., a concept map vs. a pictorial drawing), and a phenomenon can be conceptualized through models in different ways (e.g., a caloric vs. kinetic model of heat transfer for example). Moreover, applying a model to real-world circumstances must take into account the logical limits of the model as well as any underlying assumptions used to build the model. Regarding the function of models, models can be used to facilitate novel insights into a natural or mathematical system, and how they are used to predict or explain events. Models cannot be completely accurate and are almost always tentative, in the sense that they are open to further revision and development. In addition, scientists can hold more than one model for the same phenomenon depending on the context, on the purpose of the scientific research, and on the perspective of the scientist.
The pre-service physics teachers’ responses to the epistemology questionnaire were categorized as “sophisticated”, “transitional”, and “naïve” after codes were identified from their answers. Sophisticated epistemologies are aligned with expert views. Therefore, when the students answered with a naive conception of models, e.g., that models are merely small replicas of objects, their responses were categorized as naïve while their responses were categorized as sophisticated when they answered with an advanced conception of models, e.g., that models are used to reflect or explain how something functions. Specifically, similar to the Gobert and Discenna (1997)’s scoring, the students who viewed models as physical objects such as model airplanes or cars were classified as naïve whereas the students who explained models as representations of an idea or how things worked and were used to instruct, show, understand or explain how something worked were classified as sophisticated. In the case of the diagram of the water cycle, the students who stated that the water cycle model could be used as a model to show how the water cycle worked were considered as sophisticated. Regarding model building and designing, the students whose answers reflected an understanding of models as representations were considered as sophisticated. For example, the students' understanding of how close a model had to be to the real thing and what to include when making a model originated from the idea that the model had to be identical to the real object (physical view) to the model had to be close enough to be able to understand the idea (abstract view). The students having sophisticated answers had the understanding that models were a representation and tended to answer that there were multiple models and that models were changeable. Whereas, the students having naïve responses believed that models were exact replicas of the "real" thing focused on physical differences. The students’ responses were categorized as transitional if they were in between naïve and sophisticated along this continuum.
In order to determine the pre-service physics teachers’ epistemologies and to do non-parametric statistical analyses between their pre- and post-epistemologies, their responses were scored. Accordingly, naïve responses were scored as “1”, transitional responses were scored as “2”, and sophisticated responses were sored as “3”. Mean value of nine scores gathered from nine responses of each student was calculated to assign the category to the student’s epistemology. Since the mean values ranged from 1 to 3, mean values between 1 - 1.66 were categorized as naïve epistemology, mean values between 1.67 – 2.33 were categorized as transitional epistemology, and mean values between 2.34 - 3 were categorized as sophisticated epistemology.
Analysis of Models: The students’ models they created were analyzed based on the rubric whose criteria are listed in Table 2. Their models were categorized as “congruent with experts’ models”, “congruent with intermediate models”, and “congruent with novices’ models”. Their models were scored to do non-parametric statistical analyses between participants’ initial and final models. Thus, a score of “3” represented models that were congruent with those of experts, a score of “2” represented an intermediate level of models, and a score of “1” represented models that were congruent with those of novices. Mean value of three scores gathered from three perspectives (the nature of models, the function of models, and the role of models in inquiry) of each model was calculated. Since the models’ mean values ranged from 1 to 3, mean values between 1 - 1.66 were categorized as novice, mean values between 1.67 – 2.33 were categorized as intermediate, and mean values between 2.34 - 3 were categorized as expert.
Table 2. Criteria for model evaluation (Windschitl, Thompson & Braaten, 2008b).
Nature of Models
Function of Models
Role of Models in Inquiry
“3” Congruent with experts’ models
Can portray conceptual/theoretical as well as observable processes and relationships.
Represent ideas rather than “things.”
Models fallible in concept because they are based on interpretation and inference.
Models have logical limits and underlying assumptions.
Models can differ not only because of representational modes, but because a phenomenon is totally reconceptualized.
Tools to advance scientific ideas rather than only being a product of inquiry are generalizable, can be used to predict.
Tools to advance scientific ideas rather than only being a product of inquiry allow novel insights into relationships, and help generate questions for inquiry.Research questions are conceived of within the context of a model.
Hypotheses are parts of models that will be tested.
Models are revised through argument that uses data and logic, must be consistent with evidence, other models, theories.
Empirical data can be used to argue for theoretical “pieces” (structures or processes) of models.
Models can change not only as result of empirical “fine-tuning” but also because target phenomenon is reconceptualized in new way.“2” Intermediate models
Models portray processes and systems that may not be directly observable, but are taken to be real.
Models can take form of mathematical representation or set of rules.
Models of same thing can be different because there are different modes of representation.
Facilitates understanding, helps others to understand what an expert knows.
Are generalizable, used to describe different situations.
Helps analyze effects/variables of some complicated system.
Scientific inquiry is done first, then create a model based on data.
Models can help one think of things to investigate.
Hypotheses are models.
Data can be collected from models themselves.
It is important to collect data on actual phenomenon (rather than exclusively from a model) if possible.
Models are changed only if they do not match/predict data.“1” Congruent with novices’ models
Models are pictorial or physical replications of “things” considered to be real.
Object of model may be too small, too large or inaccessible to direct observation.
Relation of model to thing being modeled: object of model is more complex.
Models can be different from one another because of different “looks” at the object.To simplify, illustrate, show
Model development not recognized as part of scientific inquiry; models function only to illustrate, simplify, help communicate ideas.
Hypotheses are “best guesses” from unspecified background knowledge.
Relationships between empirical observations and theory unspecified.
Fact that data can be collected from models themselves is unacknowledged.
Argument may be synonymous with “conclusions;” directed toward determining if questions are answered rather than using patterns in data to support or refute models.Analysis of A Relationship between Epistemologies and Models: Spearman’s rank correlation coefficient test was performed to look for a relationship between the participants’ epistemologies of nature and function of models and their constructed models.
