 |
Asia-Pacific Forum on Science Learning and Teaching, Volume 19, Issue 1, Article 1 (Jun., 2018) |
The Pre-Service Physics Teachers’ Epistemologies of Nature and Function of Models
One table was prepared for each open-ended question based on the codes gathered from the students’ responses. Three out of nine tables are presented here due to the page limitation. Table 3 shows the codes gathered from the pre-service teachers’ responses about what the models are used for before and after the model-based inquiry instruction. While six students had thought that models were used for perspicuity before the MBI, three more students achieved this idea after the MBI. Even though one student (P4) had claimed that models were used for representation of a reality before the MBI, she changed her mind after the instruction and stated that models were used for perspicuity. Moreover, P10 and P11 expanded their views after the instruction and wrote that models were also used for construction of new models.
Table 3. Codes gathered from the pre-service teachers’ responses to “What are models used for?”.
P / Codes
Perspicuity Making sense
Concretization
Representation of a reality
Explanation of scientific claims
Perspicuity Making sense
Concretization
Explanation of scientific claims
Construction of new models
P1
X
X
P2
X
X
X
P3
X
X
X
X
P4
X
X
P5
X
X
X
P6
X
X
P7
X
X
X
P8
X
X
X
P9
X
X
P10
X
X
X
P11
X
X
X
Total Frequency
6
2
1
4
9
3
3
2
P: Participants
The codes based on the students’ responses about what to include in a model are presented in Table 4. Whereas three students had assumed that inventions were included in a model before the MBI, they changed their thoughts after the MBI. Only P2, P4 and P5 had written that a model comprised aspects of a subject before the instruction, four more students shared this view after the instruction. In addition, six students explored that a model could contain theory, hypothesis and formulas after the MBI.
Table 4. Codes gathered from the pre-service teachers’ responses to “How do you know what to include in a model?
P / Codes
Visuality
Inventions
Aspects of a subject
Relationships
Visuality
Aspects of a subject
Theory, hypothesis, formulas
Variables
P1
X
X
X
X
P2
X
X
X
X
P3
X
X
P4
X
X
X
P5
X
X
X
X
X
P6
X
X
X
P7
X
P8
X
X
X
X
P9
X
X
X
P10
X
X
X
P11
X
X
X
X
Total Frequency
4
3
4
1
9
7
6
2
P: Participants
According to Table 5, P1 had believed that models did not change in time whereas P3, P4 and P5 were not sure about this before the modelling activities. After they had experiences with model construction, they all believed that models could change. Four students realized that models could change because of tentativeness of science and six students understood that models could change because of scientific research and technology.
Table 5. Codes gathered from the pre-service teachers’ responses to “Are there instances that would require this model or any model to be changed? If yes, what are they?.
Before the MBI After the MBI P / Codes
Yes, models of atoms changed
Yes, if the subject is changed
Yes, when new scientific discoveries happen
Not sure
No
Yes, models of atoms changed
Yes, because of tentativeness of science
Yes, because of scientific research and technology
P1
X
X
P2
X
X
X
P3
X
X
P4
X
X
X
P5
X
X
P6
X
X
X
P7
X
X
P8
X
X
P9
X
X
X
P10
X
X
X
P11
X
X
X
X
Total Frequency
3
1
4
3
1
7
4
6
P: Participants
Table 6 illustrates that all the pre-service physics teachers improved their epistemologies after the model-based inquiry instruction. Yet, due to the fact that increases in the epistemology scores of P4, P7, P8, and P9 were less than 0.66, these students’ epistemologies stayed in the same category.
Table 6. The pre-service teachers’ model epistemologies before and after the MBI
Participants
Mean Values
Category
Mean Values
Category
Differences in mean values
P1
1.29
Naïve
2.17
Transitional
0.88
P2
1.71
Transitional
3.00
Sophisticated
1.29
P3
1.43
Naïve
2.00
Transitional
0.57
P4
1.86
Transitional
2.17
Transitional
0.31
P5
1.14
Naïve
2.57
Sophisticated
1.43
P6
2.00
Transitional
2.71
Sophisticated
0.71
P7
1.71
Transitional
2.17
Transitional
0.46
P8
1.86
Transitional
2.29
Transitional
0.43
P9
1.14
Naïve
1.54
Naïve
0.40
P10
2.43
Sophisticated
2.57
Sophisticated
0.14
P11
2.00
Transitional
2.57
Sophisticated
0.57
Overall Mean
1.69
Transitional
2.34
Sophisticated
0.65
For example, P4’s belief about models stayed in the transitional stance. Her answers to the question “What comes to mind when you hear the word model?” before and after the instruction were as follows:
“What comes to my mind when I hear the world model is concrete figure of an object whose image in my brain is abstract” (before).
“Model is the mental design of a real phenomenon. For example, a model of the atom helps us to examine the atom” (after).
She wrote the following statements about whether a model could change:
“I do not have much idea about the examples of models that changed in time. However, if aspects of the things that a model represents change, the model may change” (before).
“Many ideas can change and develop in time. New findings can be added to a finding, so that it can be changed and developed. For example, Galileo measured time with water clock in his free fall experiment but we used chronometer to measure time in our experiment” (after).
On the other hand, P9 had had naïve epistemologies in the beginning and did not show much improvement. She had thought models could only be three dimensional. Her definition of models before the MBI implementation was explanation of a subject by using an object. Her definition was more explanatory but unfortunately not sophisticated after the implementation:
“Model is a copy of a case or a phenomenon that needs to be explained. This copy should be similar to the real thing. Making scale models is modelling. Photographs are also models” (after).
Her responses to the question about water cycle had inconsistencies. She did not explain why water cycle was a model.
“Water cycle may count as a model but models do not include explanations. Models are explanations themselves” (before).
“Yes, water cycle is a scientific model. My imagination can also count as a model” (after).
Before the MBI, four students (P1, P3, P5, and P9) had held naive epistemologies of nature and function of models, whereas one student (P10) had possessed sophisticated epistemology. However, only P9 kept her naïve epistemologies and five students (P2, P5, P6, P10, and P11) could develop sophisticated epistemologies after they were introduced with modelling. P2 and P5 were the ones who performed the highest progress.
For example, P2 had not answered to the question about what to include in a model. However, after he had some experiences with modelling he could wrote that models could include hypothesis, formulas, equations, schemes, and diagrams. He also expanded his views about why models were used for:
“Models are used for making something concrete” (before).
“Models are used to explain a scientific phenomenon more easily and more effectively” (after).
The progress in P5’ model epistemology can be seen from her responses. For instance, she did not recognize water cycle as a model in the beginning of the instruction because she had assumed that the cycle was a clarification but a model was a way of proofing something. Nevertheless, at the end of the MBI instruction she stated that:
“It is hard to observe the cycle of water; hence, this cycle can be shown by using models such as pictures and animations so that students’ understanding can be reinforced” (after).
P5’s ideas about tentative nature of models became more comprehensive after the MBI implementation.
“Models can change in time but I cannot give a specific example for that. If the method changes the model changes” (before).
“Scientific models are tentative. The model that was used to explain the motion was F=m.v. However, it has been changed to F=m.a. This model may also change in the future” (after).
The whole participants’ overall mean value increased from 1.69 to 2.34 showing that their transitional epistemologies enhanced to sophisticated epistemologies. The results of Wilcoxon signed-rank test for the pre-service teachers’ epistemologies before and after the MBI were in the expected direction (z = -2.936) and sum of positive ranks was significantly higher than the sum of negative ranks (p<.01) as shown in Table 7. That is, the students’ post epistemology scores were better than their pre epistemology scores (mean rank of 6.00 vs. mean rank of 0.00). These findings present that the model-based inquiry facilitated the students’ epistemologies about nature and function of models. When the students were given a chance to generate and revise models in inquiry environment, they tended to have more sophisticated views about models and understood that models were representations and tentative. This result is not consistent with the results that emerged from the research by De Jong and van Driel (2001) and Crawford and Cullin (2004) because none of their participants achieved to the highest level in their study.
Table 7. Results of the Wilcoxson signed-rank test for the epistemologies before and after the MBI
N
Mean Ranks
Sum of Ranks
z
p
Post Epistemologies-Pre Epistemologies
Negative
0
0.00
0.00
-2.936
.003
The Pre-Service Physics Teachers’ Models
The participants created two models in two activities through the model-based inquiry instruction. They started the inquiry with their initial models and at the end of the activity, they finalized their models. Table 8 reflects how their models changed in terms of nature of models, function of models, and role of models in inquiry. Regarding Table 8, the first, the second and the third groups generated better final models than their initial models during the first activity. Besides, the second group’s final model was compatible with experts’ models. Their initial model which was in intermediate level in the category of roles of models in inquiry could test more than one hypotheses; thus, they had written more than one research question. They had thought that if they increased temperature they would decreased friction. In order to minimalize the energy lost they had kept the horizontal plane short. The second group’s initial model received the score of 2 in the nature of models category because the model included both observable and non-observable processes. Their model also obtained the score of 2 in the function of models category since it could analyze a complex system. Later, the second group revised their model because it could not test some hypothesis about a relationship between temperature and friction. Their final model was congruent with experts’ models in all categories because they wanted to use their models for generalizations, they examined friction in molecular level, their research questions were conceived of within their model, and they argued about theoretical structures. Figure 1 presents the second group’s initial and final models.
Table 8. Criteria for model evaluation (Windschitl, Thompson & Braaten, 2008b).
First Activity
Second Activity
Initial Model 1
Final Model 1
Initial Model 2
Final Model 2
Groups’ Numbers and Members
Mean Values
Category
Mean Values
Category
Mean Values
Category
Mean Values
Category
1: P1, P2
1
Novice
2
Intermediate
2
Intermediate
2.33
Intermediate
2: P3, P4
2
Intermediate
3
Expert
2
Intermediate
2
Intermediate
3: P5, P6
1.33
Novice
2.33
Intermediate
2
Intermediate
2.33
Intermediate
4: P7, P8
1
Novice
1
Novice
1
Novice
1.66
Novice
5: P9, P10, P11
1
Novice
1
Novice
2
Intermediate
2.33
Intermediate
Overall Mean
1.26
Novice
1.86
Intermediate
1.8
Intermediate
2.13
Intermediate
Figure 1. The second group’s initial and final models.
However, there was not much revision between the initial and final models of the fourth and the fifth groups during the first activity. For instance, regarding the fourth group, there were inconsistencies between theoretical pieces of their model and their constructed model. While their initial model consisted of mathematical models of potential and kinetic energy, they only measured time in their final model. Their purpose of modeling was just simplification and visualization. Consequently, their final model did not have much function. Although they stated that they neglected friction, there was not any proof for that in their model. Moreover, their model did not allow for inquiry.
The results of Wilcoxon signed-rank test were in the expected direction (z = -2.449) and sum of positive ranks was significantly higher than the sum of negative ranks (p<.05) as can be seen in Table 9. In other words, the students’ final models were better than their initial models for the first activity (mean rank of 3.50 vs. mean rank of 0.00).
Table 9. Results of the Wilcoxson signed-rank test for the initial and final models
N
Mean Ranks
Sum of Ranks
z
p
Final Model 1- Initial Model 1
Negative
Positive
Ties
Total0
6
5
11.00
3.50.00
21.00-2.449
.014
N
Mean Ranks
Sum of Ranks
z
p
Final Model 2 - Initial Model 2
Negative
Positive
Ties
Total0
9
2
11.00
5.00.00
45.00-2.810
.005
N
Mean Ranks
Sum of Ranks
z
p
Two Final Models - Two Initial Models
Negative
Positive
Ties
Total0
11
0
11.00
6.00.00
66.00-2.958
.003
All of the groups, except for the second group, increased their models’ scores when they constructed their final models during the second activity (see Table 8). However, the models stayed in the same category because the differences between final models’ scores and the initial models’ scores were not bigger than 0.66. That is, the participants already created initial models which were good in terms of their nature, function and role in inquiry. With regard to the second group, there was not any change between their scores of initial and final models. Their final model neither enabled inquiry nor generalizability. Though their model allowed them to show what they already knew about free fall, it did not open for different representations.
The results of Wilcoxon signed-rank test were in the expected direction (z = -2.810) and sum of positive ranks was significantly higher than the sum of negative ranks (p<.01) as revealed in Table 9. That is to say, the students’ final models improved according to their initial models for the second activity (mean rank of 5.00 vs. mean rank of 0.00).
Moreover, Table 8 illustrates that the pre-service teachers’ initial models for the second activity (overall mean is 1.8) were more comprehensive than their initial models for the first activity (overall mean is 1.26). Similarly, their final models for the second activity (overall mean is 2.13) were more close to experts’ models than their final models for the first activity (overall mean is 1.86). The results of Wilcoxon signed-rank test were supported this finding because they were in the expected direction (z = -2.958) and sum of positive ranks was significantly higher than the sum of negative ranks (p<.01) as given in Table 9. The students gradually constructed more quality models while experiencing model-based inquiry. Their models started to represent scientific ideas and include logical limits, directed them to inquiry, and changed based on the empirical results during the study.
These findings are not much in line with the results of Windschitl and Thompson (2006) because most of the pre-service teachers in this study used models to ground their own empirical investigations. Additionally, unlike the participants of Schwartz and Gwekwerere (2007)’s study, great majority of the participants of this research recognized the role of models in inquiry.
Relationship Between Pre-service Teachers’ Epistemologies of Models and Their Model Construction
In order to find if the pre-service physics teachers reflected their epistemologies to their models, Spearman’s rank correlation coefficient test concerning the differences between two initial models and two final models and the differences between pre and post epistemologies was calculated. These analyses revealed significant positive high correlation between the development in the students’ models they constructed and the progress in their epistemologies of models (r = .80, p < 0.01) (see Table 10). That is to say, the pre-service physics teachers could put their beliefs into their practices. While some researchers (Skott, 2001; Stipek, Givven, Salmon, & MacGyvers, 2001) advocate that the influence is from belief to practice, some (Guskey, 1986; Ruthven, 1987) argue that belief is the result of practice rather than a main influence on it. Either way, the result of this study showed consistencies between beliefs and practices. Comparison of Table 6 and Table 8 discloses that P1, P2, P5, and P6 made more advance in their model epistemologies than their peers made. These participants also improved their models more. Hence, it can be said that the more progression in model epistemologies requires the more quality revision in models.
Table 10. Result of Spearman’s rank correlation coefficient test concerning the differences between two initial models and two final models and the differences between pre and post epistemologies
Variables
N
rs
P
Differences of initial and final models - Differences of epistemologies
11
.80
.003
