Confounding conceptual integration
Keith S. Taber
Tuysuz and colleagues seem to have found chemistry and physics teachers have a different attitude to the importance of integrating concepts from across the subjects.
Conceptual integration?
Conceptual integration is very important in science. That is, science doesn't consist of a large set of unrelated facts, but rather the ability to subsume a great many phenomena under a limited number of ideas is valued. James Clerk Maxwell is widely remembered for showing that electricity, magnetism and radiation such as light (that is, what we now call electromagnetic radiation) were intimately related, and today theoretical physicists seek a 'Grand Unified Theory' that would account of all the forces in nature. Equally, the apparent incompatibility of the two major scientific ideas of the early twentieth century – general relativity and quantum mechanics – is widely recognised as suggesting a fundamental problem in our current best understanding of the world.
So, conceptual integration can be seen as a scientific value: something scientists expect to find in nature 1 and something they seek through their research.
Learners may not appreciate this. When I was teaching physics and chemistry I was quite surprised to see how little some students who studied both subjects would notice, or indeed expect, ideas taught in one course to link to those in another (e.g., Taber, 1998).
A demarcation criterion?
I have even, only partially tongue-in-cheek, suggested that a criterion for identifying an authentic science education would be that it emphasises the connections within science, both within and across disciplines (Taber, 2006). 2
Sadly, there has been limited attention to this theme within science education, and very little research. I was therefore pleased to find a references to a Turkish study on the topic. 3
A study with teachers-in-preparation
Tuysuz, Bektas, Geban, Ozturk and Yalvac (2016) undertook an interview study with students preparing for school science teaching. One of their findings was:
"Generally speaking, while the pre-service chemistry teachers think that physics concepts should be used in the chemistry lessons, the pre-service physics teachers believe that these two subjects' concepts generally are not related to each other."
Tuysuz, Bektas, Geban, Ozturk & Yalvac, 2016
Reading this in isolation might seem to suggest that those preparing for chemistry teaching (and therefore, likely, chemistry teachers) saw more value in emphasising conceptual integration in teaching than those preparing for physics teaching (and therefore, likely, physics teachers).
Why might physics teachers give less value to conceptual integration?
It is easy to try to think of possible reasons for this:
- Conjecture 1: chemistry teachers are aware of how chemistry draws upon physical concepts, and so are more minded to emphasise links between the subjects than physics teachers. 4
- Conjecture 2: physicists, and so physics teachers, are more arrogant about their discipline than other scientists (cf. "All science is either physics or stamp collecting" – as Ernest Rutherford supposedly claimed!)
- Conjecture 3: chemists are more likely to have also studied other science disciplines at a high level (and so are well placed to appreciate conceptual integration across sciences), whereas physics specialists are more likely to have mainly focussed on mathematics as a subsidiary subject rather than other sciences.
I imagine other possibilities will have occurred to readers, but before spending too much time on explaining Tuysuz and colleagues' findings, it is worth considering how they came to this conclusion.
Not an experiment
Tuysuz and colleagues do not claim to have undertaken an experimental study, but rather claim their work is phenomenology. It did not use a large, randomly selected (and, so, likely to be representative) sample of populations of pre-service science teachers (as would be needed for an experiment), but rather used a convenience sample of six students who were accessible and willing to help: three pre-service physics teachers and three pre-service chemistry teachers.
Read about sampling populations in research
It is not unusual for educational studies to be based on very small samples, as this allows for in-depth work. If you want to know what a person really thinks about a topic, you need to establish rapport and trust with them, and encourage them to talk in some detail – not just offer a rating to some item on a questionnaire. Small samples are perfectly proper in such studies.
What is questionable, is whether it is really meaningful to tease out differences between two identified groups (e.g., pre-service chemistry teachers; pre-service physics teachers) based on such samples. We cannot generalise without representative samples, so, when Tuysuz, Bektas, Geban, Ozturk and Yalvac write "Generally speaking…", their study does not really support such generalisation. The authors are only reporting what they found in their particular sample, and so the reader needs to contextualise their claim in terms of further details of the study, i.e., the reader needs to read the claim as
"Generally speaking, while the three pre-service chemistry teachers who volunteered to talk to us from this one teacher preparation programme think that physics concepts should be used in the chemistry lessons, the three pre- service physics teachers who volunteered to talk to us from this one programme believe that these two subjects' concepts generally are not related to each other."
Put in those terms, this is a very localised and limited kind of 'generally'.
This does not undermine the potential value of the study. That any future school science teachers might think that "these two subjects' concepts generally are not related to each other" is a worrying finding.
A confounded design
Another reason why it is important not to read Tuysuz's study as suggesting a general difference between teacher candidates in physics and chemistry, is because of a major confound in the study design. If the research had been intended as an experiment, where the investigators have to control variables so that there is only one differences between the different conditions, this would have been a critical flaw in the design.
The pre-service physics teachers and the pre-service chemistry teachers were taking parallel, but distinct, courses during the study. The authors report that the teaching approaches were different in the two subject areas. In particular, the paper reports that in the case of the pre-service chemistry teachers conceptual integration was explicitly discussed. The chemists – but not the physicists – were taught that conceptual integration was important. When interviewed, the chemists (who had been taught about conceptual integration) suggested conceptual integration was more important than the physicists (who had not been taught about conceptual integration) did!
- This might have been because of their different subject specialisms;
- It might have been because of the differences in the practice teaching courses taken by the two groups, such as perhaps the specific engagement of the chemists (but not the physicists) with ideas about conceptual integration during their course;
- It might have been due to an interaction between these two factors (that is, perhaps neither difference by itself would have led to this finding);
- And it might have simply reflected the ideas and past experiences of the particular three students in the chemists group, and the particular three students in the physicists group.
Tuysuz and colleagues found that, 'generally speaking', three students (who were chemistry specialists and had been taught about conceptual integration) had a different attitude to the importance of conceptual integration in teaching science to three other students (who were physics specialists and had not been taught about conceptual integration)
Read about confounding variables in research
The researchers might have just as readily reported that:
"Generally speaking, while the pre-service science teachers who had discussed conceptual integration in their course think that physics concepts should be used in the chemistry lessons, the pre-service science teachers who had not been taught about this believe that these two subjects' concepts generally are not related to each other."
Of course, such a conclusion would be equally misleading as both factors (subject specialism and presence/absence of explicit teaching input) vary simultaneously between the two groups of students, so it is inappropriate to suggest a general difference due to either factor in isolation.
Work cited:
- Taber, K. S. (1998) The sharing-out of nuclear attraction: or I can't think about Physics in Chemistry, International Journal of Science Education, 20 (8), pp.1001-1014. [Download the paper]
- Taber, K. S. (2006) Conceptual integration: a demarcation criterion for science education?, Physics Education, 41 (4), pp.286-287.
- Taber, K. S. (2013). Conceptual frameworks, metaphysical commitments and worldviews: the challenge of reflecting the relationships between science and religion in science education. In N. Mansour & R. Wegerif (Eds.), Science Education for Diversity: Theory and practice (pp. 151-177). Dordrecht: Springer. [Download chapter]
- Tuysuz, M., Bektas, O., Geban, O., Ozturk, G., & Yalvac, B. (2016). Pre-Service Physics and Chemistry Teachers' Conceptual Integration of Physics and Chemistry Concepts. Eurasia Journal of Mathematics, Science and Technology Education, 12(6), 1549-1568. https://doi.org/10.12973/eurasia.2016.1244a
Notes
1 Although science is meant to be based on objective observations of the natural world, scientists approach their work with certain fundamental assumptions about nature. These might include beliefs that
- an objective account of nature is in principle possible (that different observers can observe the same things), and
- that there is at some level a consistent nature to the universe (there are fixed laws which continue to apply over time)
assumptions that are needed for science to be meaningful. As these things are assumed prior to undertaking any scientific observations they can be considered metaphysical commitments (Taber, 2013).
[Download 'Conceptual frameworks, metaphysical commitments and worldviews']
Another metaphysical commitment generally shared by scientists as a common worldview is that the complex and diverse phenomena we experience can be explained by a limited number of underlying principles and laws. From this perspective, progress in science leads to increased integration between topics.
2 The term 'demarcation criterion' is often used in relation to deciding what should be considered a science (e.g., usually, astronomy is considered a science, and so is biochemistry; but not astrology or psychoanalysis). A famous example of a demarcation criterion, due to Karl Popper, is that a scientific conjecture is one which is in principle capable of being refuted.
Astronomers can use their theories and data to predict the date of the next solar eclipse, for example. If the eclipse did not occur when predicted, that would be considered a falsification.
By contrast, if a psychotherapist suggested a person had personality issues due to repressed, unresolved, feelings about their parents, then this cannot be refuted. (The client may claim having positive and untroubled relationships with the parents, but the therapist does not consider this a refutation as the feelings have been repressed, so they are not consciously available to the client. The problem can only be detected indirectly by signs which the therapist knows how to interpret.).
3 I became aware of the study discussed here when reading the work in progress of Louise Vong, who has been doing some research in this important topic.
4 Physics concepts are widely applied in chemistry, but not vice versa. So, this is suggesting that chemistry teachers have more need to refer to physics in teaching their subject than the converse.
However, we could also have looked to explain the opposite finding (had it been reported that pre-service physics teachers paid more attention to conceptual integration than pre-service chemistry teachers) by suggesting physics teachers have more reason to refer to chemistry topics when discussing examples of applications of concepts being taught, than chemistry teachers have to refer to physics topics.
