Teaching about nature of science through short lab activities in Hong Kong classroom

The research is pretest-posttest, quasi-experimental design, with two classes of twenty nine and thirty four grade 9 students taught respectively with the manipulated lab inquiry of apple browning and the case of dinosaur extinction as a control. The main research question is: What is the effectiveness of the manipulated lab inquiry of apple browning, as compared to the case of dinosaur extinction, in fostering students’ understandings about underdetermination of scientific theory?

The study was conducted in a secondary girl school in Hong Kong, which admitted the top 20% of primary students and was using English as a medium of instruction. The two classes taking part in the study had no large differences in academic performance and other background factors since the students were randomly assigned to the two classes from grade eight. The two classes were taught by the same biology teacher, Miss Chung, in order to reduce the impacts of the teacher variable. Miss Chung had taught biology and science for over 7 years. She is a competent science teacher with a master degree in science education, and had learned about the nature of science and its teaching in the teacher training courses. In the study, she worked closely with the researcher to develop detailed lesson plans for the two classes. The lessons were observed by the researcher to ensure that the teaching plan had been enacted properly.

Apple browning investigation

The investigation is outlined in Table 1. Different groups in a class were asked to design the investigation as guided by two different task sheets. One suggests exposure to oxygen as a hypothesis for apple browning and thus emphasizes the antioxidant property of ascorbic acid solution in the prevention of browning. Another hypothesis suggests enzyme action as the cause of apple browning and points to the acidic nature of ascorbic acid solution in preventing browning. Actually both are required for browning: the polyphenols in apple are oxidized into brown substances by polyphenol oxidase when exposed to oxygen (Nicolas el al. 1994). Therefore, when apple browning is inhibited by ascorbic acid solution, the result gives support to both hypotheses – they are underdetermined by the empirical observations. In the post-lab discussions, when students were reporting their findings and conclusions, they were surprised by that the same test and data could lend support to two different hypotheses. This formed a discrepant event that confronted the preconceptions of the students that scientific conclusion can be straightforwardly proven by data. Then, using an interactive dialogic approach, the students were guided to explicitly reflect on their preconceptions about the reliability of scientific conclusions or scientific knowledge in general, from which the concept of underdetermination and other related NOS aspects were explained.

Table 1. Manipulated lab inquiry used to teach about underdetermination of theory

	Student groups given the oxygen hypothesis	Student groups given the enzyme hypothesis
Problem	Why does apple turn brown when peeled and exposed in air?
Hypothesis	Exposure to oxygen	Enzymes released when the cells are damaged during peeling
Testing	Compare the degree of browning of apple slices with and without being immersed into ascorbic acid solution.
Results	Apple slices treated with ascorbic acid solution show less browning than those without treatment.
Conclusion	Apple browning is caused by exposure to oxygen, since ascorbic acid as an antioxidant can prevent browning	Apple browning is caused by enzymes released when cells are damaged, since ascorbic acid can prevent browning by denaturing the enzymes
NOS conception	Both conclusions are supported by the data and thus underdetermined.

Case of dinosaur extinction

As a control for the manipulated lab inquiry, the case of dinosaur extinction was used to illustrate the underdetermination of theory. There are two rival theories explaining the extinction of dinosaur 65.5 million years ago. The asteroid impact theory attributes the dinosaur extinction to a big asteroid impact producing dust that blocked the sunlight for a long time, leading to hostile climate changes and shortage of food for the dinosaurs (Schulte et al. 2010).The volcanic eruption theory, however, regards the frequent volcanic eruptions as the main cause for the climate changes leading to dinosaur extinction. The asteroid theory was supported by a huge crater formed around 65.5 million years ago, the special quartz formed by strong impact and large amount of soot left by catastrophic fire. One piece of evidence is particularly crucial: the soil layer at the time of dinosaur extinction is rich in iridium, an element rarely found on earth surface but rich in asteroids. However, much of the evidence for the asteroid impact theory can be accounted for by volcanic eruption theory as well. Volcanic eruption can also cause big fires, and most importantly, magma is also rich in iridium. The evidence is thus unable to determine which theory is correct - underdetermination. The controversy has been lingering for over 30 years until recently a group of scientists claimed that asteroid impact is the main cause for dinosaur extinction after reviewing all the available evidence (Schulte et al. 2010).

In the classroom, the teacher first introduced the impact theory as a ‘perfect’ explanation to the dinosaur extinction. Then, the arguments of the volcano theory were brought out as a discrepant event that confronted the students’ understanding about science, before the concept of underdetermination of theory was introduced. The teaching largely paralleled that of the apple browning lab inquiry except that it was not hands on and was about authentic science.

Assessment of Understandings of underdetermination

The assessment of NOS understanding in this study does not presume that one's NOS views are necessarily universal, domain general and context independent, which are the major problems of many NOS instruments (Lederman, Wade, & Bell, 1998). Therefore, the tests were constructed respectively in the contexts of the apple browning experiment and the case of dinosaur extinction for the treatment and control classes. The two tests consist of identical items: two-tier multiple choice questions that first ask students whether they think experimental conclusions/scientific theories are absolutely true and the reasons underlying their views (Table 2). The tests were administered as pretest and posttest. The tests were developed after reviewing the literature (e.g. Lederman et al. 2002; McComas and Olson 1998) and trialed on 13 students with post-test interviews. The test was then examined by another science educator to establish face and content validity. However, due to the small sample size of this study and the categorical nature of the test items, no statistical data were obtained to support the validity and reliability of the test, which would require further work on that.

Sandoval (2005) has distinguished two types of NOS understandings: practical epistemologies and formal epistemologies, with the former pertaining to how students view their school experiments and the latter about how they view formal science. To find out how students’ practical and formal epistemologies are affected by different NOS teaching approaches, each class was assessed by an additional post-test in the different context: treatment class in the context of dinosaur extinction, and control class in the context of apple browning investigation.

Table 2. Assessment items for students' views of scientific theory as truth and their underlying reasons

I think one of the theories/hypotheses is (Choose only one):
	1. absolutely true (will never become wrong in the future).
	2. true at the moment but it may become wrong in the future.
	3. never absolutely true.
because (can choose more than one):
	a. The results have proven the conclusion/theory without doubt. (Absolute proof)
	b. The experiment/evidence may have errors. (Errors)
	c. The existing results are not strong enough to prove the conclusion/theory true. More evidence is needed. (Inadequate evidence)
	d. The same results can also support another conclusion/theory. (Weak underdetermination)
	e. The same results can support an infinite number of conclusions/theories. (Strong underdetermination) f. Even if ascorbic acid can prevent apple browning, it does not necessarily mean that the browning is caused by an enzyme. Or, Even if there was really a meteorite impact or volcano eruptions happening 65 million years ago, it does not necessarily mean that they had caused the dinosaur extinction. (Contextual underdetermination)