Title: Effects of guided inquiry, virtual and physical laboratories on conceptual understanding, inquiry performance, scientific inquiry self-efficacy, and enjoyment
Authors: Siti Jamiatul Husnaini and Sufen Chen
First author’s institution: National Taiwan University of Science and Technology
Journal: Physical Review Physics Education Research, 15 010119 (2019)
While conducting experiments is an important part of the physics curriculum, they are often resource intensive and thus physics programs in developing countries may not have access to the resources needed. However, now that computers are common place in schools and universities, virtual labs may be able to give students the experience of conducting experiments without the need for specialized equipment. In addition, using virtual experiments can eliminate unrelated factors such as equipment problems and setup issues which may help prevent students from becoming discouraged and reduce the amount of mental focus (the cognitive load) needed to do the experiment by only requiring students to focus on the experiment and not also needing to understand the equipment. The goal of today’s paper is to compare these two approaches: do physical and virtual labs provide similar benefits or does one type of lab offer superior benefits compared to the other.
So why might we expect a difference in outcomes between physical and virtual labs? First, physical labs require the use of equipment which utilize multiple sensory inputs, which may lead to better learning outcomes and learning outcomes beyond just conceptual knowledge. However, these multiple sensory inputs could also result in a higher cognitive load, leading students to spend less “brain power” on developing their understanding and hence worse learning outcomes. Alternatively, virtual labs can provide visualization of phenomenon that would not be possible with physical labs. For example, virtual labs can depict the current as moving electrons while this would not be observable in a physical circuit.
The researchers randomly assigned 68 10th grade students at a public secondary school in Indonesia to either conduct a pendulum experiment relating the period of the pendulum and the length of the string physically or virtually through a PhET simulation (shown in figure 1).
The students in the physical group were given the materials to build a pendulum and used cell phone apps to measure distances, angles, and the period. Before the experiment, all students were given a pre-test and a questionnaire to measure their conceptual understanding of pendulums, the self-efficacy (their belief that they are able to perform a specific task or process) and enjoyment. The students were then given a worksheet to guide them through the inquiry process, with prompts and questions about the purpose of the experiment, the concepts needed to analyze the pendulum, the experimental question and variables, the planning and execution of the experiment, the data analysis including analyzing errors and making graphs, how to interpret their results, and designing a new experiment. After completing the experiment, all of the students completed the conceptual test and questionnaire again.
To analyze their data, the researchers split the conceptual questions into simple and difficult questions based on the number of students who answered each question correctly on the pre-test. The researchers found that both students who did the physical labs and students who did the virtual labs showed improvements on the questions rated as “simple” on the conceptual test, but they did not improve on the “difficult” questions. The students who performed the experiment virtually did better on the “difficult” questions than the students who performed the experiment physically; however, the pre-test scores of students who performed the experiment virtually were already higher than the post-test scores of the students who performed the experiment physically, suggesting that groups may not have started with equal levels of conceptual knowledge.
Next, the researchers examined how students from each experimental setup performed on the worksheet. They created a rubric to assess each area and assigned students points based on the number of criteria met. Comparing the scores between the two groups, the researchers found that the students who performed the experiment virtually and students who performed the experiment physically performed equally well on analyzing errors and plotting results. However, the students who performed the experiment physically performed better than the students who performed the experiment virtually on planning the experiment, conducting the experiment, and describing improvements to the experiment while the students who performed the experiment virtually did better than the students who performed the experiment physically on interpreting data and developing a new research question and experiment.
In terms of measures of self-efficacy and enjoyment, the students who performed the experiment virtually showed moderately increased self-efficacy while students who performed the experiment physically showed no such increase. However, there was no significant difference in reported self-efficacy on the questionnaire give post-experiment between the two groups. In addition, the reported enjoyment increased for both groups and there was also no significant difference between the groups.
Discussion and Takeaways
The results do not point to a clear winner between doing labs physically versus virtually. However, the researchers did make a few observations worthy of future investigation. First, while the physical lab seemed to promote more of the inquiry skills, it may have been a result of the students who performed the experiment virtually’s beliefs. For example, prior work has shown that undergraduates expect computer simulations to be perfect , which may explain why the researchers observed the students who performed the experiment virtually going about their investigation less systematically than the students who performed the experiment physically.
Second, cognitive load seems to have played into the results. The students who performed the experiment virtually did better on the task at the end of the worksheet than they did on tasks at the beginning of the worksheet. By the end of the experiment, the students who had to balance the task with using the equipment may have used more “mental energy” which is why they did not do as well on the sections towards the end of the worksheet.
Finally, the students who performed the experiment virtually were observed to have spent more time doing trials than the students who performed the experiment physically. The researchers commented that the students who performed the experiment virtually seemed to be approaching the experiment with a trial and error mentality and thus may not have had enough time to adequately work through all parts of the experiment.
So what are the takeaways then? First, neither virtual labs nor physical labs seem to be better than the other as each offers its own benefits. The virtual labs seemed to improve students’ understanding of difficult concepts and improve their self-efficacy while the physical labs seemed to promote more inquiry skills. Based on the results of this study, it appears that the best practice is to incorporate both types of labs into the physics curriculum.
I am a physics graduate student at Michigan State University and the founder of PERbites. I’m interested in applying machine learning to analyze educational datasets and how students use computation in physics courses.