W. P. Hanage comes up with a formula to understand scientific articles. This formula consists of questions to ask to address the validity and real-life application of these journals. Thus, this blog will address each question and the reasoning behind these questions.
Firstly, can the experiments in these scientific articles detect differences? Ultimately, every scientific article aims to use an experiment or existing evidence to address a question. What is the difference that we are observing in the data? What stands out? Do the elements that stand out have any significance in the whole picture? In conclusion, the first question asks the scientists to address and clarify the difference found in the data.
Secondly, does research suggest a correlation claim or a causation claim? A correlation is best explained as a factor is influencing the outcome of a phenomenon, but not directly causing it. For example, “smoking causes lung cancer” is a causation statement because there is numerous evidence to suggest so. However, “smoking may be related to impotence” is a correlation statement because there is evidence but it is not enough to establish a causal claim.
Thirdly, what is the mechanism? This question addresses how the outcome of an experiment can be addressed. Thus, this question ties back to the first question. Scientists observe the difference in the data obtained in their research. Then, it is now their time to establish a reason behind the difference in the data. The hypothesis comes into play here, as it provides an educated prediction as well as an explanation.
Fourthly, how much do experiments reflect reality? This question addresses how the experiment relates to real life. An explanation may apply to a single case, but not a different case. For example, medication dosage is different between people of different weights and ages. 1000 milligrams of paracetamol is helpful to reduce pain in adults and children over 16. But what about children under 16?
Finally, and the most important question, could anything else explain the results? This is known as the confounding variable question. Confounding variables are factors that affect both a factor and an outcome that is the result of a factor. For example, the lack of exercise in pregnant women leads to their weight gain. But a confounding variable here would be these women are carrying a fetus inside their bodies.
