How to Read Science Research, Chapter 5: How Experiments are Designed
In my last post, I discussed some of the terminology used in science. Today, I will explain a bit about how science experiments are designed.
The following is an excerpt from my recent book, 28 ACT Science Lessons.
Science begins with a hypothesis, which is a logical inference based on prior research and scientific knowledge. An experiment is designed to test a hypothesis and should include exactly one independent variable (or input) and one or more dependent variables (or outputs). A variable is something that can change. Ideally, an experiment includes two samples or groups: an experimental group, which receives the “treatment” that is being tested, and a control, which does not. The control is often the sample, group, or trial that gets 0 of something or the normal amount of something.
The experimental and control groups should differ only by the independent variable, but all other factors should remain constant (the same). The greater the sample size or the larger the number of repetitions, the more reliable the results, since they are less likely to be caused by randomness or error.
A scientist analyzes the data collected from the experiment and draws a conclusion, which will state whether the hypothesis was or was not supported by the experiment. If a hypothesis is tested many, many times and repeated by many, many scientists, with the same conclusion every time, that hypothesis becomes a theory. Scientists look into the data as a whole, often looking at general trends and averages, and sometimes eliminating outliers, or a few data points that are very far off from the rest and could be due to error.
For example, let’s say a scientist wants to see whether a medicine is effective at treating headaches. That scientist would find many people with headaches, who are as similar to each other as possible. Half of those people, the experimental group, will receive the medicine, while the other half (the control group) receives a placebo. A placebo is a treatment that looks and feels like the experimental treatment, but doesn’t have the actual treatment in it. The idea behind a placebo is that the people involved in the experiment don’t know whether they are getting the medicine or not, as this knowledge might affect the results. In this experiment, the independent variable is whether people received medicine or placebo and the dependent variable is what happened to their headaches.
Stay tuned for Chapter 6: Deciphering Big Science Words