How to Design a Scientific Experiment

A great deal of research is necessary when creating and executing a scientific experiment.

The idea of a scientific experiment is to come up with a hypothesis based on a series of observations, then to use research to try to prove or disprove your hypothesis. Depending on the hypothesis, you will need to use a variety of types of research to decide how to set up an experiment that will test your hypothesis.

Instructions

1. Make observations. In order to set up a scientific experiment, you will need to make observations about a particular phenomenon that you wish to investigate further. For example, you notice that birds eat more from a feeder in the winter than in the summer.

2. Make a hypothesis. A hypothesis is an educated or logical guess about how or why something occurs. For example, using the bird feeder scenario again, you may make a hypothesis that would say something to the effect of: "Cold temperatures make birds eat more."

3. Design and perform an experiment. You want to create an experiment that would test your hypothesis and the end result would be that the evidence from your experiment would either support or disprove your hypothesis. In the bird feeder scenario, you might set up an experiment that would have a group of birds with a feeder in a warm environment and a group of birds in a cold environment, and measure how much each group eats.

4. Collect data from your experiments. You need to have several consistent dependent variables that rely on your independent variable in which you collect data. In the bird feeder situation, you would want to measure the exact temperatures of the rooms in which the birds are kept, and how much food is eaten by the birds.

5. Check to see if your data from the experiment supports or rejects your hypothesis. If the data coincides with what you predicted (in the bird feeder case, the birds in the cold room ate more than those in the warm room), then your evidence supports the hypothesis. If the data does not coincide with your hypothesis, then you reject your hypothesis and come up with a new one. From there, repeat Steps 3 through 5.

Tips & Warnings

• Always have a control group when performing an experiment. You need to have a group of untested subjects so you have a baseline to compare to with the experimental group.

• Replicate experiments to ensure your initial results were not a one-time event. The experiments should all have the same results in order to support or disprove the hypothesis.

• Beware of psychological factors when setting up a scientific study. If you are using people, for example, and tell them about possible side effects of a drug you are testing, some may show signs of a "placebo effect" which would cause those taking a placebo (your control group) to show signs of being on the drug. A good way to avoid this is to perform a "blind study" where both groups of subjects do not know if they are getting a drug or a placebo, and to not tell them of side effects before hand.

Resources

• Read this Article in Spanish
• Scientific Inquiry