MadSci Network: Other

Re: Do all experiments absolutely need a controlled variable?

Date: Sun Oct 8 01:12:35 2000
Posted By: Eric Maass, Director, semiconductors / communication products
Area of science: Other
ID: 969400497.Ot

Your question makes a lot of sense, Mindy.

Actually, a control is needed in one type of experimental design, which happens to be fairly common - experiments in which you compare the current situation to a new situation. Perhaps you want to try a different way of getting to school to see if it takes less time - so you would time the way you do it now (the control), time the new way, and compare them. Perhaps a doctor has been prescribing a certain antibiotic for people who have sinus infections but is considering an antibiotic recently approved and recommended for sinus infections..he might want to have some patients try the current medication (the control group), and others the new medication, then compare how quickly and completely they recover.

However, there are other types of experimental designs. Perhaps you want to figure out a really good cake you design an experiment where you try either one or two cups of flour, one or two eggs, one or two cups of milk, one or two teaspoons of yeast, one or two tablespoons of sugar, and one or two tablespoons of cocoa. There are 2 x 2 x 2 x 2 x 2 x 2 = 64 possible combinations of flour, eggs, milk, yeast, sugar, and cocoa. This is called a full factorial experimental design. Perhaps none of the combinations is exactly the recipe you would use now ... or perhaps you have never even baked a cake before. So, there may not be a "control group", but there is still and experimental design, and you still are running an experiment, and the experiment could provide you with good insight for how to bake a really tasty cake.

Here is a website that discusses factorial experimental designs and other experimental designs. Many of these experimental designs are commonly used in a variety of industries. In the chemical industry, the combinations of chemicals and the temperatures for certain chemical reactions are varied. In the semiconductor industry, the temperatures and times of certain steps in processing can be varied. There are thousands of real life examples out there of experiments like these, which do not have a control group, but have provided ways to improve processes to make the things you use everyday.

So, in summary... many experiments involve a control or a control group... but some experiments involve different groups, none of which is defined as a control group. It depends on what you are studying and what questions you are trying to answer with your experiment.

I hope this helps.

Current Queue | Current Queue for Other | Other archives

Try the links in the MadSci Library for more information on Other.

MadSci Home | Information | Search | Random Knowledge Generator | MadSci Archives | Mad Library | MAD Labs | MAD FAQs | Ask a ? | Join Us! | Help Support MadSci

MadSci Network,
© 1995-2000. All rights reserved.