Unlock Content Over 8,500 lessons in all major subjects
Get FREE access for 5 days,
just create an account.
No obligation, cancel anytime.
What are the requirements of a scientific experiment? How do scientists turn hypotheses into theories and laws? Learn the answers to these questions and more in this lesson on the design of scientific experiments.
Have you ever thought about what goes into a real scientific experiment? Most of us get to do science investigations when we're young. School experiments are easy and fun. But what about real scientists, like chemists, physicists, and medical researchers? What kind of work do they have to do? How do they go about designing their experiments in the laboratory?
Experiments, remember, are one of the key components of the scientific method, which is a set of procedures that scientists follow to gain knowledge about the world. Other components of the scientific method are questions, hypotheses, observations, analyses, and conclusions. While experiments are only one part of a scientific investigation, they end up being accountable to the other elements. In this lesson, we'll see what goes into making a valuable experiment and learn how scientists design useful investigations.
A scientific experiment is an ordered investigation that attempts to prove or disprove a hypothesis. So its primary purpose is to test whether someone's prediction is correct. In designing experiments, scientists have to answer some pretty complicated questions, like: Does my experiment answer the question I'm trying to solve? Does it adequately test my hypothesis? Can I make observations about the results of my experiment, and will I be able to analyze those results? Finally, if I run this test, will it allow me to come up with some kind of conclusion?
Scientific experiments are different from other kinds of tests because they are required to fit in with the scientific method. Another important factor is peer review by the science community. A scientist's work isn't generally recognized unless he follows the standards set by other scientists around the world. A few basic rules apply to the design of a good experiment. Let's take a look at what a science experiment needs.
Rule #1: The experiment must show that a hypothesis is either supported or not supported. In science, we try to avoid using terms like 'right' and 'wrong,' and we don't say that hypotheses are 'proven' or 'disproven' until we're really sure about it. A single experiment is not enough to prove anything with 100% certainty.
Rule #2: The results of an experiment must be measurable and objective. Scientists use standard units to measure different properties like length, time, volume, mass, and speed. Sometimes we need special equipment to observe things in a measurable way. For example, we can't see ultraviolet light or hear infrasonic sounds. We need special devices to detect and measure those properties for us.
Rule #3 for scientific investigations: The experiment must be repeatable by other scientists. Peer reviewers want to make sure that other scientists can run the same experiment and get similar results. This is one of the reasons we standardize our measuring tools and equipment. Scientists must be able to read anyone else's report, follow the steps exactly the same way, and compare their findings to the original test. In science, new ideas aren't taken seriously until many scientists have tested them many, many times. So it's important that scientists share their techniques and confirm each other's findings.
So how do scientific ideas become part of the community knowledge base? If everyone's always double-checking each other's work, how do hypotheses become theories? How do theories become scientific law? Well, first of all, keep in mind that theory in the world of science is not the same thing as a theory in everyday language. I might have a 'theory' that my friend Jackie is going to ask her classmate Jimmy on a date, but that's not the same as a scientific theory.
In science, a theory is a statement that is generally accepted as a summary for a hypothesis or a group of hypotheses. You can also call a theory an accepted hypothesis. When one hypothesis has been tested by many different scientists and most of them have come to the same basic conclusion, then we can start calling the hypothesis a theory. There isn't any 'grand master of science' who makes the final decree about a theory. It's more like a general consensus. And a theory can still be disproven if further research reveals enough evidence to refute it.
A law is different from a theory in that it is viewed as a universal fact. A scientific law is a general statement about a group of observations that has no exceptions to the rule. Most laws can be stated as mathematical equations, like Boyle's Law and Pascal's Law. Laws in biology are statements about how living things work - for example, Mendel's Laws. To explain the results of his experiments with peas, Gregor Mendel developed the Law of Segregation and the Law of Independent Assortment. No genetics experiment has ever disproven Mendel's laws, and so his statements are still viewed as laws today.
But even Gregor Mendel had to design a valid experiment in order to receive credit for the work that he did. The use of a control is one element that really makes an experiment scientific. When scientists are trying to test one factor, they have to make sure there aren't any variables going on that could mess up their results.
For example, let's say you wanted to see whether sunflowers or daisies grow faster from seed. You'd plant a sunflower seed in one pot, a daisy seed in another, and then put the pots in a window and water them every day. But what if you put the sunflower pot in a sunny window and the daisy pot in a shady window? Your findings would be skewed, right? You wouldn't know for sure whether sunflowers always grow faster than daisies or if it was just your sunflower growing faster because it got more sun.
In science, a control is a means of ensuring that only one factor is being tested at a time. In order to make yours a controlled experiment, you would need to place both flower pots in the same window. You'd also give each plant the same amount of water, make sure the seeds were planted in the same type of soil, and plant both seeds at the same time. The more you controlled the variable factors in your experiment, the more confident you'd be that the results would accurately address your experimental question.
Get FREE access for 5 days,
just create an account.
No obligation, cancel anytime.
While we're on the subject, let's check over this flower experiment to make sure it follows the other rules. Remember, our peers in the scientific community are scrutinizing our every move! So first of all, is our experiment designed so that it either supports or refutes our hypothesis? Well, we need a hypothesis. Let's say our hypothesis is 'Sunflowers will grow faster than daisies when grown in similar conditions.' Our experiment is designed to test and compare the two growth speeds, so we've got that rule covered.
Next, do we have a means of generating measurable, objective results? Our results are related to plant height over time, so we'll need to measure the plants' heights in millimeters and the time in days and hours. We'll probably make a chart that shows how much each plant grew over increments of time.
Finally, can other scientists repeat our experiment in the same way we designed it? Well, obviously anyone can grow plants in their window. But as scientists we need to specify every detail about our experiment. How much soil are we putting in the pots? What will the average temperature be? How many hours of light will each plant receive? Exactly how much water will we give to the plants? What kind of potting soil will we use? How often will we measure the heights of the plants? How long will we continue our experiment? I know this seems like a lot of questions, but these are the rigorous requirements that scientists face in their work. If they don't defend their experiments with detailed information, then their work isn't recognized in peer review.
As you can see, scientific experimentation is no walk in the park. In order to receive credit for their work, scientists have to design and implement experiments with great complexity and precision. In fact, some scientists publish hundreds of pages of work for every study they perform! It may seem like overkill, but the challenges of the scientific community help ensure that only valid, reliable science gets fully recognized and used.
It takes a lot of work to design a good experiment. To fit in with the scientific method, experiments have to be relevant to the questions, hypotheses, observations, analyses, and conclusions of the investigation. Experiments are designed so that they support or refute a hypothesis and give results in terms of measurable, objective data. An experiment must be repeatable by other scientists so that it holds up in peer review.
When a hypothesis becomes generally accepted by the science community, it becomes a theory. Scientific principles that are viewed as fact become laws. In designing their experiments, scientists must incorporate controls to ensure that only one variable is tested at a time. Experimental design and implementation is tough, but these challenges help to keep science moving forward.
"This just saved me about $2,000 and 1 year of my life." — Student
"I learned in 20 minutes what it took 3 months to learn in class." — Student