Education.com
Try
Brainzy
Try
Plus

Planning and Conducting Your Science Fair Experiment (page 4)

By John Wiley & Sons, Inc.
Updated on Jul 8, 2010

Step Four: Create an Experimental Procedural Plan

An experimental procedural plan is a uniform, systematic approach to testing your hypothesis. When you begin this phase you should make a step-by-step list of what you will do to test your hypothesis. To start, first correlate (i.e., bring one thing into a reciprocal relationship with another) what you want to prove. You begin by selecting one thing to change in each experiment. Things that are changed are called variables. You want to be able to correlate two or more variables— the dependent variable and the independent variable. The dependent variable is the one that is being measured; the independent variable is the one that is controlled or manipulated by the experiment. For example, you may want to see whether the health and growth of a tomato plant (the dependent variable) is influenced by the amount of light the plant is exposed to (the independent variable). The correlation here is between the health of a plant and light exposure. Several other independent variables may be used instead, such as water, oxygen, carbon dioxide, nitrogen levels, and so on. However, for the sake of clarity we will use only light as a variable for this example. You should then state how you will change your independent variable and how you will measure the amount of change in the dependent variable.

Establish a Control Group

Next, an experimental group and a control group must be established. The control group provides you with a basis for comparing the experimental group. For example, you may have an experimental group of tomato plants, which is placed in a sunny window for two weeks and watered periodically. At the end of the period, the plants have grown three inches and are very green. At this point, you may conclude that sunlight does indeed increase plant growth. But before you draw this conclusion, you should determine whether the tomato plants would have grown and become green without any sunlight at all. This is where a control group of plants is needed. A control group is used for purposes of comparison with the experimental group so that you can see what occurred by changing your variables.

The control group of plants in our example would be those plants that are given the same treatment as the experimental ones, with the exception that they would not be exposed to sunlight. If the outcome of the experiment were a significant difference between the two groups, then you probably would be justified in concluding that tomato plant growth is influenced by the amount of sunlight the plant receives.

The procedural plan in this example is very simple, but it gives you an idea of the process of an experiment. In essence, the procedural plan advances from one stage to another in an organized fashion. Remember, however, that most experiments are not as simple as the one described here. Often obstacles arise and other interesting characteristics of the subject are revealed in the process. You may even discover existing differences in several trials with only one variable. In fact, this is a frequent occurrence, and it is an important reason why you must keep accurate data records.

Step Five: Conduct Your Experiment

Once you have established your procedural plan for your experiment and have received approval, it is time to collect the materials you will need. You may also need to obtain approval or permission to work in a laboratory or other professional environment. The important thing to keep in mind as you put your procedural plan into action is to collect accurate data results from repeated trials with the same variables and record all of your data for later analysis (see Chapter 5 for this important next step). The benefit to taking this approach is that it increases the accuracy of your results and conclusions. How many times do you need to repeat your experimental procedural plan? This really depends on many different factors, not the least of which is your subject matter.

If you have a mentor who is a professional scientist or engineer, you may already have the supplies and equipment you need through the mentor’s affiliation with a university or company research laboratory.

Avoiding a Failed Experiment

There are several reasons why an experiment may fail to validate a hypothesis, prove a point, or simply do what it was intended to do, for example, mistakes in the way the experiment was carried out (procedural errors), a poor or incomplete final analysis, and an erroneous hypothesis.

Procedural Errors

To avoid procedural problems, you must be consistent and meticulous with your subject variables and controls over repeated trials. For example, in the experiment involving sunlight and tomato plants, if you gave the experimental group of tomato plants more water than the control group or planted them in a soil that contained more nitrogen, you would get artificial results. This means that you are failing to control or hold your variable constant. How can you determine whether it was the sunlight alone or in combination with other factors that made the experimental tomatoes flourish? The same problem with inconsistent maintenance of controls might apply if you were studying the behavior of your friends at a party for a psychological experiment. What would happen if you made your study obvious by taking notes or pictures? Your friends probably would be influenced by your behavior and would not act in their usual manner. In this case, as the old saying goes, “you cannot measure an experiment without affecting the result.” These examples involve manipulated experiments that would yield useless data. Of course, other procedural problems may arise during an experiment, especially if poorly calibrated measuring instruments are used.

Poor Final Analysis

Even after a carefully controlled experiment is completed, errors can still occur, possibly resulting from an incorrect analysis of results. For example, if you concluded that a certain salve cures acne, on the basis of tests that were conducted on female adolescents but not male ones, your final analysis would be inconclusive. While the salve may have worked on the females you tested, it may not work on females in different age groups or on males of all age groups. Other problems with the final analysis may arise from mathematical errors or from data that are irrelevant to the topic.

Erroneous Hypothesis

When an experiment is completed, the results are sometimes quite different from those that were predicted. If this occurs, do not manipulate the results to fit the initial hypothesis. The hypothesis may have been incorrect or vague to begin with, and the experimental results were accurate. If such problems occur in your project, you can salvage your work by finding out why the results were different than expected or by explaining a new or unexpected observation or solution. This will show the judges that you understand the primary aspects that concern your project topic, including the control and handling of variables in experimentation, repeated trials, and approach to reaching conclusions. This actually happens to be a judging criterion that many students overlook. So, if your experimental results are different from what you expected after several trials, take advantage of this situation by thoroughly analyzing and knowing why you received the results you did.

Keep in mind that many scientific investigations do not support their specific goals. However, this does not weaken the validity or value of these investigations. In fact, many experiments require repeated testing and exploration to understand a particular phenomenon. Sometimes, unexpected experimental results lead to surprising discoveries and more interesting science projects!

Summary

1. The experiment is an essential part of your science project. It should test, survey, compare, and ultimately aim to solve or answer the problem or question presented.
2. You must focus your topic on an experimental approach that will clearly test your hypothesis and will uphold the scientific method.
3. After you decide on an experimental approach, you must develop a way of testing your subject. This involves defining your objective, obtaining Scientific Review Committee approval, organizing your experiment, and creating an experimental procedural plan.
4. An experimental group containing variables and a control group must be established as part of the experimental procedural plan. Several trials should be made with the same variables to ensure consistent data results from which conclusions can be made.
5. Three common ways in which an experiment can fail are procedural errors, poor final analysis, and an erroneous hypothesis.