The variable that the scientist changes during the experiment is the independent variable. Think of the experiment as a "cause and effect" exercise. The independent variable is the "cause" factor. For example, to test the hypothesis that a seed needs light to germinate, the independent variable would be whether the seed was left exposed to light or covered with soil. If you hypothesize that Cub Scout pinewood derby cars are faster when the weight is near the rear, the placement of the weight is the independent variable.
The dependent variable is what is measured or observed. It is the "effect" in the cause-and-effect relationship. In the seed experiment, seed germination would be the dependent variable. For the pinewood derby car, the time it takes the car to go down the ramp is the measurable, dependent variable. Each time you change the independent variable, such as by placing the car weight different distances from the rear, take a measurement.
In order for the test to be fair, other factors that could affect the outcome of the experiment should be kept the same, or controlled. For the seed experiment, seeds should come from the same species, source and storage conditions. The temperature, watering, planting mix and exposure period should be the same. In the pinewood derby experiment, controlled variables might include the car design; the height, length and smoothness of the ramp; initial placement of the car on the ramp; and the kind of metal in the weight. Sometimes controlled variables are called "constant variables." If these variables can't be kept constant, record their values and assess their influence on the experiment.
The National Center for Education Statistics suggests the following test to identify independent and dependent variables correctly. Put the appropriate words for your experiment in this sentence to see if they make sense: (Independent variable) causes a change in (dependent variable) and it isn't possible that (dependent variable) could cause a change in (independent variable). For the seed experiment, the test sentence would read: Light causes a change in seed germination and it isn't possible that seed germination could cause a change in light.
Independent, dependent and controlled variables need to be measurable for accurate recording. The nature of the variable affects observation and measurement. A discrete variable is measured with simple numbers that record a physical quantity or a physical value such as number of seeds germinated. A continuous variable is a numerical value that can have any number, such as distance or temperature. A categorical variable is measured by a label rather than a number, such as color. Ordered variables are characterized by size order rather than number, such as large, medium or small.