how much a variable a relationship can be discovered by experimental on changes to the relationship as the variable is changed.In an experiment, there will be: one variable-this is the feature you are measuring one or more variable-these are the variable which you change one or more variable-these are not being tested and so they stay the same.