You can make a fair test by making all the variables the same , exept one , which is the fact that the objects are different.The variables range from anywhere in between , one object has more force than another ,or they are dropped at different heights , or even on different surfaces.
First you make a game plan or procedure all thought out.If you do find a way to only have the one variable allowed then you have to record the data ,which is the information you need to collect.If you are taking the height of a bouncing ball for example, you have to catch the ball at its highest point off the bounce and not move it at all, after you catch it.After you have collected your data you do the mean.That is your final conclusion.
We did a closely related project in class today.It was totally BOSS SAUCE!!!!!!!(Which means awesome Mr.S)It was so fun trying to figure out how to get the data correctly plotted and if our procedure was right.