Since experimental errors cannot be completely eliminated, every effort should be made to minimize them. there are two ways to minimize errors in any experiment. They are by selection of approprate measuring devices and by making several measurements for each variable and taking average. The first minimizes systematic errors and the second reduces the random errors. The minimization of systematic and random errors is illustrated by the following examples:

A. SIMPLE PENDULUM

The formula g=4 Pi^2*L/T^2 indicates that the error in the determination of g will come from errors in the measurements of L and T because the remaining factors (4 and Pi^2) are constants. We also saw that the error in the measurement of T was doubled because of square of T in the formula. Therefore, it is twice more important to reduce the fractional error in the measurement of T than to reduce the fractional error in the measurement of L.

B. YOUNG'S MODULUS

In measuring various physical quantities necessary in any experiment, one should carefully look at relative sizes of the quantities. For example in the Young's modulus experiment [Formula: Y=MgL/Pi^2rl],M and L are large while r and l are very small. This indicates that extra care must be used in measuring r and l and devices used to measure r and l should have the lowest possible leastcount; otherwise the relative fractional errors contributed by these factors will be much higher than relative fractional errors contributed by the other two factors (M and L). Also, since the radius is squared in the formula, diameter should be measured with the best available micrometer. Finally, if one variable (factor) in an experiment can only be measured with an accuracy of 1% then there is no point in trying to measure the other factors to less than 0.1% accuracy because the overall accuracy of the experiment will be determined by the factor with the largest percentage error.