# Error Propagation

The analysis of uncertainties (errors) in measurements and calculations is essential in the physics laboratory. For example, suppose you measure the length of a long rod by making three measurement x = xbest ± ∆x, y = ybest ± ∆y, and z = zbest ± ∆z. Each of these measurements has its own uncertainty ∆x, ∆y, and ∆z respectively. What is the uncertainty in the length of the rod L = x + y + z? When we add the measurements do the uncertainties ∆x, ∆y, ∆z cancel, add, or remain the same? Likewise , suppose we measure the dimensions b = bbest ± ∆b, h = hbest ± ∆h, and w = wbest ± ∆w of a block. Again, each of these measurements has its own uncertainty ∆b, ∆h, and ∆w respectively. What is the uncertainty in the volume of the block V = bhw? Do the uncertainties add, cancel, or remain the same when we calculate the volume? In order for us to determine what happens to the uncertainty (error) in the length of the rod or volume of the block we must analyze how the error (uncertainty) propagates when we do the calculation. In error analysis we refer to this as error propagation. There is an error propagation formula that is used for calculating uncertainties when adding or subtracting measurements with uncertainties and a different error propagation formula for calculating uncertainties when multiplying or dividing measurements with uncertainties. Let’s first look at the formula for adding or subtracting measurements with uncertainties. Adding or Subtracting Measurements with Uncertainties. Suppose you make two measurements, x = xbest ± ∆x y = ybest ± ∆y What is the uncertainty in the quantity q = x + y or q = x – y? To obtain the uncertainty we will find the lowest and highest probable value of q = x + y. Note that we would like to state q in the standard form of q = qbest ± ∆q where qbest = xbest + ybest. (highest probable value of q = x + y): (xbest+ ∆x) + (ybest + ∆y) = (xbest+ ybest) + (∆x +∆y) = qbest + ∆q (lowest probable value of q = x + y): (xbest- ∆x) + (ybest - ∆y) = (xbest+ ybest) - (∆x +∆y) = qbest – ∆q Thus, we that ∆q = ∆x + ∆y is the uncertainty in q = x + y. A similar result applies if we needed to obtain the uncertainty in the difference q = x – y. If we had added or subtracted more than two 1

87cm ∆q = (∆x) 2 + (∆y ) 2 = (0....measurements x.04cm qbest = xbest + ybest = 3. Again we would like to state q in the standard form of q = qbest ± ∆q where now qbest = xbest ybest. (highest probable value of q = xy): (xbest+ ∆x)(ybest + ∆y) = xbestybest + xbest ∆y +∆x ybest + ∆x ∆y = qbest + ∆q = xbestybest + (xbest ∆y + ∆x ybest) = qbest + ∆q (lowest probable value of q = xy): 2 . y.06cm q = 5.. . .52 cm ± 0.. ∆x = 0.. . the result would be ∆q = ∆x + ∆y + ....05 cm y = 2.87cm ± 0..35cm. + (∆z ) 2 Ex. The result will be the same if we consider q = x/y. if the uncertainties ∆x.05)2 + (0..... ∆y = 0.35cm = 5. ∆y......52cm + 2....04) 2 = 0.. x = xbest ± ∆x y = ybest ± ∆y What is the uncertainty in the quantity q = xy or q = x/y? To obtain the uncertainty we will find the highest and lowest probable value of q = xy.. the result is ∆q = (∆x) 2 + (∆y ) 2 + ..05cm ybest = 2. ∆z are random and independent.04 cm Calculate q = x + y We would like to state q in the standard form of q = qbest ± ∆q xbest = 3.35 cm ± 0.. x = 3.... . ∆z respectively ..06cm Multiplying or Dividing Measurements with Uncertainties Suppose you make two measurements. ∆y. z each with its own uncertainty ∆x....... + ∆z Now...52cm..

.52cm. + qbest y best x best z best Now. then the product ∆x ∆y ≈ 0..∆x)(ybest ... ∆z are random and independent. ∆y. .08cm y = 189.52cm   189... .53cm)=9.. a similar result applies if we needed to obtain the uncertainty in the division of q = x/y.08cm ∆y = 0. ybest = 189.... if the uncertainties ∆x..05cm   =    +  = 1...05cm qbest = xbestybest = (49.. .∆x ybest + ∆x ∆y = qbest – ∆q = xbestybest – (xbest ∆y + ∆x ybest) = qbest – ∆q Since the uncertainties ∆x and ∆y are assumed to be small.52cm ± 0.. ∆y. y..xbest ∆y . z each with its own uncertainty ∆x.∆y) = xbestybest ..08cm   0.. Dividing by xbestybest gives x ∆y y ∆x ∆q = best + best xbest y best xbest y best xbest y best ∆q ∆y ∆x = + qbest y best x best Again.. ..... ∆x = 0.53cm   2 2 3 .. the result would be ∆z ∆q ∆y ∆x = + + .63691E − 3   49. If we had multiplied or divided more than two measurements x..38553 x 103 cm2  ∆x ∆q =  x qbest  best   ∆y  +  y   best 2  0. x = 49.05cm Calculate q = xy We would like to state q in the standard form of q = qbest ± ∆q xbest = 49.... +   z   best 2     2 Ex.. Thus...53cm ± 0. we see that ∆q = xbest ∆y +∆x ybest in either case...(xbest. the result is  ∆y ∆q =  y qbest  best   ∆x  +  x   best 2   ∆z  + .52cm)(189....53cm. ∆z respectively..

is given by the expression ∆q ∆x = n qbest xbest Ex.1cm 3 ∆q ∆x = n = qbest xbest ∆q  0.08 cm. Let q = x3 where x = 5.∆q = (1.08cm  = (3)  3 190.1cm  5.93cm 3 ≈ 7cm 3 q = 190cm 3 ± 7cm 3 4 . We would like to state q in the standard form of q = qbest ± ∆q n=3 ∆x = 0.63691 x 10-3)qbest = (1.75cm  ∆q = 7.3632cm2 ≈ 20 cm2 q = 9390 cm2 ± 20 cm2 Uncertainty for a Quantity Raised to a Power If a measurement x has uncertainty ∆x. Calculate the uncertainty ∆q in the quantity q.08cm xbest = 5.75cm ± 0.38553 x 103 cm2) ∆q = 15.63691 x 10-3) (9. then the uncertainty in q = xn.75cm 3 qbest = xbest = 190.