Scribd
Upload
10 views2 pages

205 W1 Meaurement Errors 2

Uploaded by

khujastakhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views2 pages

205 W1 Meaurement Errors 2

Uploaded by

khujastakhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Christian7e c03.tex V2 - 08/28/2013 1:37 P.M.

Page 71

3.6 WAYS OF EXPRESSING ACCURACY 71

3.5 Rounding Off


If the digit following the last significant figure is greater than 5, the number is rounded Always round to the even number,
up to the next higher digit. If it is less than 5, the number is rounded to the present if the last digit is a 5.
value of the last significant figure:
9.47 = 9.5
9.43 = 9.4
If the last digit is a 5, the number is rounded off to the nearest even digit:
8.65 = 8.6
8.75 = 8.8
8.55 = 8.6
This is based on the statistical prediction that there is an equal chance that the last
significant figure before the 5 will be even or odd. That is, in a suitably large sampling,
there will be an equal number of even and odd digits preceding a 5. All nonsignificant
digits should be rounded off all at once. The even-number rule applies only when
the digit dropped is exactly 5 (not . . . 51). (For example, if we want four significant
figures, then 45.365 rounds to 45.36, whereas 45.3651 rounds to 45.37.)

3.6 Ways of Expressing Accuracy


There are various ways and units in which the accuracy of a measurement can be
expressed. In each case, it is assumed that a “true” value is available for comparison.

ABSOLUTE ERRORS
The difference between the true value and the measured value, with regard to the
sign, is the absolute error, and it is reported in the same units as the measurement.
If an analyst reports 2.62 g of copper as 2.51 g, the absolute error is −0.11 g. If the
measured value is the average of several measurements, the error is called the mean
error. The mean error can also be calculated by taking the average difference, with
regard to sign, of the individual test results from the true value.

RELATIVE ERROR
The absolute or mean error expressed as a percentage of the true value is the relative
error. The above analysis has a relative error of (−0.11/2.62) × 100% = −4.2%. The
relative accuracy is the measured value or mean expressed as a percentage of the true
value. The above analysis has a relative accuracy of (2.51/2.62) × 100% = 95.8%.
We should emphasize that, more often than not, neither number is known to be “true,”
and the relative error or accuracy is based on the mean of two sets of measurements;
that the true value is “true” is an assumption, unless that value is a certified value in a
reference standard.
Relative error can be expressed in units other than percentages. In very accurate
work, we are usually dealing with relative errors of much less than 1%, and it is
convenient to use a smaller unit. A 1% error is equivalent to 1 part in 100. It is also
equivalent to 10 parts in 1000. This latter unit is commonly used for expressing small
uncertainties. That is, the uncertainty is expressed in parts per thousand. The number
Christian7e c03.tex V2 - 08/28/2013 1:37 P.M. Page 72

72 CHAPTER 3 STATISTICS AND DATA HANDLING IN ANALYTICAL CHEMISTRY

0.11 expressed as parts per thousand of the number 2.62 would be 0.11 parts per 2.62,
or 42 parts per thousand. Parts per thousand is often used in expressing precision of
measurement. For even smaller relative amounts, parts per million (1 ppm = 1 part in
1,000,000) and parts per billion (1 ppb = 1 part in 1,000,000,000) are commonly used.

Example 3.6
The results of an analysis are 36.97 g, compared with the accepted value of 37.06 g.
What is the relative error in parts per thousand?

Which is a bigger relative error: Solution


395 ppb or 0.412 ppm?
Solution: Multiply ppm by 1000 to Absolute error = 36.97 g − 37.06 g = −0.09 g
get the value in ppb. Thus, the
−0.09
latter value is 412 ppb, which is Relative error = × 1000‰ = −2.4 parts per thousand
obviously greater than 395 ppb 37.06
and constitutes the bigger error. ‰ indicates parts per thousand, just as % indicates parts per hundred. 

3.7 Standard Deviation—The Most Important Statistic


“If reproducibility be a problem, Each set of analytical results should be accompanied by an indication of the precision
conduct the test only of the analysis. Various ways of indicating precision are acceptable.
once.”—Anonymous The standard deviation σ of an infinite set of experimental data is theoretically
given by

(xi − μ)2
σ = (3.1)
N
where xi represents the individual measurements and μ represents the mean of an
average, x = (xi /N) infinite number of measurements (which should represent the “true” value). This
equation holds strictly only as N → ∞, where N is the number of measurements.
In practice, we must calculate the individual deviations from the mean of a limited
that x → μ as N → ∞, although
number of measurements, x, in which it is anticipated
we have no assurance this will be so; x is given by (xi /N).
For a set of N measurements, there are N (independently variable) deviations
from some reference number. But if the reference number chosen is the estimated
mean, x, the sum of the individual deviations (retaining signs) must necessarily add up
to zero, and so values of N − 1 deviations are adequate to define the Nth value. That
is, there are only N − 1 independent deviations from the mean; when N − 1 values
have been selected, the last is predetermined. We have, in effect, used one degree of
freedom of the data in calculating the mean, leaving N − 1 degrees of freedom for
calculating the precision.
See Section 3.15 and Equation As a result, the estimated standard deviation s of a finite set of experimental
3.17 for another way of estimating data (generally N < 30) more nearly approximates σ if N − 1, the number of degrees
s for four or less numbers. of freedom, is substituted for N (N − 1 adjusts for the difference between x and μ).

(xi − x)2
s= (3.2)
N−1

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy