Scribd
Upload
3 views

Chapter4

Chapter 4 discusses the application of statistics in analytical chemistry, focusing on concepts such as mean, standard deviation, and Gaussian distribution. It provides examples of calculating averages and confidence intervals, as well as methods for comparing means and standard deviations using Student's t-test and the F-test. Additionally, it addresses the identification of outliers using Grubbs' test and the method of least squares for calibration curves.

Uploaded by

EK
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
3 views

Chapter4

Chapter 4 discusses the application of statistics in analytical chemistry, focusing on concepts such as mean, standard deviation, and Gaussian distribution. It provides examples of calculating averages and confidence intervals, as well as methods for comparing means and standard deviations using Student's t-test and the F-test. Additionally, it addresses the identification of outliers using Grubbs' test and the method of least squares for calibration curves.

Uploaded by

EK
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/ 20

Chapter 4

Statistics in Analytical Chemistry


We commonly express experimental results in the form:
mean ± standard deviation = ± s
Example
A factory produces light bulbs and four light bulbs were used to test the life time
and the lifetime values are listed below: 821, 783, 834, and 855 hours.
Find the average and standard deviation
821 + 783 + 834 + 855
x= = 823
4

S= (821 − 823) 2 + (783 − 823) 2 + (834 − 823) 2 + (855 − 823) 2 =30


4 −1

A Gaussian curve in which μ = 0 and σ = 1.


A Gaussian curve whose area is unity is called
a normal error curve. In this case, the abscissa,
x, is equal to z, defined as z = (x − x)/s.
|Z|value, Area from 0 to |Z| |Z|value, Area from 0 to |Z| |Z|value, Area from 0 to |Z|

0.0 0.0000 1.4 0.4192 2.8 0.4974

0.1 0.0398 1.5 0.4332 2.9 0.4981

0.2 0.0793 1.6 0.4452 3.0 0.4987

0.3 0.1179 1.7 0.4554 3.1 0.4990


0.4 0.1554 1.8 0.4641 3.2 0.4993
0.5 0.1915 1.9 0.4713 3.3 0.49952
0.6 0.2257 2.0 0.4772 3.4 0.49966
0.7 0.2580 2.1 0.4821 3.5 0.49977
0.8 0.2881 2.2 0.4861 3.6 0.49984
0.9 0.3159 2.3 0.4893 3.7 0.49990
1.0 0.3413 2.4 0.4918 3.8 0.49928
1.1 0.3643 2.5 0.4938 3.9 0.49995
1.2 0.3849 2.6 0.4953 4.0 0.49997
1.3 0.4032 2.7 0.4965 Infinity 0.5
Example
The lifetime of light bulb obeys Gaussian’s distribution. The average lifetime is
823 h and standard deviation is 85 h. The manufacturer offers to replace free of
charge any bulb that burns out in less than 600 h. If the factory plans to sell a million
bulbs, how many extra bulbs should be kept as replacements?

x = 823 S = 85 x = 600

z = 600-823 = -2.60
85

Area is 0.4953

Probability from –infinity to -2.60 is


1 - 0.5 - 0.4953 = 0.0047 = 0.47 %

1000000 X 0.47 % = 4700


6
4-2. Student t
The carbohydrate content of a glycoprotein (a protein with sugars
attached to it) is determined to be 12.6, 11.9, 13.0, 12.7, and 12.5 g of
carbohydrate per 100 g of protein in replicate analyses. Find the 90%
confidence intervals for the carbohydrate content.

Uncertainty

The most common estimates of uncertainty are the standard


deviation and the confidence interval.
Frequently, we use the standard deviation as the estimated
uncertainty. 7
Comparison of Means with
Student's t
You purchased a Standard Reference Material coal sample
certified by the National Institute of Standards and Technology to
contain 3.19 wt% sulfur. You are testing a new analytical method
to see whether it can reproduce the known value. The measured
values are 3.29, 3.22, 3.30, and 3.23 wt% sulfur, giving a mean of
and a standard deviation of s = 0.041. Does your answer agree with
the known answer?

8
9
Confidence interval

10
A reliable assay shows that the ATP (adenosine triphosphate)
content of a certain cell type is 111 μmol/100 mL. You developed
a new assay, which gave the following values for replicate analyses:
117, 119, 111, 115, 120 μmol/100 mL (average = 116.4). Can you
be 95% confident that your result differs from the “known” value?

11
Comparison of Means with
Student's t
Case 1 We measure a quantity of two samples by same method
that give two different answers, each with its own standard
deviation. Do the two results agree with each other “within
experimental error”?

Case 2 We measure a quantity of one samples by same method


but by two different people that give two different answers, each
with its own standard deviation. Do the two results agree with each
other “within experimental error”?

12
If tcalculated > ttable (95% at n1 + n2 − 2 degrees of freedom) , the
difference is significant.

13
Lord Rayleigh: light scattering,
blackbody radiation, and elastic
waves in solids.
Nobel Prize, 1904. Discovery of
the inert gas argon.
A small discrepancy between two
sets of measurements of the density
of nitrogen gas.

14
The average mass of a gas from air is 2.310 11 g, with a
standard deviation of s1 = 0.000 143 (for n1 = 7 measurements).
The mass of gas from chemical sources is 2.299 47 g, with s2 =
0.001 38 (n2 = 8 measurements), are the two gas different?

15
Comparison of Standard Deviations with the
F Test

We always put the larger standard deviation in the


numerator so that F ≥ 1. If Fcalculated > Ftable, then the
difference is significant. 16
Comparison of Standard Deviations with the
F Test

17
Grubbs Test for Bad Data
• Consider the 12 results: 10.2, 10.8, 11.6, 9.9, 9.4, 7.8,
10.0, 9.2, 11.3, 9.5, 10.6, and 11.6. Is 7.8 a “bad point”?

18
™need at least 4 measurements
Method of least squares

19
Calibration Curve
A good example of method of least squares

Sy =
∑ i)
( d 2

n−2

n
Sm = S y
D

Sb = S y
∑ i)
( x 2

You might also like