CH 5-7 Statistical Data in Analyasis
CH 5-7 Statistical Data in Analyasis
Chapter 5-7
Errors, Random Errors, and Statistical
Data in Chemical Analyses
Error (偏差,誤差):
1) The difference between a measured value and the “true” or
“known” value.
2) The estimated uncertainty in a measurement or experiment.
3-2
1
20/8/31
Figure 5-1 Results from six replicate determinations for iron in aqueous
samples of a standard solution containing 20.00 ppm of iron(III).
3-3
3-4
2
20/8/31
3-5
3-6
3
20/8/31
(5-1)
The symbol Σxi means to add all the values xi for the
replicates.
3-7
3-8
4
20/8/31
3-9
¡ Calculate the mean and the median for the data shown in Figure 5-1.
3-10
5
20/8/31
Figure 5-1 Results from six replicate determinations for iron in aqueous
samples of a standard solution containing 20.00 ppm of iron(III).
3-11
3-12
6
20/8/31
(5-2)
3-13
3-14
7
20/8/31
¡ Absolute Error
§ The absolute error E in the measurement of a quantity xi is
given by the equation
(5-3)
3-16
8
20/8/31
¡ Relative error
§ Relative error is also expressed in parts per thousand (ppt).
e.g.
3-17
3-18
9
20/8/31
NH
S O
NH3Cl
OH
N
benzyl isothiourea hydrochloride nicotinic acid
absolute average
deviation = 0.22
Figure 5-1 Results from six replicate determinations for iron in aqueous
samples of a standard solution containing 20.00 ppm of iron(III).
3-20
10
20/8/31
¡ Refer to Figure 5-3: Notice that the scatter in the data, and
thus the random error, for analysts 1 and 3 is significantly
less than that for analysts 2 and 4.
3-21
3-22
11
20/8/31
3-23
3-24
12
20/8/31
3-25
3-26
13
20/8/31
¡ Instrument errors
§ All measuring devices are sources of systematic errors.
For example, pipets, burets, and volumetric flasks may
hold or deliver volumes slightly different from those
indicated by their graduations.
§ These differences arise from using glassware at a
temperature that differs significantly from the calibration
temperature, from distortions in container walls due to
heating while drying, from errors in the original calibration,
or from contaminants or scratches on the inner surfaces of
the containers. Calibration eliminates most systematic
errors of this type.
3-27
¡ Instrument Errors
§ Electronic instruments are subject to instrumental
systematic errors. These uncertainties have many
sources. For example, errors emerge as the voltage of a
battery-operated power supply decreases with use.
Instrument errors are detectable and correctable.
3-28
14
20/8/31
¡ Method Errors
§ The non-ideal chemical or physical behavior of the
reagents and reactions upon which an analysis is
based often introduce systematic method errors.
3-29
¡ Method Errors
§ In Figure 5-3, the results by analysts 3 and 4 show a
negative bias that can be traced to the chemical nature of
the sample, nicotinic acid.
§ The compounds containing a pyridine ring are incompletely
decomposed by the sulfuric acid; hence, the negative
errors in Figure 5-3 are likely systematic errors from
incomplete decomposition of the samples.
§ Errors inherent in a method are often difficult to detect and
are thus the most serious of the three types of systematic
error.
3-30
15
20/8/31
NH
S O
NH3Cl
OH
N
benzyl isothiourea hydrochloride nicotinic acid
Kjeldahl method
The Kjeldahl method or Kjeldahl digestion is a method for the
quantitative determination of nitrogen in chemical substances.
Degradation:
Sample + H2SO4 → (NH4)2SO4(aq) + CO2(g) + SO2(g) + H2O(g)
Liberation of ammonia:
(NH4)2SO4(aq) + 2NaOH → Na2SO4(aq) + 2H2O(l) + 2NH3(g)
Capture of ammonia:
B(OH)3 + H2O + NH3 → NH4+ + B(OH)4–
Back-titration:
B(OH)3 + H2O + Na2CO3 → NaHCO3(aq) + NaB(OH)4(aq) + CO2(g) + H2O
NH
S O
NH3Cl
OH
N
benzyl isothiourea hydrochloride nicotinic acid
1-32
16
20/8/31
¡ Personal Errors
Measurements requiring personal judgments.
§ Judgments of this type are often subject to systematic,
unidirectional errors.
§ An analyst who is insensitive to color changes tends to use
excess reagent in a volumetric analysis. Physical
disabilities are often sources of personal determinate
errors.
§ A universal source of personal error is prejudice or bias.
3-33
¡ Personal Errors
Number bias is another source of personal error that varies
considerably from person to person.
3-34
17
20/8/31
3-35
3-36
18
20/8/31
3-37
3-38
19
20/8/31
3-39
3-40
20
20/8/31
¡ Independent Analysis
§ If standard samples are not available, a second
independent and reliable analytical method can be used in
parallel with the method being evaluated.
3-41
3-42
21
20/8/31
¡ Define bias.
1-43
3-44
22
20/8/31
1-45
¡ Table 6-1 shows all the possible ways the four errors
can combine to give the indicated deviations from the
mean value.
3-46
23
20/8/31
3-47
3-48
24
20/8/31
(c)
3-50
25
20/8/31
3-51
26% of the data reside in the cell containing the mean and median
value of 9.982 mL and that more than half the results are within ±
0.004 mL of this mean.
1-52
26
20/8/31
3-53
3-54
27
20/8/31
3-55
3-56
28
20/8/31
3-57
3-58
29
20/8/31
3-59
Figure 6-4 Normal error curves. The standard deviation for curve B
is twice that for curve A; that is, σB = 2σA. (a) The abscissa is the
deviation from the mean in the units of measurement.
3-60
30
20/8/31
3-61
31
20/8/31
(6-1)
3-63
3-64
32
20/8/31
3-66
33
20/8/31
3-67
(6-4)
3-68
34
20/8/31
(6-4)
3-69
3-70
35
20/8/31
Substituting into Equation 6-5 leads to
3-71
(6-5)
3-72
36
20/8/31
Gaussian Curve
1-73
3-74
37
20/8/31
3-75
3-76
38
20/8/31
3-77
3-78
39
20/8/31
3-79
3-80
40
20/8/31
¡ Other than sample standard deviation, three other terms are often
employ in reporting the precision.
1. The variance (s2) is
(6-8)
3-81
s s
RSD = s r = RSDinppt = × 1000 ppt
x x
s
CV = RSDinperce nt = × 100% (6-9)
x
3-82
41
20/8/31
3-83
• For the set of data in Example 6-3, calculate (a) the variance,
(b) the relative standard deviation in parts per thousand, (c)
the coefficient of variation, and (d) the spread.
3-84
42
20/8/31
3-85
sy = ±0.06 à 2.63±0.06
3-86
43
20/8/31
3-87
3-88
44
20/8/31
3-89
¡ 信度(reliability)是指測量結果的一致性、穩定性及可靠性。
3-90
45
20/8/31
3-91
3-92
46
20/8/31
Ex.
3-93
3-94
47
20/8/31
3-95
3-96
48
20/8/31
3-97
3-98
49
20/8/31
3-99
迴歸分析(Regression Analysis)是一種統計學上分析數據的方
法,目的在於了解兩個或多個變數間是否相關、相關方向與
強度,並建立數學模型以便觀察特定變數來預測研究者感興
趣的變數。
3-100
50
20/8/31
1-101
3-102
51
20/8/31
The second assumption: any deviation of the individual points from the
straight line arises from error in the measurement. That is: no error in x
values of the points (concentration).
N 2
1-103
(8-10)
(8-11)
(8-12)
3-104
52
20/8/31
(8-14)
¡ 2. The intercept b:
(3-24)
(8-15)
3-105
(8-17)
53
20/8/31
3-107
54
20/8/31
1-110
55