ABHISHEK ANTIBIOTICS PPT MICROBIOLOGY // USES OF ANTIOBIOTICS TYPES OF ANTIB...
Chapter 3 Experimental Errors Statistics
1. SLIDE | 1
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
BSK1153
ANALYTICAL CHEMISTRY
CHAPTER 3
EXPERIMENTAL ERRORS AND STATISTICS
DR WAN NORFAZILAH WAN ISMAIL
FACULTY OF INDUSTRIAL SCIENCES AND TECHNOLOGY
norfazilah@ump.edu.my
2. SLIDE | 2
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Measurement and Readings
Depends on what apparatus or instruments you used or read.
Example:
50 mL burette with 0.1 graduation, readings must be to the nearest 0.01 mL
Calibrated mm ruler, readings must be to the nearest 0.1 mm
Others:
Analytical balance??
Top loading balance??
pH meter??
3. SLIDE | 3
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Errors in Chemical Analysis
There are two types of error:
1. Systematic error - always too high or too low (improper shielding and
grounding of an instrument or error in the preparation of standards).
2. Random error - unpredictably high or low (pressure changes or
temperature changes).
Precision = ability to control random error.
Accuracy = ability to control systematic error.
4. SLIDE | 4
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
5. SLIDE | 5
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Basic Statistics in Analytical Chemistry
Mean
ത
x =
σi xi
N
Deviation di = |xi - ҧ
𝑥|
Range ω = xhighest - xlowest
Variance
variance =
σi di
2
N − 1
Standard deviation
s =
σi(xi − ത
x)2
N
Relative standard deviation Relative standard deviation =
s
ത
x
Standard error of mean (Sm) s
n
6. SLIDE | 6
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Exercise
For data 0.725, 0.756, 0.752, 0.751 and 0.760. Calculate:
i. Mean
ii. Median
iii. Standard deviation
iv. Variance
v. Relative standard deviation
vi. Range
0.749
0.752
0.0125
1.898 × 10−4
0.0167
0.035
7. SLIDE | 7
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Statistics to Data Evaluation
Statistics
Statistics
Q-test
Q-test
Rejecting
outliers
Rejecting
outliers
t-test
t-test
Compared
measured result
with a known value
Compared
measured result
with a known value
Compare replicate
measurement
Compare replicate
measurement
Compare individual
differences
Compare individual
differences
Qexp =
gap
range
*Reject if Qexp > Qtable
Different from known value
if tcalc > ttable
μ = ത
x ±
ts
N
8. SLIDE | 8
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Q-test (Rejection of data)
Values of Q for Rejection of Data.
Number of
observations
Q (90% confidence) Q (95% confidence) Q (99% confidence)
3 0.941 0.970 0.994
4 0.765 0.829 0.926
5 0.642 0.710 0.821
6 0.560 0.625 0.740
7 0.507 0.568 0.680
8 0.468 0.526 0.634
9 0.437 0.493 0.598
10 0.412 0.466 0.568
9. SLIDE | 9
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Exercise
The concentrations of metal, M in a water sample were determined in a
set of five measurements: 4.85, 6.18, 6.28, 6.49, and 6.69 ppm. Can we
reject observation 4.85 as an outlier at:
i. 95% confidence level?
ii. 99% confidence level?
Reject
Accept
11. SLIDE | 11
Dr. Wan Norfazilah Wan Ismail | FIST | norfazilah@ump.edu.my
Exercise
A soda ash sample is analyzed in the UMP laboratory by titration with
standard hydrochloric acid. The analysis is performed in triplicate with
the following results: 93.58%, 93.43% and 93.50% Na2CO3. Calculate
the range where the true value lies at 95% confidence level.
The mean = 93.50%
standard deviation, s = 0.075% Na2CO3
At 95% confident level and two degree of
freedom, t = 4.303
Confident limit = x ± ts / N
= 93.50 ± (4.303 x 0.075) / 3
= 93.50 ± 0.19 %
true value = 93.31 – 93.69%