Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener...
-
Upload
stuart-cross -
Category
Documents
-
view
213 -
download
0
Transcript of Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener...
![Page 1: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/1.jpg)
Quality Assurance & Quality Control
University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux
University of South Carolina Don Edwards
![Page 2: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/2.jpg)
Instruction Materials Primary References
Michener and Brunt (2000) Ecological Data: Design, Management and Processing. Blackwell Science.
Edwards (2000) Ch. 4 Brunt (2000) Ch. 2 Michener (2000) Ch. 7
Dux, J.P. 1986. Handbook of Quality Assurance for the Analytical Chemistry Laboratory. Van Nostrand Reinhold Company
Mullins, E. 1994. Introduction to Control Charts in the Analytical Laboratory: Tutorial Review. Analyst 119: 369 – 375.
![Page 3: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/3.jpg)
Outline:
Define QA/QC QC procedures
Designing data sheets Data entry using validation rules, filters,
lookup tables QA procedures
Graphics and Statistics Outlier detection
Samples Simple linear regression
![Page 4: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/4.jpg)
QA/QC “mechanisms [that] are designed to prevent
the introduction of errors into a data set, a process known as data contamination”
Brunt 2000
![Page 5: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/5.jpg)
Errors (2 types)
Commission: Incorrect or inaccurate data in a dataset
Can be easy to find Malfunctioning instrumentation
Sensor drift Low batteries Damage Animal mischief
Data entry errors
Omission Difficult or impossible to find Inadequate documentation of data values, sampling
methods, anomalies in field, human errors
![Page 6: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/6.jpg)
Quality Control “mechanisms that are applied in advance,
with a priori knowledge to ‘control’ data quality during the data acquisition process”
Brunt 2000
![Page 7: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/7.jpg)
Quality Assurance “mechanisms [that] can be applied after
the data have been collected, entered in a computer and analyzed to identify errors of omission and commission” graphics statistics
Brunt 2000
![Page 8: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/8.jpg)
QA/QC Activities Defining and enforcing standards for
formats, codes, measurement units and metadata.
Checking for unusual or unreasonable patterns in data.
Checking for comparability of values between data sets.
Brunt 2000
![Page 9: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/9.jpg)
Outline:
Define QA/QC QC procedures
Designing data sheets Data entry using validation rules, filters,
lookup tables QA procedures
Graphics and Statistics Outlier detection
Samples Simple linear regression
![Page 10: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/10.jpg)
Flowering Plant Phenology – Data Entry Form Design
Four sites, each with 3 transects Each species will have
phenological class recorded
![Page 11: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/11.jpg)
Plant Life Stage______________ _____________________________ _____________________________ _____________________________ _____________________________ _____________________________ _____________________________ _____________________________ _______________
What’s wrong with this data sheet?
Data Collection Form Development:
![Page 12: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/12.jpg)
Plant Life Stageardi P/G V B FL FR M S D NParpu P/G V B FL FR M S D NPatca P/G V B FL FR M S D NPbamu P/G V B FL FR M S D NPzigr P/G V B FL FR M S D NP
P/G V B FL FR M S D NPP/G V B FL FR M S D NP
PHENOLOGY DATA SHEETCollectors:_________________________________Date:___________________ Time:_________Location: black butte, deep well, five points, goat draw
Transect: 1 2 3
Notes: _________________________________________
P/G = perennating or germinating M = dispersingV = vegetating S = senescingB = budding D = deadFL = flowering NP = not presentFR = fruiting
![Page 13: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/13.jpg)
Outline:
Define QA/QC QC procedures
Designing data sheets Data entry using validation rules, filters,
lookup tables QA procedures
Graphics and Statistics Outlier detection
Samples Simple linear regression
![Page 14: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/14.jpg)
![Page 15: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/15.jpg)
Other methods for preventing data contamination
Double-keying of data by independent data entry technicians followed by computer verification for agreement
Randomly check 10% of data to calculate frequency of errors
Use text-to-speech program to read data back Filters for “illegal” data
Computer/database programs Legal range of values Sanity checks
Edwards 2000
![Page 16: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/16.jpg)
Table ofPossibleValues
and Ranges
Raw DataFile
Illegal DataFilter
Report ofProbable
Errors
Flow of Information in Filtering “Illegal” Data
Edwards 2000
![Page 17: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/17.jpg)
Illegal-data filter in SASData Checkum; Set all;
Message=repeat(“ ”,39);If Y1<0 or Y1>1 then do; message=“Y1 is not
on the interval [0,1]”; output; end;If Y3>Y4 then do; message=“Y3 is larger than
Y4”; output; end;…….(add as many checks as desired)If message NE repeat(“ ”, 39);Keep ID message;
Proc Print Data=Checkum;
![Page 18: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/18.jpg)
Outline:
Define QA/QC QC procedures
Designing data sheets Data entry using validation rules, filters,
lookup tables QA procedures
Graphics and Statistics Outlier detection
Samples Simple linear regression
![Page 19: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/19.jpg)
Identifying Sensor Errors: Comparison of data from three Met stations, Sevilleta LTER
![Page 20: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/20.jpg)
Identification of Sensor Errors: Comparison of data from three Met stations, Sevilleta LTER
![Page 21: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/21.jpg)
QA/QC in the Lab: Using Control Charts
![Page 22: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/22.jpg)
Laboratory quality control using statistical process control charts:
Determine whether analytical system is “in control” by examining Mean Variability (range)
![Page 23: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/23.jpg)
All control charts have three basic components:
a centerline, usually the mathematical average of all the samples plotted.
upper and lower statistical control limits that define the constraints of common cause variations.
performance data plotted over time.
![Page 24: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/24.jpg)
16
18
20
22
24
26
0 2 4 6 8 10 12 14 16 18 20
Time
Mean
LCL
UCL
X-Bar Control Chart
![Page 25: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/25.jpg)
Constructing an X-Bar control chart
Each point represents a check standard run with each group of 20 samples, for example
UCL = mean + 3*standard deviation
![Page 26: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/26.jpg)
Things to look for in a control chart:
The point of making control charts is to look at variation, seeking patterns or statistically unusual values. Look for: 1 data point falling outside the control limits 6 or more points in a row steadily increasing
or decreasing 8 or more points in a row on one side of the
centerline 14 or more points alternating up and down
![Page 27: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/27.jpg)
Linear trend
0
2
4
6
8
10
12
0 2 4 6 8 10 12
time
![Page 28: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/28.jpg)
Outline:
Define QA/QC QC procedures
Designing data sheets Data entry using validation rules, filters,
lookup tables QA procedures
Graphics and Statistics Outlier detection
Samples Simple linear regression
![Page 29: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/29.jpg)
Outliers An outlier is “an unusually extreme value
for a variable, given the statistical model in use”
The goal of QA is NOT to eliminate outliers! Rather, we wish to detect unusually extreme values.
Edwards 2000
![Page 30: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/30.jpg)
Outlier Detection
“the detection of outliers is an intermediate step in the elimination of [data] contamination” Attempt to determine if contamination is
responsible and, if so, flag the contaminated value.
If not, formally analyse with and without “outlier(s)” and see if results differ.
Edwards 2000
![Page 31: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/31.jpg)
Methods for Detecting Outliers Graphics
Scatter plots Box plots Histograms Normal probability plots
Formal statistical methods Grubbs’ test
Edwards 2000
![Page 32: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/32.jpg)
X-Y scatter plots of gopher tortoise morphometrics
Michener 2000
![Page 33: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/33.jpg)
Example of exploratory data analysis using SAS Insight®Michener 2000
![Page 34: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/34.jpg)
Box Plot Interpretation
Upper quartileMedian
Lower quartile
Upper adjacent value
Pts. > upper adjacent value
Lower adjacent valuePts. < lower adjacent value
Inter-quartile range
![Page 35: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/35.jpg)
Box Plot Interpretation
Inter-quartile range
IQR = Q(75%) – Q(25%)
Upper adjacent value = largest observation < (Q(75%) + (1.5 X IQR))
Lower adjacent value = smallest observation > (Q(25%) - (1.5 X IQR))
![Page 36: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/36.jpg)
Box Plots Depicting Statistical Distribution of Soil Temperature
![Page 37: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/37.jpg)
Statistical tests for outliers assume that the data are normally distributed….
CHECK THIS ASSUMPTION!
![Page 38: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/38.jpg)
Y
-3 -2 -1 0 1 2 3
0.0
0.1
0.2
0.3
0.4
a. The Normal Frequency Curve
Y
-3 -2 -1 0 1 2 3
0.0
0.2
0.4
0.6
0.8
1.0
b. Cumulative Normal Curve Areas
Normal density and Cumulative Distribution Functions
Edwards 2000
![Page 39: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/39.jpg)
Quantiles of Standard Normal
Y
-2 -1 0 1 2
78
910
1112
Normal Plot of 30 Observations from a Normal Distribution
Edwards 2000
![Page 40: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/40.jpg)
Quantiles of Standard Normal
Y
-2 -1 0 1 2
01
23
4
a. a right skewed distribution
Quantiles of Standard Normal
Y
-2 -1 0 1 2
0.0
1.0
2.0
3.0
b. a left-truncated Normal distribution
Quantiles of Standard Normal
Y
-2 -1 0 1 2
-50
5
c. a heavy-tailed distribution
Quantiles of Standard Normal
Y
-2 -1 0 1 22
46
810
12
14
16
d. a contaminated Normal distribution
Edwards 2000
Normal Plots from Non-normally Distributed Data
![Page 41: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/41.jpg)
Grubbs’ Text is a Statistical Test to determine if a point is an outlier.
![Page 42: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/42.jpg)
Grubbs’ test for outlier detection:
Tn = (Yn – Ybar)/S
where Yn is the possible outlier,
Ybar is the mean of the sample, and
S is the standard deviation of the sample
Contamination exists if Tn is greater than T.01[n]
![Page 43: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/43.jpg)
Example of Grubbs’ test for outliers: rainfall in acre-feet from seeded clouds (Simpson et al. 1975)
4.1 7.7 17.5 31.432.7 40.6 92.4 115.3118.3 119.0 129.6 198.6200.7 242.5 255.0 274.7274.7 302.8 334.1 430.0489.1 703.4 978.0 1656.01697.8 2745.6T26 = 3.539 > 3.029; Contaminated
Edwards 2000But Grubbs’ test is sensitive to non-normality……
![Page 44: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/44.jpg)
0 500 1500 2500
05
10
15
20
rainfall
a. rainfall
Quantiles of Standard Normal
rain
fall
-2 -1 0 1 2
0500
1500
2500
b. Normal plot
0.5 1.5 2.5 3.5
02
46
810
log10(rainfall)
c. log10(rainfall)
Quantiles of Standard Normal
log10(r
ain
fall)
-2 -1 0 1 2
0.5
1.5
2.5
3.5
d. Normal plot
Edwards 2000
Checking Assumptions on Rainfall Data
Contaminated
Normallydistributed
![Page 45: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/45.jpg)
Simple Linear Regression:check for model-based……
Outliers Influential (leverage) points
![Page 46: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/46.jpg)
Influential points in simple linear regression:
A “leverage” point is a point with an unusual regressor value that has more weight in determining regression coefficients than the other data values.
Edmonds 2000
![Page 47: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/47.jpg)
x
y
0 10 20 30 40 50
50
100
150
outlier andleverage pt
leverage pt
outlier
Edwards 2000
Influential Data Points in a Simple Linear Regression
![Page 48: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/48.jpg)
x
y
0 10 20 30 40 50
50
100
150
outlier andleverage pt
leverage pt
outlier
Edwards 2000
Influential Data Points in a Simple Linear Regression
![Page 49: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/49.jpg)
x
y
0 10 20 30 40 50
50
100
150
outlier andleverage pt
leverage pt
outlier
Edwards 2000
Influential Data Points in a Simple Linear Regression
![Page 50: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/50.jpg)
x
y
0 10 20 30 40 50
50
100
150
outlier andleverage pt
leverage pt
outlier
Edwards 2000
Influential Data Points in a Simple Linear Regression
![Page 51: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/51.jpg)
body weight (kg)
bra
in w
eig
ht (g
)
0 2000 4000 6000
01000
2000
3000
4000
5000
African Elephant
Asian Elephant
Man
Edwards 2000
Brain weight vs. body weight, 63 species of terrestrial mammals
Leverage pts.
“Outliers”
![Page 52: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/52.jpg)
log10 body weight (kg)
log10 b
rain
weig
ht (g
)
-2 -1 0 1 2 3 4
-10
12
3
Elephants
Man
Chinchilla
mispunched point
Edwards 2000
Logged brain weight vs. logged body weight
“Outliers”
![Page 53: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/53.jpg)
Outliers in simple linear regression
0
50
100
150
200
250
0 50 100 150 200 250 300 350 400 450
Beta-Glucosidase
Ph
os
ph
ata
se
Observation 62
![Page 54: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/54.jpg)
Outliers: identify using studentized residuals Contamination may exist if
|ri| > t /2, n-3
= 0.01
![Page 55: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/55.jpg)
Simple linear regression:Outlier identification
n = 86
t/2,83 = 1.98
![Page 56: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/56.jpg)
Simple linear regression:detecting leverage points
hi = (1/n) + (xi – x)2/(n-1)Sx2
A point is a leverage point if hi > 4/n, where n is the number of points used to fit the regression
![Page 57: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/57.jpg)
Regression with leverage point:Soil nitrate vs. soil moisture
![Page 58: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/58.jpg)
Regression without leverage point
Observation 46
![Page 59: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/59.jpg)
Output from SAS:Leverage points
n = 336
hi cutoff value: 4/336=0.012
![Page 60: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/60.jpg)
Process from raw data to verified data to validated data (reviewed by a qualified scientist) is called data maturity, and implies increasing
confidence in the quality of the data through time
![Page 61: Quality Assurance & Quality Control University of New Mexico Kristin Vanderbilt William Michener James Brunt Troy Maddux University of South Carolina Don.](https://reader036.fdocuments.in/reader036/viewer/2022070410/56649f285503460f94c406e7/html5/thumbnails/61.jpg)
Data Quality (Confidenc
e in the Data)
Data Maturity
Raw Data
Verified Data
Validated Data