"2 Variances" test for Paired Data in Minitab - How to proceed

B

Berg.Jlle

#1
Hello.

I could not find any post about my question of using of "2 Variances" test for paired data. Sorry whether a similar thread already exists... :eek:
So, can I use "Stat>Basic Statistics>2 Variances" to assess paired data ?

:thanx:
 
Elsmar Forum Sponsor

Miner

Forum Moderator
Leader
Admin
#2
What is the question to be answered?
  1. Is there a difference between the means? Paired t-Test
  2. Is there a relationship between the pairs? Correllation and Regression
The Test for Equal Variances is not really appropriate for paired data. If you are testing for the difference in means, you should be concerned that the delta between the paired measurements is normally distributed.
 
B

Berg.Jlle

#3
What is the question to be answered?
  1. Is there a difference between the means? Paired t-Test
  2. Is there a relationship between the pairs? Correllation and Regression
The Test for Equal Variances is not really appropriate for paired data. If you are testing for the difference in means, you should be concerned that the delta between the paired measurements is normally distributed.
Miner,

My purpose is to answer the following hypotheses for a set of paired data.

H0: σA = σB or σA - σB = 0
H1: σA ≠ σB or σA - σB ≠ 0

So my question is if I can use "Stat>Basic Statistics>2 Variances" to assess differences of standard deviation or variances for paired data ?


Data:

* The same 35 mobile phones were measured in both equipment A and B.
HTML:
#	  A		  B
1	9,70		9,29
2	8,72		7,91
3	10,24		9,99
4	11,28		11,10
5	11,20		11,10
6	11,73		11,60
7	7,82		7,08
8	9,77		9,41
9	11,10		11,10
10	8,91		8,92
11	9,31		8,62
12	8,31		7,95
13	8,15		7,93
14	9,02		9,06
15	9,23		9,60
16	7,88		7,65
17	9,43		8,84
18	9,60		9,53
19	10,13		10,20
20	9,63		9,40
21	9,67		9,51
22	9,63		9,14
23	11,34		12,20
24	9,91		10,50
25	9,81		10,10
26	9,49		9,81
27	11,97		12,70
28	10,87		11,50
29	12,38		13,10
30	9,35		8,86
31	11,66		11,80
32	8,39		7,75
33	10,54		10,10
34	10,90		10,70
35	11,92		12,90
Best regards.
 

Miner

Forum Moderator
Leader
Admin
#4
The short answer is yes, you can use the Test for Equal Variances on this data. However, there are other tests that will better utilize the information that you have available.

See, the Test for Equal Variances throws out a lot of information contained in your data. That is, it will ignore the fact that you have paired data and treat it the sample as unpaired data. Yes you will have some useful results, but not as useful as others.

For example, you can analyzed this type of paired data using a Bland Altman plot.

If you are a Minitab user, there are macros that will create the Bland Altman plot. If you use Minitab 16, there is a new feature called Orthogonal Regression that is designed for this specific scenario.
 
B

Berg.Jlle

#5
Miner,

Thank you for the reply. By the way I use Minitab 16.

I read about the orthogonal regression in the Minitab ("Bloodpressure.MTW" example), but I didn't understand the interpretation of results neither the Y/X error variance ratio = 0.90. How was this value obtained ? :frust:

Please, if it's not asking too much, could you guide me with the orthogonal analysis for my paired data ?
:eek:

Regards.
 

Bev D

Heretical Statistician
Leader
Super Moderator
#6
Can you post your data in excel?
Can you tell us more about what you trying to understand? Not which statistical test you are trying to perform but what real world question(s) you are trying to answer
 
B

Berg.Jlle

#7
Can you post your data in excel?
Can you tell us more about what you trying to understand? Not which statistical test you are trying to perform but what real world question(s) you are trying to answer
Hi Bev D,

Surely I can post the data in excel, please see attached.

So, the situation is:
I have two equipment (A and B) measuring the voltage at a certain point of mobile phones. I used a sample of 35 mobile phones that were measured in both equipment A/B and data were recorded in the attached table.

>> What I am trying to answer in the "real world" is as follows:

1) If the equipment are equal (alpha 0.05) about to the means.
To this question I used "Paired t" test that shows no significant difference between the means of A and B.

2) If the equipment are equal (alpha 0.05) about to the variances or standard deviation, i.e. if the variability are equal for A and B.
This is my doubt.

:thanx:
 

Attachments

Miner

Forum Moderator
Leader
Admin
#8
Take a look at the attached analysis. Ask any specific questions that you may have, and I will respond.

The biggest concern that I would haveover this data is that you would expect the data to fall on either side of a 45 degree line. The Orthogal regression equation should be close to B = A. In this case the slope of the line is close to 1.3 instead of 0. There is also an overall bias greater than -2 for the means. Now, the 2-sample t and the test for equal variances both indicate that this is not significant, and that is true FOR THE RANGE OF VALUES STUDIED. If you exceed this range, there is every indication that it will become significant.
 

Attachments

B

Berg.Jlle

#9
Take a look at the attached analysis. Ask any specific questions that you may have, and I will respond.

The biggest concern that I would haveover this data is that you would expect the data to fall on either side of a 45 degree line. The Orthogal regression equation should be close to B = A. In this case the slope of the line is close to 1.3 instead of 0. There is also an overall bias greater than -2 for the means. Now, the 2-sample t and the test for equal variances both indicate that this is not significant, and that is true FOR THE RANGE OF VALUES STUDIED. If you exceed this range, there is every indication that it will become significant.
Hi Miner. :)

I checked the file attached and I have some questions:

1) Using the regression orthogonal, we have the equation B = - 2.854 + 1.280*A. In order to consider that the two instruments (A and B) measure the same thing, the slope must be 1 and the constant 0, instead of 1.280 and -2.854, respectively.
Since these conditions are not met, can I conclude that the instruments A and B do not measure the variability equally. Right ?

2) Using the test for "2 Variances", this shows that there is no significant difference between the devices A and B, contrary to what was stated with regression orthogonal test in item #1.
Can I conclude that the test "2 Variances" does not show the reality and that the variability of instruments A and B are really different ?

3) To make the orthogonal regression analysis you used the "Error Variance Ratio" (B/A) equals to 1.
How did you come to that conclusion/valor ?


:thanx:
 

Miner

Forum Moderator
Leader
Admin
#10
Questions 1 & 2) If you take the results of the orthogonal regression, the Bland Altman Plot, the paired t-Test and the test for equal variances together, I would draw the following conclusion. The slope between tests A and B is not equal to 1 and there is a bias between the two tests. Within the range of data that you evaluated, this difference is not statistically significant and the two test method could be treated as equivalent. HOWEVER, there is a strong indication that if you were to exceed this range of data, that difference would become statistically significant. As I see it you have three options:
  1. If you intend to stay within the range of data, use the two tests as equivalent.
  2. Expand the range of testing, solidify the orthogonal regression equation, and use it to correct the results of the new test.
  3. Reject the new test and try another.
Question 3) That is based on the best information that you have available at the time and was supported by the test for equal variances. If you follow option 2 above, you will have better data, and should revise the Error Variance Ratio to fit your new data.
 
Thread starter Similar threads Forum Replies Date
optomist1 Interpreting Minitab Test For Equal Variances Using Minitab Software 3
optomist1 Two Way ANOVA Test For Equal Variances Using Minitab Software 10
M z Test vs. F Test - The most appropriate for comparing the Variances Statistical Analysis Tools, Techniques and SPC 3
N Test for Equal Variances - Two factor data set - Minitab Using Minitab Software 3
P Bartlett or Levene Test Software - Homogeneity of Variances Software Quality Assurance 8
ScottK Minitab 13 - Test for Equal Variances Using Minitab Software 3
K Flow of Descriptive Stats Analysis and Unequal Variances Help (Minitab) Using Minitab Software 3
S Data have unequal variances Six Sigma 4
R Gage R&R - Anova Method - Standard Deviations vs. Variances Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 12
H DIN 12880:2007-05 Clause 7.2 Test Equipment Description CE Marking (Conformité Européene) / CB Scheme 0
E Usability for Near Patient Testing Test Kit Human Factors and Ergonomics in Engineering 2
B SAE J 524 Pressure Test/Possible conflicting requirements? Manufacturing and Related Processes 0
J Automated test lead switcher? Calibration and Metrology Software and Hardware 0
J Applicable Great Britain regulations for a custom-built test rig or evaluation kit used for research purposes only. Design and Development of Products and Processes 0
U Leakage test Standard for Medical Device IEC 60601 - Medical Electrical Equipment Safety Standards Series 3
A MSA and Poison (Red Rabbit) Test Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
K ISO 17025 Method Validation and Verification for Test Lab ISO 17025 related Discussions 4
J IEC60601-1: HiPot Test Method for Internally Powered Devices IEC 60601 - Medical Electrical Equipment Safety Standards Series 3
P Comparing Two Test Variables Using Attribute Data Inspection, Prints (Drawings), Testing, Sampling and Related Topics 0
M Disinfection: Test recommended ISO 13485:2016 - Medical Device Quality Management Systems 0
H Microbiology count test failure Other Medical Device Related Standards 1
S AMS 2769 - weekly frequency test AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 2
S How to determine sample size to test when the acceptance criteria is x >= Standard value? 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 7
Kevin Walters AS9100 and AS9120 AATT Course and Test AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 1
W Conducting an MSA Test on scales Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
T Lot Release Test and Certificate of Analysis 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 6
John Broomfield Construction Inspection and Test Plan Misc. Quality Assurance and Business Systems Related Topics 5
Z COVID-19 Antigen Test Kit (Self-test) Registration in Peru Other Medical Device Regulations World-Wide 0
R GearBox Test Bench Calibration General Measurement Device and Calibration Topics 2
N 60601-2-34 Electrosurgery interference Test IEC 60601 - Medical Electrical Equipment Safety Standards Series 0
M Gage R&R automated test equipment Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
F Self-test minimum requirements IEC 60601 - Medical Electrical Equipment Safety Standards Series 5
H Regarding the validity of EMC & Safety test reports without NABL & ILAC-MRA mark. CE Marking (Conformité Européene) / CB Scheme 2
Z Implement methods of Round Robin test Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
D 62304 code review, unit test, integration test example IEC 62304 - Medical Device Software Life Cycle Processes 2
D DOD First Article Test Plan & Report Manufacturing and Related Processes 4
Stoic Details of Operational Qualification (OQ) test design for plastic extrusion processes, effect of material property noise, and GHTF/SG3/N99-10 US Medical Device Regulations 2
A Earth (ground) leakage current routine test IEC 60601 - Medical Electrical Equipment Safety Standards Series 1
G Audit & Agreements for "Test Laboratory" Supplier? US Medical Device Regulations 4
G Developing our custom test methods and test equipment US Medical Device Regulations 5
K Looking for a tape to perform cross hatch test on powder coating parts Reliability Analysis - Predictions, Testing and Standards 3
G Change Control for Test Samples Inspection, Prints (Drawings), Testing, Sampling and Related Topics 3
MaHoDie Summative Usability test Human Factors and Ergonomics in Engineering 2
A Calibration of spring loaded test fixture General Measurement Device and Calibration Topics 5
E Pressure vessels hydraulic test IEC 60601 - Medical Electrical Equipment Safety Standards Series 0
C Spinal implant predicate device test results 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 0
S Statistical Test Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 0
Ed Panek Looking for Test Resource General Measurement Device and Calibration Topics 1
R Microbiological test (USP61, USP62, USP71, ISO 11137.1, ISO11137.2) --- Testing flow chart Other Medical Device Related Standards 0
S Discussion on OBL and OEM test data for submission as per new EUMDR EU Medical Device Regulations 0

Similar threads

Top Bottom