Can anyone please help me find the outliers in my data?

Watch

Announcements

Consultation launched for GCSE and A-level assessments in 2021. Have your say >>

Applying to uni? Find your group chat here >>

start new discussion reply

Page 1 of 1

Go to first unread

Skip to page:

Loreto2018

Badges: 20

Rep:

Report Thread starter 1 month ago

So I did an Elisa lab experiment and im confused about which bits of my data are outliers in my patient samples and my standards. Can anyone please help me out?
Standards:
0.0454 0.0407 0.0489
0.0581 0.0582 0.0541
0.0720 0.0626 0.0593
0.0900 0.0880 0.0722
0.1297 0.1341 0.1036
0.2264 0.1782 0.1787
0.3377 0.2281 0.2413
0.0377 0.0291 0.0343

Patient:
0.0625 0.0609 0.0623
0.2391 0.0347* 0.2967
0.4204 0.2973 0.3202
0.0487 0.0519 0.0517
0.0570 0.0542 0.0506
0.1417 0.1294 0.1375
0.0428 0.0436 0.0467
0.3075 0.1963 0.2133

* is where i think the outlier is, please correct me if im wrong.
Thanks.

Last edited by Loreto2018; 1 month ago

Theloniouss

Badges: 20

Rep:

Report 1 month ago

I don't understand the data, but a common method of finding outliers is >upper quartile + 1.5*IQR or <lower quartile - 1.5*IQR

Jpw1097

Badges: 17

Rep:

Report 1 month ago

(Original post by Loreto2018)
So I did an Elisa lab experiment and im confused about which bits of my data are outliers in my patient samples and my standards. Can anyone please help me out?
Standards:
0.0454 0.0407 0.489
0.0581 0.582 0.0541
0.0720 0.0626 0.0593
0.0900 0.0880 0.0722
0.1297 0.1341 0.1036
0.2264 0.1782 0.1787
0.3377 0.2281 0.2413
0.0377 0.0291 0.0343

Patient:
0.0625 0.0609 0.0623
0.2391 0.0347* 0.2967
0.4204 0.2973 0.3202
0.0487 00519 0.0517
0.0570 0.0542 0.0506
0.1417 0.1294 0.1375
0.0428 0.0436 0.0467
0.3075 0.1963 0.2133

* is where i think the outlier is, please correct me if im wrong.
Thanks.

Can you correct the errors in your data. It seems as though you have put an extra ‘0’ here and there, if not, you have more outliers.

Loreto2018

Badges: 20

Rep:

Report Thread starter 1 month ago

(Original post by Theloniouss)
I don't understand the data, but a common method of finding outliers is >upper quartile + 1.5*IQR or <lower quartile - 1.5*IQR

Yeah I did try to use that method on excel but I would get some negative numbers for the lower bounds. So for example if you look at the one I have marked with a star, wouldn't that be an outlier out of the 3 data points there? I calculated the upper and lower bounds for each triplicate using functions on excel but they are all within range which is strange because I'm sure there's some outliers in my data, especially the one I have marked.

Last edited by Loreto2018; 1 month ago

Loreto2018

Badges: 20

Rep:

Report Thread starter 1 month ago

(Original post by Jpw1097)
Can you correct the errors in your data. It seems as though you have put an extra ‘0’ here and there, if not, you have more outliers.

Hi yeah well spotted I did make a few errors typing that but actually I forgot to includes some extra zeros. I've corrected them now. Basically they are in triplicates and I have to find the mean of each triplicate for both standards and patients. But before I do that I need to find the outliers.

Loreto2018

Badges: 20

Rep:

Report Thread starter 1 month ago

(Original post by Theloniouss)
I don't understand the data, but a common method of finding outliers is >upper quartile + 1.5*IQR or <lower quartile - 1.5*IQR

So there are 8 triplicates (3 values) for both the patient and standards. And for each triplicate I need to find the average but first I have to exclude any outliers within each triplicate. I'm just having trouble identifying which ones would be outliers.

Theloniouss

Badges: 20

Rep:

Report 1 month ago

(Original post by Loreto2018)
So there are 8 triplicates (3 values) for both the patient and standards. And for each triplicate I need to find the average but first I have to exclude any outliers within each triplicate. I'm just having trouble identifying which ones would be outliers.

I'd say the one you've identified looks like an outlier, but with only three data points (and if you're excluding the outlier, two) it's not really possible to say.

Loreto2018

Badges: 20

Rep:

Report Thread starter 1 month ago

(Original post by Theloniouss)
I'd say the one you've identified looks like an outlier, but with only three data points (and if you're excluding the outlier, two) it's not really possible to say.

That's what i thought but when i calculated the LB for the data sets 0.2391, 0.0347, 0.2967 i got a value of -0.0596 and the UB was 0.4644 and all of these values are within range which means that it wouldn't be an outlier right?

Theloniouss

Badges: 20

Rep:

Report 1 month ago

(Original post by Loreto2018)
That's what i thought but when i calculated the LB for the data sets 0.2391, 0.0347, 0.2967 i got a value of -0.0596 and the UB was 0.4644 and all of these values are within range which means that it wouldn't be an outlier right?

As I said, it's not really possible to know. By that definition it might not be, but there isn't really a single definition of an outlier and it depends on why you're looking for outliers and so on.

Loreto2018

Badges: 20

Rep:

#10

Report Thread starter 1 month ago

#10

(Original post by Theloniouss)
As I said, it's not really possible to know. By that definition it might not be, but there isn't really a single definition of an outlier and it depends on why you're looking for outliers and so on.

Ok well I will probably just keep the one I've identified in my data then because my tutor said if in doubt about an outlier, it should be kept in. Thanks for your help.