Question Regarding Standard Deviation of Rate Vs. Time

nai2499

New Member
Hello, I am not particularly well versed in statistics but I am doing some research and encountered something I didn't understand and was hoping someone here could help.

For my research, we need to find the pump speed with the most consistency when filling a vial to 50ml. I am not sure if I should use the standard deviation of the time it takes to fill the 50 ml or the rate (ml/s). Now to me, it seems like there shouldn't be any difference but clearly there is. By time, Pump 8 is best and by rate Pump 6 is best.

Could anyone tell me (1) if this is a good way to find the consistency of the pump and (2) why the standard deviation of the time is different from the rate. ( I understand why the number itself is different but not why the ranking of deviation is different. )

Here is the link to the google doc. I hope this is allowed but if not I will attach the PDF. Thank you very much!

Attachments

• 33.8 KB Views: 2

katxt

Well-Known Member
One SD is the SD of Time, the other SD of Rate = 50/Time. The division means that you need to go through the CV's or use the delta method if you prefer.
In short, this turns out to be SD(Rate) = SD(Time)*50/mean(Time)^2 so all is well.
As to the preferred method I would have thought that the SD of the delivered volume was the best gauge. If the pump delivers the 50 ml by working for a specific time then the SD time would be important.

AngleWyrm

Active Member
For my research, we need to find the pump speed with the most consistency when filling a vial to 50ml. I am not sure if I should use the standard deviation of the time it takes to fill the 50 ml or the rate (ml/s). Now to me, it seems like there shouldn't be any difference but clearly there is. By time, Pump 8 is best and by rate Pump 6 is best.
Mean for the rate in ml/s wasn't calculated in the original, so I madea copy and added it (blue outlined cells)

I also agree the time to complete the fill operation is a better measure.
The average speed in ml/sec varies considerably between pumps, and so a small variance in the smallest stream may be of less merit than the most consistent delivery times.

Last edited:

katxt

Well-Known Member
Or, more likely, use the one with the best relative SD = SD/mean