Validity, Accuracy, Reliability (1 Viewer)

[Ekman]

Hey everyone, I just need some clarification here,

For performing a first hand investigation on the precipitation of ions, is it true that:

-The validity of the experiment is based on how well you can control contaminants and other external factors (i.e. temperature and pH)

-The accuracy of the experiment is based on how much yield of the ions you can precipitate out

-The reliability of the experiment is based on the consistency of your results, hence if you only perform the experiment once, then your results are unreliable?

Thanks

 

[InteGrand]

Hey everyone, I just need some clarification here,

For performing a first hand investigation on the precipitation of ions, is it true that:

-The validity of the experiment is based on how well you can control contaminants and other external factors (i.e. temperature and pH)

-The accuracy of the experiment is based on how much yield of the ions you can precipitate out

-The reliability of the experiment is based on the consistency of your results, hence if you only perform the experiment once, then your results are unreliable?

Thanks

Reliability requires multiple trials of the experiment to be performed. Reliability cannot be evaluated when only one trial has been performed. It is meaningless to speak of reliability if there's only one trial. Once multiple trials are performed, reliability can be gauged based on how close the data from each experiment is (via things like standard deviation, coefficient of variation, etc.).

Accuracy is based on the yield you achieved compared to the theoretical value you should have received. The percentage error can be calculated as:

.

Your percentage accuracy is then just: .

A percentage accuracy of 90% or greater is generally good for typical school lab experiments.

For an experiment to be valid, it should have all variables controlled. I think other criteria for validity included that the results were reliable and accurate (but I'm not 100% sure on this).