It is important to understand that measurements
(observations) may be precise but not accurate if they are closely
grouped together around a value that is different from the
expectation
("true" value) by a certain amount. Also, observations may be
accurate but not precise if they are well distributed about
the expected value but are significantly dispersed from one
another. Finally, observations can be both precise and accurate
if they are closely grouped around the expected value (the
distribution
mean).
Is it better to be accurate then Precise?
This question is at the heart of any experimental scientist. Any scientist will tell you it better to be both but that kind of a cop out. Here is the break down of precision and accuracy.
To run an experiment and to get experimental value that is accurate but not precise, then you have a great deal of random error but have little systematic error.
To run an experiment and to get experimental value that
is precise but not accurate, then you have a great deal of systematic error
but have little random error.
So in my opinion i think to be precise and not accurate is better the the other way around. To have poor precision means that you didn't plan well enough, but to have poor accuracy means that you lack lab skills. This isn't a hard and fast rule because when performing some experiments it may be hard to avoid random or systematic error.
