Episode description

Here we consider whether when collecting data we are able to distinguish between the signal (hits) and noise (false alarms). I make the case the author early on is doing a good job of explaining "random error" when conducting experiments. However, broadly speaking this is an issue of increasing precision in our measurements. No mention seems to be made, crucially, in understanding the possibility of systematic error (a problem for accuracy). How do precision and accuracy differ? Why won't repeating our experiments and collecting more data help guard against certain kinds of errors? All this and more discussed in this episode.