Repeat and Average

One reading can be off, so good scientists measure the same thing several times, add the readings and share them out to find the average — and a reading that does not fit the rest is an anomaly worth a second look.