replication
How repeated observations of each score were reduced to the single score compared here, or null when the scores were not observed repeatedly.
Recorded because two studies over the same runs can reach different conclusions purely through this choice, so a result that does not say which was used cannot be interpreted.