Remove 2011 Remove Analytics Remove Slice and Dice
article thumbnail

Measuring Validity and Reliability of Human Ratings

The Unofficial Google Data Science Blog

It then goes on to show how a new framework called cross-replication reliability (xRR) implements these concepts and how several different analytical techniques implement this framework. If they roll two dice and apply a label if the dice rolls sum to 12 they will agree 85% of the time, purely by chance.