Chimpy
First Post
Can the method used to determine the average review score be adjusted? I say this because I think it's too easy for someone to skew a score by giving a high or low rating.
Statistically, the mean is poor figure representative figure because it is influenced too much by a small number of extreme pieces of data. Perhaps for ENWorld ratings, a median or modal value would be a better choice?
For example, take these numbers:
1, 6, 8, 8, 9, 9, 9, 9, 9, 9, 10
The mean is 7.9, but the median is 9.0. The median is more representative of what most of the figures are. The mean is skewed down because of the single low number.
An alternative method might be to only take the middle, say, 90% if the scores.
Statistically, the mean is poor figure representative figure because it is influenced too much by a small number of extreme pieces of data. Perhaps for ENWorld ratings, a median or modal value would be a better choice?
For example, take these numbers:
1, 6, 8, 8, 9, 9, 9, 9, 9, 9, 10
The mean is 7.9, but the median is 9.0. The median is more representative of what most of the figures are. The mean is skewed down because of the single low number.
An alternative method might be to only take the middle, say, 90% if the scores.