On 8 Dec, 22:17, "The Medway Handyman"
wrote:
Roger Chapman wrote:

You really are displaying an appalling lack of scientific (and in this
particular case mathematical) knowledge. The Met Office in their
wisdom show the 95% confidence limits as well as the actual figure
for each year. If you take 95% to be +/- two standard deviations then
for the 2 to be equal a reduction of 0.3 of a standard deviation in
the higher and an increase of 0.3 in the lower (or any other
combination with the same sum) is all that is required.

In other words they are fudging the figures.

Two different years - yes or no?


What Roger is saying is that you are not going to get a definitive
temperature to 10 decimal places for each year so you can't compare
years like that. You have to take into account the statistics of the
data sets and there will be some fluctuations around the broad
trends. This means that one data set may show a different warmest
year from another while still demonstrating the same general trend.
Roger showed that the data is indeed in general agreement.