dikranmarsupial
I don’t think that is true, I would have thought that if this uncertainty were taken into account in computing the p-value it would make it more difficult to get a low p-value (as there is more noise/uncertainty, so H0 can “explain” a wider range of observations).
As I’ve asked a few other times with no responses on this blog, should it make any difference if the researcher decides to use p<0.1 instead of p<0.05 ?
Should it make any difference when that decision is actually made?
You have taken Prof. Curry to task for her statements regarding one possible null hypotheses relating to the "hiatus". Hopefully you will continue to build upon these exchanges in the future.
In the meantime, it would be helpful to hear your opinions concerning the fact that the Karl, et al., hiatus-killing paper did not achieve p<0.05 (and did not achieve p<0.1 for land or ocean for the years 1998-2012).
From Karl, et al., at page 2:
It is also noteworthy that the new global trends are sta-tistically significant and positive at the 0.10 significance level for 1998–2012 (Fig. 1 and table S1) using the approach described in (25) for determining trend uncertainty. In contrast, IPCC (1), which also utilized the approach in (25), reported no statistically significant trends for 1998-2012 in any of the three primary global surface temperature datasets. Moreover, for 1998–2014, our new global trend is 0.106± 0.058°C dec−1, and for 2000–2014 it is 0.116± 0.067°C dec−1 (see table S1 for details). This is similar to the warming of the last half of the 20th century (Fig. 1). A more comprehensive approach for determining the 0.10 significance level (see supplement) that also accounts for the impact of annual errors of estimate on the trend, also shows that the 1998–2014 and 2000–2014 trends (but not 1998–2012) were positive at the 0.10 significance level.