AUGH. Stop with the Professor Jones thing already! The whole question was *designed* to get a talking point for people who don't like global warming.
Question: "Do you agree that from 1995 to the present there has been no statistically-significant global warming?"
Jones: "Yes, but only just. I also calculated the trend for the period 1995 to 2009. This trend (0.12C per decade) is positive, but not significant at the 95% significance level. The positive trend is quite close to the significance level. Achieving statistical significance in scientific terms is much more likely for longer periods, and much less likely for shorter periods."
This is a total no-brainer: statistical significance is a measure of how likely it is that your results are due to variance in the sample--broken stations, noisy data, rounding errors, and plain-old weather. In ANY time series involving short-term variations, measurements over a short period of time have lower significance in predicting a long-term trend because the data is very noisy.
You couldn't, for example, tell me with much confidence whether the earth is warming over decades based on a single hour's observation, or samples running from June to January. It's like trying to predict how much you'll weigh in ten years by measuring your weight for a week.
Because of the extreme seasonal and year-to-year variability of climate, it takes about *thirty years* to extract a statistically significant measure of a small-scale (relative to seasonal change) trend like global warming. The question is therefore meaningless--even though the data is consistent with global warming, the probability that that consistency is due to chance is too high given limited time. More importantly, when you *do* include enough samples to reach two-sigma significance (that 95% confidence he's talking about), those data *do* indicate global warming.
The question was clearly intended to confuse those who don't understand statistical significance--and it clearly worked, as evidenced by Fox: