Quote:Not understanding what "significant" means makes it clear you don't have a background in science. If results are statistically significant it means that you accept or reject the null hypothesis (something like there is no warming or warming = 0). There are various test to determine the degree of significance usually based on the probability of having a 5% (p < 0.05, 95% confidence) chance of finding a difference as large or larger than the one in your study given that the null hypothesis is true. I'll give you a very basic illustration of how this works. 

 

Say I am interested in the effect of solar radiation on tropospheric warming, and using average temperatures over a span of 50 years as an endpoint. Say you took 100 temperature measurements for each of those 50 years (you might need to run a power calculation to determine if this is an adequate sample size).You could use other endpoints but I am using temperature as one in this example. 

 

HO (null hypothesis): Solar activity has no significant effect on tropospheric warming.

HAalternative hypothesis): Solar activity has a significant effect on tropospheric warming. 

 

So in this very very simplistic exercise you might run a ANOVA (this would probably be a bad test because the data is likely not normally distributed and the test is not very robust, Kruskal-Wallis, PERMANOVA, etc might be better). on the averages using a program like R.

 

The output will have a test statistic (F-value) and the degree of significance. Let's say we were testing at p < 0.05. If our results indicate p = 0.002, then we would reject the null hypothesis and accept the alternative. We reject the null because at the significance level of 0.05, the results obtained happen too frequently for us to be confident that solar activity has a effect on tropospheric warming. The opposite can also happen. If p = 0.06, the you would accept the null and the results would indicate that solar activity has no effect on tropospheric warming. Again this is very very basic. 

 

You could do other things like create a linear mixed model but I won't get into that.

Blah Blah blah. A TLR treatise on statistics. You apparently want to claim that the evidence that CO2 is involved in the warming is statistically significant.  But that's not pertinent to the comment I was referring to. How significant is the warming? What percentage of the recent warming is caused by CO2?