P value and statistical significance: The two-tailed P value equals 0.0437 By conventional criteria, this difference is considered to be statistically significant. Confidence interval: The mean of Group One minus Group Two equals -1.61392857142900 95% confidence interval of this difference: From -3.17781916844734 to -0.05003797441066 Intermediate values used in calculations: t = 2.1462 df = 21 standard error of difference = 0.752
If anyone could shed some light, I'd be very grateful.
You can understand it this way:
if you filp 1000 coins, you were expecting around 500 heads 500 tails. But you got 100:900. You want to test whether the coin is biased or that the weird no. is due to random chance, and you do 100 more trials. End up 95 of the 100 trials gives ~100:900, and you can say that this weird ratio is not due to random chance, and that the coin is biased.
The degree of freedom and standard error are just numbers that help you calculating t value.
If you still dont get it, you might also want to post this in Math forum?