Significance in Least Squares

[CycloWizard]

I'm drawing a complete blank on this one for some reason. I have a bunch of data and I fit it with a line. How do I determine if the slope found from least squares regression is statistically different from zero? /brain fart

 

[lollybo]

Since the population parameters B0 and B1 are unknown, the sample points that you drew have a LS line with an estimated slope (B1 hat) and an estimated intercept (B0 hat). To test the sample statistic if it is statistically significant, If the sample size is large enough, you can assume it is normally distributed with a mean of the population parameter B1(in this case, it is zero) and a standard deviation of s/sqrt(Sxx). Find the critical value, and if it is reasonably small, you can assume the null hypothesis can be rejected.

 

[CycloWizard]

Thanks. I forgot that I had posted that, but since then I've used an analysis of variance approach that seems to give me what I'm looking for.