# Significance test comparing slopes from linear fit

#### iiwanovic

##### New Member
I would like to compare four slopes I got from linear fits through five data points for each fit. The fitting results (for the slopes) are as follows:
1: 0.08885 ± 0.00991
2: 0.08744 ± 0.0118
3: 0.10288 ± 0.00669
4: 0.0926 ± 0.01285

My hypothesis is that they are all not significantly different from each other and I would like to put that into a number. Can I run an unpaired t-test putting N=5? I guess I should loose a degree of freedom due to the linear fit. Or are there other tests better suited for this problem? Thanks a lot for your help in advance.

#### hlsmith

##### Not a robit
Would you want to compare fits based on residuals? Visualization are always great too. If you plan to generalize the model, you will also need to contemplate overfitting.

#### iiwanovic

##### New Member
The output of the origin fit:

Equation y = a + b*x
Plot ln(d) ln(d) ln(d) ln(d)
Weight Instrumental
Intercept 4.85867 ± 0.03419 4.61573 ± 0.07007 4.30094 ± 0.03586 3.68573 ± 0.06233
Slope -0.08885 ± 0.00991 -0.08744 ± 0.0118 -0.10288 ± 0.00669 -0.0926 ± 0.01285
Residual Sum of Squares 56.84869 155.07012 163.38903 64.12215
Pearson's r -0.98183 -0.97374 -0.99371 -0.97232
R-Square(COD) 0.96399 0.94817 0.98746 0.9454
Adj. R-Square 0.95198 0.93089 0.98328 0.9272