Engle Granger and Error Correction Model coursework

[Woolsey]

I'm a 3rd year Economics student and I have some coursework where I'm using an Engle-Granger two step model and include an error correction model if necessary, to test whether expectations theory holds (Whether there's a relationship between long and short run interest rates), however I am a bit stuck as the textbook is unclear of what to do when residuals from the cointegrating equation are found to be non-stationary.

- Firstly, I have tested both variables and they are both I(1).

- Then I regressed long-term rates on Short-term using OLS
Equation:
spainlr c spainsr

-I have ADF tested the residuals from the equation above and they are non-stationary.

So from this,
1) Am I correct in concluding that there's no long run relationship between the two and that they are not cointegrated?

2) Would I also be correct in saying that an error correction model is not applicable as the variables are not cointegrated?

3) Is there anything else that I can do to this model to test for a relationship (short or long term)?

4) Finally are there any other statistical tests that I should be running on anything?

Any help would be greatly appreciated,
Thanks