1: \begin{abstract}

2: 

3: There are major discrepancies between recent ICFT (Intermediate Coupling

4: Frame Transformation) and DARC (Dirac Atomic R-matrix Code) calculations

5: (Fern\'andez-Menchero {\em et al.}~2014, {\em Astron.\ Astroph.}~{\bf 566} A104,

6:  Aggarwal {\em et al.}~2016 {\em Mon.\ Not.\ R.~Astr.\ Soc.}~{\bf 461} 3997)

7: regarding electron-impact excitation rates for transitions in several Be-like ions, as

8: well as claims that the DARC calculations are much more accurate and the ICFT results might even be wrong.

9: To identify possible reasons for these discrepancies and to estimate the accuracy 

10: of the various results, we carried out independent B-Spline R-Matrix (BSR) calculations

11: for electron-impact excitation of the Be-like ion N$^{3+}$.

12: Our close-coupling expansions contain the same target states (238 levels overall) 

13: as the previous ICFT and DARC calculations, 

14: but the representation of the target wave functions is completely different.

15: We find close agreement among all calculations for the strong transitions

16: between low-lying states, whereas there remain serious discrepancies for the weak transitions

17: as well as for transitions to highly excited states.

18: The differences in the final results for the collision strengths are mainly due to differences

19: in the structure description, specifically the inclusion of correlation effects, rather 

20: than the treatment of relativistic effects or problems with the validity of the 

21: three methods to describe the collision. 

22: Hence there is no indication that one approach is superior to another,  

23: until the convergence of both the target configuration 

24: and the close-coupling expansions have been fully established.

25: 

26: \end{abstract}

27: