Hi rafaelcortes88,
That difference in Scaled Scores is probably due to a few factors. The scoring algorithm for the Official GMAT is remarkably complex; it includes a variety of factors - the relative difficulty level of the questions, whether you were expected to get them correct or incorrect, the relative position of the question, what's going on "around" the question, whether the question even counted or not (some questions are "experimental" and are worth 0), if you didn't answer questions at the end, etc.
Based on these scores, I'd be curious about some things:
1) How did you perform in the first half of each section?
2) Did you run out of time and not answer any questions (if so, then how many?)?
3) Did you run short of time and have to guess on a bunch of questions (if so, then how many?)?
4) Were either of these CATs a retake (had you taken either of them before)?
5) Did you take the ENTIRE CAT both times (including the Essay and IR sections)?
6) When reviewing these CATs, how often did you "narrow it down to 2 choices and then guess?" How often did you "guess wrong?"
7) Did you "skip over" an entire RC passage and just guess on the accompanying questions?
8) Did you do ANYTHING that would be considered non-testlike while taking either of these CATs? Think hard; the little details matter.
GMAT assassins aren't born, they're made,
Rich