To answer your original question -- there are two types of modal analysis that can be used in ETABS: Eigen Vector analysis and Ritz Vector analysis.
Eigenvector analysis is the traditional modal analysis. Essentially, the modes are based on computing a stiffness and mass matrices (taking into account each degree of freedom in the structure), and then solving for the fundamental frequencies. Load combinations, and load cases have no effect on the eigenvector fundamental modes - they are solely dependent on the mass and stiffness of the structure. They are deemed the "Fundamental Modes of Vibration" thus directly dependent on only the structure and it's properties.
Raleigh-Ritz Vector analysis is another method of modal analysis. In ritz vector analysis, starting load vectors (i.e. your dead load, your seismic loading, nonlinear degrees of freedom, etc.) are used to determine the approximate fundamental modes of vibration. As your starting load vectors are based on your loading, the loading distribution will have an effect on your ritz vectors.
The modes determined by the eigenvector analysis and the ritz vector analysis will not be identical, but are usually similiar. I believe in most cases CSI recommends the use of Ritz vectors for the analysis.
In regards to mode shapes, with a simplified 3-story 2-D frame with the masses lumped at each story the modes will be as follows:
First Mode -- All masses move in the same direction (i.e. similiar to ASCE 7 pseudo lateral force loading)
Second Mode -- There will be one "kink" in the deflected shape.
Third Mode -- There will be two "kinks" in the deflected shape.
Obviously, once you have a three-dimensional modal that has torsional effects and many degrees of freedom many more mode shapes will be introduced, but typically these translational mode shapes should be dominant for most regular structures.
Keep in mind I'm not an expert on the matter. I've taken a Dynamics in Structures class taught by Chopra a few years back (his book is a good reference although I'll be the first to admit it is not easy to read through), but haven't done a lot of this recently.
I agree with the others, that asking questions is good and I don't think anyone should fault someone for asking questions. I ask them all the time. However, being the design engineer of a 30 story building when you are asking some basic questions in the use of ETABS is a little uncomfortable. Designing a building of that size is a lot of responsibility and there are a lot of "subtle" check boxes in ETABS that can have large impacts... i.e. assuming you are doing response spectrum analysis you'll have to worry about 1) mass source, 2) scaling of spectrum, 3) modal mass participation factors, etc. which will all have large impacts on the final results.