You could give the relative yields of the products either as isolated yields, or based on some spectroscopic analysis like GC-MS, HPLC or NMR. It depends on your facilities and the rentention factors of the compounds using specteoscopic means. Without knowing the exact nature of your compounds, I would suggest 1H NMR from the crude. Internal standards are usually prefered in these cases.
You know the selectivity of any position in your active site of molecule depends on the chemo and stereo selective site.also depend on what your reaction program.
Thank you Mr. Sandtrov. Is it actually necessary to characterize the side products or obtaining the yield of pure product after crystallization will be sufficient to calculate the selectivity.
Yield and selectivity are not the same. If you want to report reaction selectivity, you need to either isolate the side product(s) (which can be very hard if they are formed in small amounts), or use NMR (or other techniques like GCMS or HPLC) on the crude reaction mixture. Without knowing the details of your work, I think an isolated yield of the product and a crude NMR with an internal standard should give you sufficient information to provide yield and selectivity.
It is not that difficult to give that in batch even when you are not able to detect all of your products. However, it is important that you can quantify your starting compound and the desired product. Thus:
Selectivity (%)= mol desired product / (mol starting compound-mol starting compound left after reaction)*100.