>I have a suspicion that I’m going to get slightly different dispersion distances for n = +1 and n = -1, possibly because of the grating, being plastic, it’s not completely flat but is curving. Depending on where the dot occurs, if it’s on a part that’s more curved, relative to the opposite side of the zero order image, we might get discrepancies there.
That would be really strange, I think. The light from the laser dot on the distant object illuminates the full grating (just like a star illuminates the full aperture if the telescope.) The grating then produces a number of beams at different angles according to the spectrum order, which are then brought to focus by the lens. Distortions in the grating could blur the spectrum because the effective line spacing varies across the grating but I don’t understand how it could affect the +- orders differently. If you are seeing different dispersions in +- orders I suspect It would be more likely to be due to aberrations in the lens (eg barrel/pincushion type distortions for example.