If Arp is right that quasars have an "internal redshift" (in other words are much closer than expected) then you would not expect quasars to have intervening absorption at greater redshift. So that does not provide evidence in favour of the cosmological redshift at all. But I didn't start this thread to try and debate this, but to do my own analysis to test it because I want to be sure myself what is correct.

The very large scatter in the quasar redshift versus brightness diagram is a reason to suspect that redshift accurately measures distance. Of course it might be that they really have a huge range of brightnesses. One other thing that I would like to do with a quasar survey is to look at the various properties of quasars, particularly any parameters that might be indicative of their age, as that would be correlated with intrinsic redshift if Arp is right.

What survey would be best to use that has most other measurements of quasar properties (e.g. colour indices, other emissions, etc)? It does not matter if the sample size is not great as long as they are reasonably representative.

The idea is to see whether it is possible to make an estimate of their intrinsic redshift and produce a much tighter external redshift (being observed minus estimated internal) versus brightness diagram like for galaxies.

I think that if the scatter can be considerably reduced then that would be proof of a non-cosmological component to redshift.