I think in most fuel injected aircraft with a well functioning mechanical pump and pipes it does not seem you will need boost pump ON on fuel tank change (assuming it is wet) once the engine at running at cruise rpm
Obviously, switching from a dry fuel thank, I am sure you will need the boost pump ON (even on a high wing aircraft 
)
Some checklists say to start on less-full tank, taxi on it, change to fullest tank for engine checks and take-off, with electric pump on for take-off after the checks.
That’s exactly what the TB20 has, and I thought it was standard for low wing. The reason is simple: verify both paths comprising the fuel system, before takeoff.
Regardless I start, taxi, and take off on the same tank. At a safe height I change.
I would do that only if for some reason there was not enough engine running time to verify both paths, before one had to take off. Perhaps if somebody had launched an ICBM with the airport being the target (I can certainly think of some UK airport cafes being a suitable target) and I had to get away quick
I still can’t see a technical reason, for a typical low wing aircraft, for the fuel pump to be ON during tank change. I absolutely know that the TB20 doesn’t need that.
Some checklists say to start on less-full tank, taxi on it, change to fullest tank for engine checks and take-off, with electric pump on for take-off after the checks.
Regardless I start, taxi, and take off on the same tank. At a safe height I change.
The DR1050 always takes off on front tank, which has some gravity feed if not in a climb.
On a number of occasions I’ve had a problem on changing tanks. Not detected until I switched the electric pump off. Causes:- leaks in fuel system, and once a partially blocked finger filter in the rear tank.
The M20K POH specifically states “Fuel Boost Pump – OFF” in the “Takeoff” section (pg. 4-10) as well as the “Before Landing” one (pg. 4-14). I don’t recall it being “ON” anywhere earlier in the normal procedures. The only reason to turn it on is during high ambient temperatures, to overcome vapor locks, and in emergency situations (power loss, etc).
We have done this before (in fuel management) but I change tanks in the TB10 quite often to keep the aircraft balanced – there is no aileron trim and if you let any significant imbalance develop then it will pull noticeably to one side.
Also because the gauges are next to useless, and hence the above forms part of my fuel management strategy for any non-trivial A to B flight. The notes on my knee board contain ‘paper gauges’ where I draw and cross out blocks of available fuel at timed intervals, usually 15 mins which is approx. 10 litres.
I have never noticed any sort of effect on the engine or any fluctuation of the fuel pressure gauge when changing tank, whether with the pump on or off. Of course the pressure goes up a bit with the pump on, but the change of tank doesn’t seem to do anything. Hence I change per my fuel management schedule without regard for what’s below, but then I have not made a longer water crossing than Weymouth to Guernsey.
I don’t change to the fuller tank in my pre-landing checks, not unless a change is due per the schedule.
As a supplementary question I wonder how many people change tanks after a specific time interval or for some other predetermined reason regardless of whether it is otherwise a good time to do so?
The chances are the engine will not quit straight away but with a minute or two if it is going to. Waiting to change tanks until such time as you arent over “hostile” terrain or water always seems to me to make sense. I wouldnt change tanks over the middle of the north sea for example if I could help it.
What about changing to the fuller tank before landing?
Interestingly POH checklist for my Grumman tiger (lycoming 0-360)makes no mention of electric pump use for tank change, so I just monitor fuel pressure during change, never seen a fluctuation.
On an carbureted engine, very low pressure will deliver enough fuel. Gravity flow is enough when the tanks are above the engine.
I think that the fuel supply design in the Warrior is marginal, such that you can have unexplained fuel pressure deviations. This happened to me crossing the Skagerak (Brrr!) and was only resolved by changing both Electric and Engine pumps, with no fault found in either. Further evidence is the low pressure permitted (0.5 psi) for what reason? Added to which the fuel changeover valves are notoriously leaky, leading to a smell of fuel when changing tanks. Btw, my variations never went below 2.5 psi, but why any drop at all?
I suspect the ‘pump on’ procedure is to plaster over known design defects when using the engine pump only.
I can only assume the reason to turn on the electric pump on (for example) a carburated Lycoming is to refill the float bowl with certainty after moving the tank selector through the off position has taken longer than planned. However the mechanical pump would seem to do it just as well.
Interesting that this has come up, I was thinking about this question only yesterday.
