If you use a spot meter and measure flush against a light box and get an EV reading of 12, and then place a light red filter (B+H 090) on the light box and measure again and get an EV reading of 11, what is that telling you?
Is your filter compensation 1 stop when B+H says it should be factored at 5X, or 2.2 stops by my reckoning?
I don't think you could go by the reading or published factor unless the spectral sensitivity of the meter matched that of the film. Supposedly, this is what the Zone VI modification of the Pentax analog spot meter did. The best course would probably be to just run a test bracketing under the light source you will use for the photographs you'll take. I would think daylight vs. tungsten or fluorescent speeds could differ significantly with a filter such as light red. For example, with warmer tungsten light the filter factor would be considerably less than with a colder source. There is more of the spectrum present for the filter to block with a daylight source so the effect of the filter would be relatively greater compared to warmer tungsten light sources.
Filter factors are at best a good guess - and nothing more
They also tend to be slightly generous for exposure - just to make sure you get something on the Neg
Good as your Spotmeter is - the sensor isn't matched to the response characteristics of your film (and all the films are different) - so putting the filter in front of the meter isn't accrate either
As in the previous post - the filter factor is also affected be the temperature of the light you are working with.
This is true out doors outside in the landscape almost as much as it is in doors under artificial lighting - have you ever noticed how ineffective a dark red filter is completely cloudy sky - not just the colour of the sky but the depth of the shadows too
Flare is a factor as well. If there is a lot of lightbox white around your filter as you measure it, the flare will alter your reading. Some spotmeters are poorly baffled against flare.