As far as determining turbulence by airspeed changes...this is a ridiculous concept.
While I have very limited experience of flying in turbulence, I agree. The qualitative guidelines for turbulence severity talk about what happens to unsecured objects. This implies that what we're really measuring is accelerations. Were aircraft to routinely be equipped with g-meters, would this be a more appropriate quantitative measure?