If you refer to the manufactures of speakers, then even here in Denmark, some speakers are not measured in anechoic chambers - because it's expensive and you can get a decent result, with a gated measurement. Like Atkinson points out - they often sum a nearfield from the bass-unit with a gated from the tweeter/midrange.I was astonished by his admission that they can't be bothered to use an anechoic chamber.
Geddes fx. use a moore-smoothing - where you have more and more resolution, the higher you get in frequency. But I do not think it's important to use a specific method, rather than being honest and consistent. It is very easy to create false or invalid data - so again - it can be tricky business to end up with a clean, great souding and finished/final/complete project.