(Building off this earlier question about spin-testing airliners.)
For normal-category aircraft1 in the U.S. (at least), only single-engine (or, in the case of some gliders, no-engine) airplanes need to demonstrate the ability to recover from a one-turn (or, if the aircraft is intended to be certified for intentional spinning, six-turn) spin. This means that, for instance, the single-engine Cessna 208 (MTOW 3630 kg) has to be successfully recovered from a one-turn spin, whereas the much-smaller Piper PA-34 (MTOW 2155 kg) is exempt from this requirement solely by virtue of having a second engine.
There is nothing about multiengine aircraft that makes them inherently resistant to spinning compared to single-engine aircraft; indeed, if anything, it would be expected to make spins somewhat more likely (think "aftermath of botched or absent pilot response to an asymmetrical engine failure"2). If, instead, the number of engines is being used as a proxy for the size and intended audience of the aircraft (as suggested in this comment on the aforelinked question), then I don't see any reason not to simply use a dividing line based directly on weight (especially since, as noted above, there are many large single-engine airplanes and small multiengine airplanes).3
Given all this, why aren't multiengine airplanes (especially small multiengine airplanes, such as the aforementioned PA-34) required to demonstrate spin recovery like their single-engined compatriots are?
1: Utility and aerobatic airplanes do have to demonstrate spin recoverability even if they have more than one engine.
2: Especially for multiengine tractor-propeller aircraft with wing-mounted engines, where, when flying at very low airspeeds, simply shutting down one engine or losing it to a mechanical failure can be enough to immediately stall its respective wing (due to said wing losing the air previously being blown over it at high speed by said engine's propeller).
3: Granted, even large multiengine airliners can find themselves spinning on occasion, something that (inadvertently) happened to (for instance) the first 307 and the first 737;4 however, I accept that an airplane requiring an ATPL and a specific type rating, not to mention a second pilot, is (by virtue of hopefully-good piloting) rather less likely (barring a major structural failure) to be placed in a spin than one requiring merely a PPL, and also that very large aircraft such as these are likely to become debris clouds if spun.
4: For the record, both the Stratoliner and the 737 were successfully recovered from their spins, although the former then almost immediately disintegrated in flight due to an overly-aggressive pullout from the subsequent dive.5
5: Hey, I can nest footnotes!