This is an armchair question by a non-pilot, so I acknowledge going in that it may prove to be 'dumb'. Nevertheless:
ET302 was moving fast--fast enough to suggest they had plenty of juice to climb. It seems fair to rule out engine failure or something like an explosion that might have greatly increased drag.
Meanwhile, early evidence (the commentary from the pilot watching ET302 leave, plus the available flight data) suggests that ET302's pilots had, and knew they had a problem right after takeoff, while flaps were probably still extended. As discussed, MCAS isn't supposed to come into play until flaps are retracted and in any case it's intended to counter issues that arise during turns and at slower speeds than what ET302 was doing. Regardless, a pilot that's aware of MCAS--as Ethiopian has said its pilots were--should theoretically be able to disable it and fly manually if they suspect an issue.
Next there's the hypothetical matter of faulty AoA sensors--e.g., the sensors indicate a stall and the pilots don't want to put the nose up to climb lest they exacerbate the situation. This might make sense at night or during poor visibility, but this happened in the morning with good visibility. Surely, at ~1000' AGL and 300+ kn, the pilots would be able to determine that they were not entering a stall and that the sensor was wonky. Again, you'd think this might lead to a decision to fly manually--and conditions seemed conducive to doing so, if it came to that.
Anyway, finally to the question. If a crew encounters control problems soon after takeoff, is it accepted practice to stay at lower altitude, or would it be more normal to climb (if possible) to a higher altitude that allows for more margin for error while determining the nature and extent of the problem? Basically I'm wondering to what extent we can infer from the flight staying at ~1000' AGL that they COULDN'T climb, vs. that they CHOSE not to climb because things were weird and they were planning to immediately return to the airport.
(In my own uneducated, totally speculative view, their speed suggests they were probably trying to climb yet couldn't. That, in turn, suggests two possibilities: that they didn't disengage the automatic systems, whether through choice or ignorance, and were indeed fighting the plane a la JT610; or that the control problem was unrelated to MCAS or other automatic systems regardless of whether they disengaged them. Loss of hydraulic power, stripped jackscrews, or other 'maintenance' issues have caused control loss in past incidents, except this was a new aircraft and ET has a decent reputation. Meanwhile, terrorism or explosions also seem unlikely both because of the speed/drag issue noted above, and especially because I don't think we'd have a worldwide grounding if there was evidence for extrinsic causation. So, to me, and again purely speculatively, some version of the 'pilots fighting the plane' scenario seems most supported by logic and current evidence. If so, it obviously remains to be determined whether this entailed some kind of pilot error, or a new wrinkle that the pilots couldn't have foreseen.)