A 'space launch' should include even suborbital launches, so long as they cross in to space (100km altitude, for example, though slightly different parameters could be used). If you only wanted orbital missions that would involve different wording.
I sort of feel that failed launches shouldn't count, only because this allows for easy adjudication against efforts that were not serious attempts to begin with. Also if we are 'keeping score' so to speak we shouldn't be giving credit for failures! But this is not a strong opinion.
Test launches should count.

@BorisBartlog The awkwardness of the 100km line in my mind is that it would include ballistic missile trajectories, and the stats would be completely dominated by the Iran-Israel war. This seems very clearly outside of the original spirit of how we had created the question.

@eapache Hm, yes. I hadn't considered that.

@eapache
To me "all space launches worldwide" means the purpose is to launch stuff to space.
An ICBM purpose is to deliver a missile to a target on earth. The fact is might go through space is incidental and doesn't reach level required to be the purpose of the missile to go to space.

So I don't think ICBMs and anything similar should count.
A launch of an experiment that is not deployed from the rocket and a key requirement is to return the experiment safely to the ground, is in my opinion in a different category where the purpose includes intending to take the experiment to space.

Perhaps the above is a fine line to draw but I think it makes sense to me. Purpose seems adequately important to reach this decision IMO.

Having put purpose as such a critical factor, I seem to end up thinking failed launches should count. I guess you could say if there is a failed attempt to put something up and it fails then you try again this might be only one intention and possibly only count once but I doubt I would want to do that. It just makes it messy and difficult to distinguish from testing the rocket.

A launch with no payload which is just a test seems like the purpose of the test it to get the rocket at least into space in order to test its ability to do so, so I end up suggesting this does count. So Starship launches do count if the intent is to go above 100km (regardless of altitude actually reached). If rocket explodes or otherwise goes wrong before liftoff then it hasn't launched and doesn't count

I also tend to think suborbital flights like New Shephard launches intended to go above 100km should count as it is "all space launches worldwide" and not "all orbital space launches worldwide"

(I don't believe I have any holdings in the market nor voted here. Just suggestions of what I think, feel free to ignore if you disagree.)

My current opinion is that failed launches should count, but not suborbital ones. If we trust wikipedia's numbers there were 330 such launches in 2025, of which 165 belonged to SpaceX. That is exactly half, but not "more than half", so the question should resolve "no".