Quantum Mechanics is strange, and we don't really have any idea of why it is as it is. We see particles apparently communicating infinitely fast to produce apparently random yet coordinated outcomes. We can predict with amazing accuracy the probabilities of outcomes and yet it seems that no-one can predict precisely what will happen. This mystery has led to a range of models of what is actually happening with quantum states, and there are more than just the two models that are widely known: Copenhagen and Many Worlds, but I'll start with those.
The “Copenhagen” interpretation of Quantum Mechanics says that quantum states remain undecided until some kind of measurement takes place, and at that time of measurement the state collapses into one of the possible outcomes. The mystery with this interpretation is that no-one can say what “measurement” actually means. It could be the interaction of the quantum state with some laboratory equipment, or it could be as soon as there is an interaction with a single atom or particle. This interpretation has led to vast amounts of nonsense about “observer effects”, in which the mind is supposed to have some role, but that's not a sensible view of things – the human mind is a physical system which is no more special than any other physical system when it comes to interaction with quantum states.
The “Many Worlds” interpretation says that quantum states don't collapse, because all the possible outcomes always happen. We don't see all those possible outcomes because we are physical systems like any other, and our existence is split when we encounter an quantum state so that there is one of us seeing each possibility. This interpretation is widely used in physics, but I find it unhelpful, as it doesn't answer the question of why THIS copy of me sees what I see – each copy of me will be unable to predict which of the outcomes it sees.
Now, some other ideas:
The “Transactional” interpretation says that as soon as a quantum state interacts with anything at all, an interaction (“transaction”) back and forth through time occurs during which one of the possible outcomes is picked (at random). The link through time avoids the issue of instant communication through space. There is no need for any idea of “measurement”, and there is no “interaction at a distance”.
Roger Penrose's quantum gravity interpretation says that quantum states can remain undecided only while they are below the level at which there is a significant difference in the gravitational effect of possible outcomes. Once that level has been reached the quantum system is forced to collapse into a single state.
I could go on and on. The point is that there are many more ways of looking at quantum mechanics. It's more than just a matter of either “observer effects” or “quantum worlds”. Right now, we have no way to determine by experiment which of these interpretations is correct, so anyone drawing any conclusions from a particular view of quantum mechanics is on dodgy ground!