The question, what's the meaning of observation in the sense of quantum physics is not so trivial as it might seem. There are a lot of pretty "esoteric" claims even by very famous people. E.g., John von Neumann, who was the first to give a mathematically strict formulation of non-relativistic quantum mechanics, revealing the Hilbert-space structure, applying the spectral theorems for non-bound self-adjoint operators to make sense of continuous spectra, etc. had ideas about the "measurement problem" which sound very strange (at least to me). It's known as the Princeton interpretation of quantum mechanics, and I don't want to get into any detail about this. The upshot is that he thinks the result of a measurement is established only when a conscious mind has become aware of it. For further details have a look at
https://en.wikipedia.org/wiki/Von_Neumann–Wigner_interpretation
In my opinion that's nonsense. Nowadays the experimentalists, say at the LHC doing experiments in ultrarelativistic proton-proton or heavy-ion scatttering, take notice of the outcome of an experiments much later than it was done. The results of the measurement are the momenta of zillions of produced particles in all kinds of detectors in the big experiments (ATLAS, CMS, LHCb, ALICE) in form of data in huge computer files. Then the experimentalists uses these "raw data" from the computer file and analyzes them, making plots etc. Then we get aware of what's come out, e.g., that in certain invariant-mass distributions after careful background subtraction and statistical analysis "a Higgs-boson like particle is seen".
In my opinion, it's very mondane, when an observation is made: It's made at the moment when the observable of interest of the quantum system (say an elementary particle) is somehow registered by a macroscopic measurement device and this information is somehow irreversibly stored.
This is not very different as with any probabilistic experiment. Throwing a die leads with some chance to a number between 1,...,6 which is practically unpredictable and thus can be seen as a probability experiment. If the die is not manipulated the probability to get a certain result is 1/6, and this can be tested repeating many (uncorrelated) experiments, and we can give a significance in how far the prediction of the probability 1/6 for each possible outcome is true or not. The more often we repeat the experiment the better we can tell whether the probabilistic prediction is compatible with the experiment or not.
The difference to the quantum mechanical probability is just that according to quantum mechanics, even when we know everything about the quantum system, i.e., we have prepared it in some pure state, not all observables have a determined value. This is not due to some lack of knowledge about the state of the system but because, according to quantum theory, the observable has no determined value due to the preparation of the system in the pure state. Knowing the pure state only tells us the probability to find a certain value when an observable is measured, and again we can make a probability experiment by repeating the measurement of the observable many times, the more often the better. A certain value of the observable is known as soon it is measured by an appropriate (macroscopic) device and this information somehow stored so that we can take notice of it, as in the example of throwing dice. For the double-slit experiment the particle's position is established as soon as it hits the screen, and you can count the pixels on your photo plate (or CCD in a more modern lab) to figure out, whether the predictions of quantum theory are correct with regard to the position probability distribution of not, and the more particles you run through the apparatus the more precise becomes your statistical significance for success or failure of this prediction.
This is the socalled minimal statistical interpretation, which is the one making the least assumptions and which is, in my opinion, the most natural one. Of course, it's a bit disappointing that we don't know more about the system than the probabilities for the outcome of a measurement, but that's how Nature seems to be, according to quantum theory. Physics doesn't provide explanations for why Nature is as it is but only describes, how Nature is (or at least that part of Nature that is objectively and reproducibly established by observations).