# What is an ensemble of particles in quantum mechanics?

According to Griffith's Introduction to Quantum mechanics in section 1.5 Momentum, he said that " the expectation value is the average of repeated measurements on an ensemble of identically prepared systems, not the average of repeated measurements on one and the same system"

So, my question is what does the ensemble of particles actually mean and how making a large identical copies of a single particle helps to determine the expectation values of any dynamical variable?

• If you were to repeat the same measurement on the same system, it would lock the system into that measured value, which would likely be different from the ensemble average. So, the expectation value can not be interpreted as the average of repeated measurements on the same system. Mar 29 at 11:03
• So, you are saying that, our wavefunction is in superposition of different states and this different states act as a single particle (in our classical sense) which on measurement gives a different value with averages out as the expectation value of that dynamical system? Am I Right? Mar 29 at 11:12
• I'm saying that measurement loosely speaking implies a collapse of the wavefunction into the state corresponding to the measured value. If you repeat the measurement, and there is no time for the system to evolve naturally to another state, you'll get the same value for the measurement. Even if you leave time, the distribution of measured values and the mean of those will not correspond to the expectation value in general. Mar 29 at 11:17
• Ok, Thanks for the solution 😁 Mar 29 at 11:49
• Warning: Griffith's statement shall not be confused with the "ensemble interpretation" of QM. Ensemble interpretations: the wave function describes an ensemble of identically prepared systems, in contrast with the “orthodox” interpretations (the wave function provides as complete a description as is possible of an individual system). Griffith is still describing the "orthodox" POV, explained in the comment of @Raskolnikov. See: "Statistical Interpretation of QM" doi.org/10.1103/RevModPhys.42.358 and "Ensemble interpretations of QM" doi.org/10.1016/0370-1573(92)90088-H Apr 1 at 13:25

Measurement (in classical sense) always involve uncertainty and has to be done repeatedly in order to estimate the value of the measured quantity and its variance: $$\overline{x} = \frac{1}{N}\sum_{i=1}^Nx_i,\\ \text{var}(x) = \overline{(x -\overline{x})^2} = \frac{1}{N-1}\sum_{i=1}^N(x_i-\overline{x})^2$$ In classical situations we can perform measuring repeatedly on the same object. E.g., if one wants to measure the weight of an object, we weigh it several times and calculate the average - if we don't see it done in everyday life, it is because we are either using very precise instruments (or conversely do not bother about precision.) But repeated measurements are routinely done in any serious technological or scientific situations. In most situations measurements done on a single object should produce the same results as measurements on different but identical objects - in the latter case we call it sample averaging or ensemble averaging (the former is more statistical term, the latter is more common on physics.)