Measurement

Quantum measurement is fundamentally unlike classical observation. When you read a classical bit, you learn its value and the bit is unchanged. When you measure a qubit, you do two things simultaneously: you extract one classical bit of information, and you irreversibly destroy the qubit’s quantum state. Superposition collapses to a definite outcome. Entanglement with other qubits is broken. Whatever quantum information was not encoded in the measurement outcome is gone.

This destructive nature is not an engineering limitation; it is a physical law. It shapes the entire design philosophy of quantum algorithms.

The details

A qubit in state $|\psi\rangle = \alpha|0\rangle + \beta|1\rangle$ gives outcome $0$ with probability $|\alpha|^2$ and outcome $1$ with probability $|\beta|^2$, according to the Born rule. After measurement, the qubit is in the state corresponding to the outcome: $|0\rangle$ or $|1\rangle$.

Measurement is always relative to a chosen basis. The standard computational basis $\{|0\rangle, |1\rangle\}$ corresponds to projecting onto the Z-axis of the Bloch sphere. Other bases are also used:

• X-basis $\{|+\rangle, |-\rangle\}$: implemented by applying a Hadamard gate before measuring in the Z-basis
• Y-basis $\{|i\rangle, |-i\rangle\}$: implemented with an $S^\dagger$ gate followed by Hadamard then Z-basis measurement
• Arbitrary basis: any unitary rotation before measurement changes the effective measurement basis

Multi-qubit measurements can be performed on individual qubits (single-qubit measurement on each) or as joint measurements that project onto entangled basis states. Bell measurements, used in quantum teleportation, project two qubits jointly onto the four Bell states rather than the four product basis states.

In quantum circuits, measurement is drawn as a meter symbol at the end of a qubit wire:

|ψ⟩ ──[H]──[CNOT]── ... ──[M]── (classical bit output)

Mid-circuit measurement is also possible: measure some qubits partway through, use the classical outcome to conditionally apply gates to remaining qubits. This is essential for error correction syndrome extraction and for certain adaptive algorithms.

from qiskit import QuantumCircuit

qc = QuantumCircuit(2, 2)
qc.h(0)
qc.cx(0, 1)
# Both qubits are entangled; measuring gives correlated outcomes
qc.measure(0, 0)  # Measure qubit 0, store result in classical bit 0
qc.measure(1, 1)  # Measure qubit 1, store result in classical bit 1
print(qc.draw())

Why it matters for learners

Understanding measurement is essential for understanding why quantum algorithms are structured the way they are. Because measurement destroys superposition, you cannot read out intermediate results to check your work. Every gate must be applied blind, in a carefully designed sequence, so that the final measurement yields the correct answer with high probability.

This is why quantum interference is so important: algorithms use interference to concentrate amplitude on the correct answer before measurement. The measurement itself is trivial; the engineering is in shaping the state beforehand.

Measurement also explains the need to run circuits many times. A single run yields one classical outcome. To estimate probabilities, check for errors, or confirm an algorithm’s correctness, you repeat the circuit thousands of times (called shots) and aggregate the distribution.

Common misconceptions

Misconception 1: Measurement just reveals a pre-existing value. Before measurement, the qubit has no definite value. The outcome is genuinely indeterminate until measurement occurs. This is not ignorance about a hidden classical fact; it is a fundamental property of quantum mechanics confirmed by Bell inequality experiments.

Misconception 2: You can measure a qubit without disturbing it. Quantum measurement is necessarily disturbing. Gentle or partial measurements exist (weak measurements) but they trade information gain for less disturbance; they do not fully avoid it. Any measurement that gives you information about a qubit’s state changes that state.

Misconception 3: Measuring in the wrong basis gives garbage output. Measuring in a different basis gives a valid outcome in that basis; it is not garbage. However, it may not give you the information you want. If a qubit is in $|+\rangle$ and you measure in the Z-basis, you get a random 0 or 1, which loses the phase information that distinguishes $|+\rangle$ from $|-\rangle$. Choosing the right measurement basis is part of algorithm design.

See also

• Born Rule
• Superposition
• Qubit
• Quantum Interference
• Decoherence