Skip to content
Tech Interview Prep
Technical interview guide

Entanglement & Measurement

How entangled qubits produce correlations with no classical analog, why measurement is irreversible, and the no-cloning theorem's practical implications.

Read
52 min
Practice MCQs
25
Interview Q&A
25
Edition
v2
Editorial status
Reviewed

Scope: IBM Quantum learning and Qiskit current documentation; Bell and CHSH primary literature reviewed 2026-09-04.

Curated: · Written: · Reviewed:

25 questions

MCQ-2A team prepares the two-qubit state |Φ+⟩ = (|00⟩ + |11⟩)/√2 and hands one qubit to Alice and the other to Bob, who are in separate labs. Alice measures her qubit in the computational basis {|0⟩, |1⟩} and records her outcome. Bob then measures his qubit in the same basis. What correlation between their recorded outcomes does the Born rule predict?
MCQ-7A 2-qubit register is prepared in the state |ψ⟩ = (3/5)|00⟩ + (4/5)|11⟩. A measurement is performed in the computational basis, and the outcome is recorded as the projector P = |00⟩⟨00|. What is the probability that this measurement returns outcome 0 (i.e. the state collapses onto |00⟩)?
MCQ-9A transmon qubit exposes only Z-basis (computational-basis) readout. A protocol must report the expectation value of the Pauli X operator on each qubit, so every shot needs an X-basis measurement. What should the pulse sequence apply immediately before the readout?
MCQ-1A two-qubit pure state is prepared in the computational basis as |ψ⟩ = (1/2)(|00⟩ + |01⟩ + |10⟩ + |11⟩). Its coefficient matrix, with rows indexed by qubit A and columns by qubit B, is C = (1/2)[[1,1],[1,1]]. Which conclusion about this state is correct?
MCQ-8A two-qubit system is prepared in the state |ψ⟩ = (1/√3)(|00⟩ + |01⟩ + |10⟩). The first qubit is measured in the computational basis and the outcome is 0. What is the normalized post-measurement state of the two-qubit system?
MCQ-4A team prepares two qubits in the state |ψ⟩ = (√3/2)|00⟩ + (1/2)|11⟩ and sends only qubit A to a remote node, discarding qubit B. What is qubit A's reduced density matrix ρ_A?
MCQ-5A two-qubit state is prepared as |ψ⟩ = (3/5)|00⟩ + (4/5)|11⟩, and a colleague reports that its Schmidt rank is 2. A second state is prepared as |φ⟩ = (1/√2)(|01⟩ − |10⟩). Which statement about these two states is correct?
MCQ-20A simulator prepares a two-qubit register in a pure state |ψ⟩_AB, then traces out qubit B. The reduced density matrix ρ_A comes back with eigenvalues 0.5 and 0.5, so S(ρ_A) = 1 bit. What does that nonzero entropy establish about the register?
MCQ-21A two-qubit pure state is prepared as |ψ> = cos(π/12)|00> + sin(π/12)|11>. You need to report its concurrence. Which value is correct?
MCQ-6A two-qubit state is prepared and tomographed. Its density matrix is reported as ρ = 1/2|Φ+><Φ+| + 1/2|Φ-><Φ-|, where |Φ±> = (|00> ± |11>)/√2. A teammate concludes that ρ is entangled because it is a convex mixture of two maximally entangled Bell states. What is the correct assessment?
MCQ-10A two-outcome measurement on a qubit is proposed with effects E0 = [[1/2, 0], [0, 3/2]] and E1 = [[1/2, 0], [0, -1/2]]. The device is tested on state |1> = (0,1)^T. Which statement correctly gives the probability of outcome 1 and the validity of this proposal?
MCQ-11A simulator's mid-circuit measurement node is given only a POVM and rebuilds the post-measurement state from the effects alone. Two backends declare the same measurement on a qubit initialised to |0>: effects E_0 = E_1 = I/2, so both report p(0) = p(1) = 0.5, and both then run the same Z-basis measurement on that qubit. Backend A returns 0 on that second measurement for every shot; backend B returns 1 on exactly the shots whose first outcome was 1. Through the node as it stands, both backends emit the same second-measurement record, so at most one of them can be right. Which change to the node fixes B without breaking A?
MCQ-22A one-qubit circuit starts in |0>, applies H, then T, then H, then measures. An engineer moves the measurement earlier, placing it between H and T, and adds a final measurement. What is the final measurement distribution after the change, and why?
MCQ-3A two-qubit circuit prepares a Bell pair: H on q0, then CNOT(q0, q1). A teammate inserts a Z gate on q0 between the H and the CNOT, leaving everything else unchanged, and then an H is added to each qubit before readout so both qubits are measured in the X basis. With 4000 shots and no noise, which counts does the modified circuit give?
MCQ-12A two-qubit source produces pairs. In 10,000 shots measured in the Z basis, the outcomes are 5,000 |00> and 5,000 |11>, with no |01> or |10>. The team claims the source is entangled. They now measure both qubits in the X basis. Which result would show that the data so far are still consistent with a separable mixture of |00> and |11>?
MCQ-13Alice and Bob share entangled photon pairs in the singlet state. Alice always measures in the Z basis. Bob randomly chooses the Z or X basis for each photon. In 1000 trials, Bob chooses Z 500 times and X 500 times. Alice's outcome is + in 250 of the Z-basis trials and 250 of the X-basis trials. Bob then selects only the trials where his own outcome was +. In the Z-basis subset, Alice's outcome is + in 0 of 250 trials; in the X-basis subset, Alice's outcome is + in 125 of 250 trials. A team claims this proves Bob's basis choice changes Alice's local statistics. Which assessment is correct?
MCQ-15A Bell test runs four measurement settings and reports correlators E(a,b) = 0.9, E(a,b′) = −0.9, E(a′,b) = 0.9, E(a′,b′) = 0.9, with S = E(a,b) − E(a,b′) + E(a′,b) + E(a′,b′). What does S imply about the run?
MCQ-16A photonic Bell test logs 10,000 coincidences for each of the four analyser-setting pairs. Alice switches between settings a1 and a2, Bob between b1 and b2, and each detection is recorded as +1 or −1. E(a,b) is the mean of the product of the two outcomes over the coincidences at that pair, and the measured values are E(a1,b1) = −0.98, E(a1,b2) = −0.34, E(a2,b1) = −0.17, E(a2,b2) = −0.94. What is the CHSH value S, and what does it imply about a local hidden-variable description of these data?
MCQ-24An experimentalist evaluates a two-qubit hardware channel by measuring the Clauser-Horne-Shimony-Holt (CHSH) parameter S = <A0 B0> + <A0 B1> + <A1 B0> - <A1 B1>. Why does an observed value of S = 2.5 definitively demonstrate bipartite entanglement without reconstructing the full density matrix rho?
MCQ-14A two-photon Bell test reports S = 2.42 ± 0.05 on the CHSH inequality, with the measurement settings chosen by a hardware random-number generator. The detectors fire on 8% of the photons the source emits, and the team keeps only the runs in which both detectors fire. What does this run establish about local hidden-variable models?
MCQ-19A CHSH test on polarization-entangled photon pairs runs between two labs, with analyzer settings chosen after the photons leave the source, so loss is the only open issue. Each arm's calibrated end-to-end detection efficiency is 0.40. In one run the coincidence counter keeps 2,000,000 events where both arms fired, and the analysis script drops 6,000,000 events where exactly one arm fired. Over the kept events the analysis returns S = 2.62 +/- 0.01. What should the team report?
MCQ-18A two-qubit CHSH test runs on IBM Quantum's `ibm_brisbane`. Qubit A has readout error 1% and qubit B has readout error 5%, so each qubit's recorded bit is flipped with those probabilities independently of the true outcome. The qubits are prepared in `|Φ+> = (|00>+|11>)/√2` and measured in the Z basis, where the ideal correlator `E = P(00)+P(11)-P(01)-P(10)` is 1.00. What does the asymmetric readout error do to the observed joint outcome frequencies and the measured E?
MCQ-17A team estimates the expectation value of a Pauli observable P (eigenvalues ±1) on a superconducting processor. Each shot returns +1 or −1, and after 10,000 shots the sample mean is 0.60. They report the result as 0.60 ± σ, where σ is the standard error of the mean. What is σ?
MCQ-23A team reconstructs a two-qubit state by linear inversion from 9 Pauli-basis settings with 1000 shots each. The resulting matrix has eigenvalues 1.03, 0.02, −0.01, −0.04 (trace 1.00), and they report the negativity N = 0.05 as evidence of entanglement. Bootstrap resampling of the shot data gives a 95% interval of [0.00, 0.09]. Which conclusion is correct?
MCQ-25When estimating bipartite entanglement measures (such as subsystem purity or second Renyi entropy) across a subsystem of size $N_A$ qubits via local randomized measurements, what fundamental scaling limitation governs the statistical sample complexity compared to targeted Pauli witness estimation?