← Meta Interview Insights

Meta·Data Scientist·Technical Phone Screen·Senior

Senior
Jul 2026

Summary

Meta DS interview with a gnarly causal inference question about a feature that only auto-loads for opted-in power users, so you can't just run a clean A/B. The whole thing was a deep dive into quasi-experimental methods and how to communicate bias risk to stakeholders.

Questions Asked (4)

Q1

A feature is auto-enabled only for power users who opted in, so you can't randomize cleanly. Which quasi-experimental method would you use (DiD, Synthetic Control, IV, or PSM), and why? Walk through the identifying assumptions for your choice.

A/B Testing & ExperimentationTechnical Trade-offs
Author's notes

This is where I spent most of my mental energy.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Acknowledge the selection bias from opt-in power users, then recommend Difference-in-Differences (DiD) as the most practical choice, while briefly noting why the others are less suitable. Walk through the key identifying assumptions—parallel trends, no anticipation, and no spillovers—and discuss how to test or mitigate violations.

Pro tip: Emphasize that the parallel trends assumption is untestable but can be supported by pre-trend analysis and placebo tests; also mention that if power users are fundamentally different, consider combining DiD with propensity score matching (PSM) to create a more comparable control group.

1. Diagnose the selection bias

Explain that opt-in power users are not representative, so randomization is impossible and naive comparisons would be biased. This sets the stage for why a quasi-experimental method is needed.

2. Choose DiD and justify

Select DiD because it controls for time-invariant differences between power users and others, and can leverage a natural experiment (e.g., a policy change or feature rollout) that affects one group but not the other. Briefly contrast with IV (needs a valid instrument), Synthetic Control (needs a donor pool and pre-period fit), and PSM (only controls for observed confounders).

3. State the identifying assumptions

List the core DiD assumptions: parallel trends (absent treatment, the outcome trends would be the same for both groups), no anticipation effects, and no spillovers (treatment of one group does not affect the other). Also mention stable composition and no other concurrent shocks.

4. Discuss validation and robustness

Describe how to test assumptions: plot pre-treatment trends, run placebo tests (e.g., fake treatment dates), check for anticipation, and consider robustness checks like changing the control group or using synthetic control as a sensitivity analysis.

5. Acknowledge limitations and alternatives

Admit that DiD may not fully address unobserved time-varying confounders. If parallel trends is questionable, suggest combining with PSM or using synthetic control if a good donor pool exists. Mention that IV could be used if a valid instrument is available.

Key Points to Mention

  • Selection bias due to opt-in power users
  • Parallel trends assumption and how to test it (pre-trend plots, placebo tests)
  • No anticipation and no spillover assumptions
  • Why IV requires a valid instrument (relevance and exclusion restriction)
  • Synthetic control requires a donor pool and good pre-period fit
  • PSM only controls for observed confounders, not unobserved ones
  • Combining DiD with PSM for robustness

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

How would you design placebo tests and pre-trend checks for your chosen method, and what robustness checks would you run?

A/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

Talked through running the analysis on a period before the feature launched to see if the treatment effect shows up where it shouldn't.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the chosen method (e.g., difference-in-differences, synthetic control, or switchback) and the causal assumptions it relies on. Then explain how you would design placebo tests and pre-trend checks to validate those assumptions, and finally outline a set of robustness checks to probe sensitivity to violations and alternative specifications.

Pro tip: Emphasize that pre-trend checks are necessary but not sufficient—placebo tests on unaffected outcomes or pre-periods can catch violations that pre-trends miss. Also, mention that you would pre-register your robustness checks to avoid p-hacking.

1. Clarify the method and its assumptions

Briefly state the chosen method (e.g., DiD, synthetic control) and the key identifying assumptions, such as parallel trends or no anticipation. This sets the foundation for the tests you propose.

2. Design pre-trend checks

Explain how you would test for pre-existing trends, e.g., event-study plots, placebo tests on pre-periods, or formal tests like the parallel trends test. Mention that you would check for differential trends in the pre-treatment period.

3. Design placebo tests

Describe placebo tests such as using a fake treatment date, a fake treatment group, or an unaffected outcome. Explain how these tests can detect violations of assumptions and provide evidence for the validity of the main results.

4. Outline robustness checks

List robustness checks such as alternative specifications, different control groups, sensitivity analysis (e.g., Rosenbaum bounds), and testing for heterogeneous effects. Mention that you would also check for spillover effects and anticipation.

5. Interpret and communicate results

Discuss how you would interpret the results of these tests and communicate them to stakeholders, including any limitations and potential biases. Emphasize transparency and reproducibility.

Key Points to Mention

  • Parallel trends assumption and event-study plots
  • Placebo tests: fake treatment date, fake treatment group, unaffected outcomes
  • Robustness checks: alternative specifications, control groups, sensitivity analysis
  • Anticipation effects and spillover effects
  • Pre-registration of analysis plan to avoid p-hacking
  • Use of synthetic control or other methods if parallel trends fail

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

What sensitivity analyses would you run, and how would you use negative controls or Rosenbaum bounds?

A/B Testing & ExperimentationRoot Cause Analysis
Author's notes

Negative controls I had a decent answer for: pick an outcome the feature logically shouldn't affect and check if your method still finds an effect.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing sensitivity analysis as a way to test the robustness of causal conclusions to unmeasured confounding and modeling assumptions. Then, describe specific techniques like negative controls and Rosenbaum bounds, explaining when and how you would apply them. Finally, tie it back to the business context at Meta, emphasizing how these methods build trust in experimental results.

Pro tip: Emphasize that sensitivity analyses are not just statistical exercises but tools to communicate uncertainty to stakeholders and guide decision-making. Show that you can balance rigor with practicality by prioritizing analyses that address the most plausible threats to validity.

1. Clarify the causal question and assumptions

Restate the goal of the analysis and the key identifying assumptions (e.g., no unmeasured confounding, correct model specification). Explain that sensitivity analyses test how violations of these assumptions affect conclusions.

2. Identify plausible threats and choose sensitivity analyses

Discuss potential biases (e.g., unmeasured confounding, selection bias, measurement error) and select appropriate sensitivity analyses such as negative controls, Rosenbaum bounds, or E-values. Justify your choices based on the context.

3. Explain negative controls

Describe how negative control outcomes or exposures can detect unmeasured confounding. Give an example: if a negative control outcome shows an effect, it suggests bias. Explain how to interpret results and adjust if needed.

4. Explain Rosenbaum bounds

Detail how Rosenbaum bounds quantify the strength of unmeasured confounding needed to overturn a significant result. Explain the sensitivity parameter (Gamma) and how to interpret it (e.g., Gamma=1.5 means confounder would need to increase odds of treatment by 50% to explain away effect).

5. Interpret and communicate findings

Summarize how you would integrate results from multiple sensitivity analyses to assess robustness. Discuss how to present findings to stakeholders, including limitations and confidence in causal claims.

Key Points to Mention

  • Definition and purpose of sensitivity analysis in causal inference
  • Negative control outcomes and exposures: examples and interpretation
  • Rosenbaum bounds: Gamma parameter, interpretation, and limitations
  • E-values as an alternative or complement to Rosenbaum bounds
  • The importance of pre-specifying sensitivity analyses to avoid p-hacking
  • How to communicate uncertainty and robustness to non-technical stakeholders

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q4

How would you explain the directional bias risks of this quasi-experimental approach compared to a proper randomized experiment, to a non-technical stakeholder?

A/B Testing & ExperimentationStakeholder Management
Author's notes

Honestly the part I felt best about.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging the stakeholder's need for a clear, non-technical explanation, then use a relatable analogy to illustrate how quasi-experimental methods can systematically over- or under-estimate effects compared to randomization. Emphasize that randomization is the gold standard because it balances all factors, while quasi-experiments may leave residual confounding that biases results in a particular direction.

Pro tip: Use a concrete example from the stakeholder's domain (e.g., ad campaigns) to make the bias tangible, and always quantify the risk when possible (e.g., 'we might overestimate lift by 10-20%') to build trust and show business impact.

1. Set the context

Briefly explain why we use quasi-experimental methods (e.g., when randomization isn't feasible) and that they aim to estimate causal effects but can be biased.

2. Introduce the analogy

Use a simple analogy, like comparing two groups of plants where one gets more sunlight, to show how unaccounted differences can skew conclusions.

3. Explain directional bias

Clarify that bias can go either way: we might think an effect is larger or smaller than it truly is, and this direction depends on the unmeasured factors.

4. Contrast with randomization

Highlight that randomization acts like a fair coin flip, balancing all factors (known and unknown) on average, so any difference is likely due to the treatment.

5. Discuss implications and mitigation

Explain how this bias could affect decisions, and mention ways to assess or reduce it (e.g., sensitivity analysis, triangulation with other methods).

Key Points to Mention

  • Randomization ensures exchangeability of groups, eliminating systematic differences.
  • Quasi-experiments rely on assumptions like parallel trends, which may not hold.
  • Directional bias means the estimated effect is consistently off in one direction, not just noisy.
  • Unmeasured confounders can push results away from the true effect.
  • Sensitivity analysis can quantify how robust results are to unmeasured confounding.
  • Always communicate uncertainty and avoid overstating causal claims from quasi-experiments.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.