A Data Scientist at NORC at the University of Chicago serves at the intersection of rigorous social science research, advanced statistical methodology, and modern data engineering. You are not just building models; you are providing the empirical foundation for policy decisions, healthcare analytics, and large-scale public interest research. Your work directly impacts how organizations understand complex societal trends, health outcomes, and human behavior.
The role is both intellectually demanding and mission-driven. Because NORC at the University of Chicago operates as a non-partisan research institution, you will face unique challenges in data integrity, privacy, and the interpretability of your findings. You will collaborate with subject matter experts, statisticians, and project leads to transform raw, often messy datasets into actionable insights that hold up under intense academic and public scrutiny.
Recruiter Screen
reportedWhoever runs this call is usually not a practitioner. They take notes, and a hiring manager skims those notes later, so the real question is whether your work survives being written down by someone outside the field. Test every project sentence against that: could a non-specialist repeat it correctly without knowing what a propensity score is? Carry a plain-language version of each project and one reason you want this particular role that you could not copy onto another application. Vagueness at this stage reads as inexperience, even when the underlying work was genuinely deep.
What to demonstrate
- Whether a non-specialist can restate your projects accurately, since their paraphrase is what reaches the hiring manager
- Whether your reason for wanting the role points at the work itself rather than the company's reputation
- Whether your language signals the level being screened for: what you decided yourself versus what you were handed
How to prepare
- Write a two-sentence, jargon-free version of each major project: the question nobody could answer, and the decision your work changed. Read it to someone outside data and have them repeat it back
- Point your 'why this role' answer at something concrete in the job description or the product surface you would be working on, and keep it to two sentences
- Have two questions ready about measurement: which metric the team is held to, and who acts on an analysis once it lands
Technical Assessments
reportedMuch of what gets scored here happens out loud while you type. Nobody can see your reasoning inside a half-written query, so five silent minutes read as being stuck even when they are not. State the plan in plain language first: which tables, what grain you are aggregating to, and the one filter that defines the population. Then write it. The narration doubles as insurance, because a wrong plan gets caught early and cheaply while a wrong query gets caught at the end with no time left to redo it. A timed statistics section, where one exists, is a separate test with its own clock.
What to demonstrate
- Whether the query you write matches the plan you just described
- What you do with a hint, meaning whether the correction gets absorbed or the first approach gets defended
- Whether you can debug your own wrong output by reading the result set and naming which part of the query produced the anomaly
How to prepare
- Solve three problems while screen-sharing into a recording, then watch it back and mark every stretch longer than thirty seconds where you said nothing
- Practise compressing the plan into one sentence before typing, then check afterwards whether the finished query actually matched it
- Time yourself on statistics questions that carry a business reading, such as what a confidence interval does and does not claim, rather than re-reading notes without a clock
Panel Interview
reportedWhere a loop ends with a senior leader, that conversation is rarely another skills test. The technical signal already exists by then, so the questions tend to open up: what you would look at first, where a metric you have heard about could mislead, what you would push back on. The decision being made is scope, which in practice means level and how much you would be trusted to own unsupervised. Treating it as a formality is the usual mistake. An open question late in the day is still being scored, and a vague answer reads as someone who has not run anything themselves.
What to demonstrate
- Whether your view of the business has anything specific behind it, given that you are working only from what is public and are expected to say so
- Whether the scope of work you describe owning matches the scope of the role, instead of sitting a level below it
- Whether you can disagree with something concrete and stay useful about it, rather than agreeing with everything said in the room
- Whether your questions are ones only this person could answer, as opposed to ones the recruiter already covered
How to prepare
- Build one view you could defend for two minutes using only public information: what the funnel probably looks like, which metric likely drives decisions, and where that metric could mislead. Being wrong for a stated reason survives this round; having no view does not
- Write down the largest piece of work you have owned from question to decision, who else touched it, and what you decided alone, then check that it reads at the level you are interviewing for
- Prepare one thing you would want changed if you joined and phrase it as a question rather than a verdict, so it opens a conversation instead of closing one
PracHub editorial advice for the preparation topics above.
Consent and age rules silently truncate the data, not just the joins.
Where age_gated is TRUE, behavioural logging and cross-system joins are restricted, so those learners are missing from the very tables used to compute engagement. An analysis that simply inner-joins will produce a population skewed to older learners and self-serve accounts while appearing complete. Check the age_gated share of every population you report on, and state it.
Roster sync creates and deactivates accounts in bulk, and it is not behaviour.
An automated roster import at term start can create thousands of accounts in one hour and deactivate thousands more at term end. Signup, activation, and churn series computed without filtering enrollment_source IN ('roster_sync','admin_bulk') will show spikes and cliffs driven entirely by an integration job. It also breaks cohort retention: a roster-created account that never activates is a provisioning artefact, not a churned learner.
Reading a dozen metrics with no multiplicity control
Nominate one primary metric before launch and treat the rest as guardrails or exploratory, with Bonferroni or Benjamini-Hochberg applied when you intend to make claims from them. Twenty independent tests at 0.05 under the null produce at least one false positive about 64 percent of the time.
Interpreting a change before checking data quality and logging
Spend the first pass on row volume by day, null rates, duplicate keys, and whether the step change lands on a release or tracking-migration date. A discontinuity that coincides with a deploy is an instrumentation hypothesis before it is a behavioural one.
Choose a category, try a prompt, then open its approach, worked solution or follow-up when you need it.
Can you explain the difference between correlation and causation in th…
Can you explain the difference between correlation and causation in the context of observational studies?
Approach
- Quantify uncertainty explicitly rather than reporting a point estimate alone.
- Sanity-check the answer against a simple bound or a simulated case.
- Write down the assumption the method needs before you use the method.
Follow-up
- What sample size would you need to detect an effect half this size?
- Which assumption here is most likely to be violated in practice?
Simulate the gain a bottom-quartile selection manufactures
Observed quiz scores are a true score plus independent noise, both normal, with reliability r equal to true-score variance over observed-score variance. Simulate 200,000 learners, standardise the pretest to mean 0 and SD 1, select the bottom quartile on the pretest, and measure the mean pretest-to-posttest change with no intervention at all. Report the manufactured gain for r in {0.5, 0.7, 0.9, 1.0}, and give the closed form your simulation should reproduce. Use numpy only, no statistical packages.
Approach
- Parameterise by r directly: draw the true score with variance r and each noise term with variance 1 - r, so the observed score has unit variance and reliability exactly r by construction. Draw one true score and two independent noise terms per learner.
- Select on the pretest and only the pretest. Selecting on the true score, or on the average of the two measurements, removes the correlation between selection and the pretest's own noise, and the artefact disappears. That sensitivity is the lesson.
- Compute the mean of posttest minus pretest over the selected group. The true score cancels in the difference, so whatever remains is entirely the difference of two noise draws conditioned on the first being low.
- Check against the closed form. For jointly normal standardised scores, E[posttest | pretest] = r * pretest, so the expected gain is (1 - r) * |E[pretest | bottom quartile]|, and E[pretest | bottom quartile] = -phi(z_0.25) / 0.25 = -1.2711.
- Extend the script with a control group selected by the identical rule from an untreated population and show the difference of differences returns to zero. That is the design fix you would actually propose, not a caveat in a footnote.
Worked solution 25 min
- For each r: t = rng.normal(0, sqrt(r), n); e1, e2 = rng.normal(0, sqrt(1 - r), n) twice; pre = t + e1; post = t + e2.
- sel = pre <= np.quantile(pre, 0.25); gain = (post[sel] - pre[sel]).mean().
- Compare gain to (1 - r) * 1.2711 and print the absolute difference.
- Repeat the whole loop over the four r values and tabulate simulated against closed form.
- Add the untreated control arm selected by the same rule and report the difference of differences.
Follow-up
- A published case study reports large gains specifically for learners who started in the bottom quartile. What do you ask for before believing any of it?
- The posttest is twice as long as the pretest, so its reliability is higher. Does the manufactured gain grow or shrink, and why?
- Give a design that measures a real effect on exactly this selected population without a randomised control.
Attach the calibration in force without merge_asof
You have responses (response_id, content_item_id, submitted_at, is_correct) and calibration_history (content_item_id, valid_from, valid_to, irt_a, irt_b), where intervals within an item are non-overlapping and the open interval carries valid_to as NaT. Attach to each response the irt_a and irt_b in force at submitted_at. You may not use pandas.merge_asof and you may not apply row-wise. responses has about 5 million rows, calibration_history about 40 thousand. Return the input frame plus two columns, NaN where no interval covers the timestamp.
Approach
- Sort calibration_history by (content_item_id, valid_from) and factorise content_item_id across both frames into a shared integer code, so an unknown item on the response side is detectable immediately rather than joining to nothing.
- Convert both timestamps to int64 seconds and build one composite key per side, code * 2**32 + seconds. With codes well under two billion and seconds near 1.8e9, that stays inside int64 and makes a single global search possible. State the second-level resolution assumption; interval boundaries are set at second granularity, so nothing is lost.
- Call np.searchsorted(calibration_keys, response_keys, side='right') - 1 once over the whole array. That gives, for each response, the position of the latest interval starting at or before it within the same item, because the composite key orders by item first.
- Invalidate bad candidates in two places: index -1, and a candidate whose code does not match the response's code, which happens when a response precedes its item's first interval and lands on the previous item's last row.
- Invalidate a third case that is easy to miss: the candidate has a non-null valid_to and submitted_at is at or after it, meaning the response falls in a coverage gap. Without this the previous interval's parameters leak onto uncovered responses.
- Take the parameters positionally with np.take and write NaN where any invalidation fired.
Follow-up
- A backfill wrote overlapping intervals for 200 items. How do you detect that before the join rather than after, and what do you do with those responses?
- The join is correct but peak memory is unacceptable. What changes, and which part of the approach survives?
- How would you test this without a golden output to compare against?
How do you select the appropriate model for a binary outcome variable?
How do you select the appropriate model for a binary outcome variable?
Approach
- Handle the rows that do not match: a LEFT JOIN with a NULL check is usually the question.
- State the window function and its partition and ordering out loud before writing it.
- Say which table is the grain you start from, and join outward from it.
Follow-up
- What breaks if events arrive late or out of order?
- How would you verify this result without re-running the same query?
Write a query using SQL window functions to calculate a rolling averag…
Write a query using SQL window functions to calculate a rolling average of survey responses over the last 30 days.
Approach
- Handle the rows that do not match: a LEFT JOIN with a NULL check is usually the question.
- State the window function and its partition and ordering out loud before writing it.
- Check whether any join is one-to-many before aggregating, or the sums inflate.
Follow-up
- What breaks if events arrive late or out of order?
- How would you verify this result without re-running the same query?
Verified mastery events with a delayed retention check
From fct_assessment_response(response_id, learner_id, objective_id, content_item_id, attempt_no, submitted_at, scored_at, score_points, max_points, scoring_mode, mastery_state_after, mastery_prob_after), count verified mastery events for one calendar month. A mastery event is the first time a (learner_id, objective_id) pair reaches mastery_state_after = 'mastered'. It is verified when a later response on the same objective, submitted at least 7 days after the crossing, scores at or above 0.8 of max_points. Return verified events, eligible events, the verified rate, and the share of crossings with no eligible check.
Approach
- Find the first crossing with ROW_NUMBER() OVER (PARTITION BY learner_id, objective_id ORDER BY submitted_at, response_id) over rows where mastery_state_after = 'mastered', keeping rn = 1 across all time, then filtering that result to the month. Filtering to the month first would count a re-mastering after a lapse as a fresh event.
- Attach the check with a LATERAL or correlated subquery selecting the earliest response on the same (learner_id, objective_id) with submitted_at >= crossing_at + INTERVAL '7 days'. Take the earliest, not the best: best-of turns the metric into did they ever eventually pass, which only rises as time goes by.
- Score the check as score_points >= 0.8 * max_points, guarding max_points > 0 and requiring score_points IS NOT NULL. A rubric row still awaiting grading has scored_at IS NULL and must be excluded from eligibility rather than counted as a failure.
- Set the denominator to pairs with an eligible check, and report eligible / crossings separately. That coverage share is the honest caveat and it falls for objectives that sit at the end of a course.
- State the reporting lag the definition forces: the month cannot close until the last day's crossings have had 7 days plus the rubric scoring lag to produce a check.
Worked solution 35 min
- Build the all-time first-crossing CTE and assert its row count equals COUNT(DISTINCT (learner_id, objective_id)) within it.
- Filter crossings to the target month and count them.
- Add the LATERAL earliest-eligible-check join and count how many crossings get one.
- Compute verified_events, verified_rate and eligible_coverage, then re-run the whole thing with a 14-day delay to see how coverage responds.
Follow-up
- A learner who never returns to an objective has no eligible check. Excluding them assumes that missingness is unrelated to retention. How would you test that assumption with the data you have?
- If the engine team lowers the mastery threshold, this metric's numerator and denominator both move. Which guardrail catches that, and what would you watch on the item difficulty distribution?
- Objectives late in a course get fewer checks. How do you stop that from making late objectives look systematically better or worse?
How do you manage competing priorities when supporting multiple resear…
How do you manage competing priorities when supporting multiple research initiatives?
Approach
- Fix the population and the time window before naming any metric.
- Name one primary metric, then the guardrail that stops it being gamed.
- Decompose the metric into the rates that drive it, and say which one you would check first.
Follow-up
- What would you do if the primary metric and the guardrail moved in opposite directions?
- How would you detect that the metric is being gamed rather than genuinely improving?
A key performance metric for our research dashboard has suddenly dropp…
A key performance metric for our research dashboard has suddenly dropped by 10%. How do you investigate the root cause?
Approach
- State what result would change your recommendation, so the answer is falsifiable.
- Name one primary metric, then the guardrail that stops it being gamed.
- Fix the population and the time window before naming any metric.
Follow-up
- How would you detect that the metric is being gamed rather than genuinely improving?
- Which segment would you cut first, and what would that rule out?
How would you design a metric to measure the success of a new public h…
How would you design a metric to measure the success of a new public health data initiative?
Approach
- State what result would change your recommendation, so the answer is falsifiable.
- Decompose the metric into the rates that drive it, and say which one you would check first.
- Restate the decision this analysis has to support, and who acts on the answer.
Follow-up
- How would you detect that the metric is being gamed rather than genuinely improving?
- Which segment would you cut first, and what would that rule out?
If we introduce a new feature to our survey platform, how would you de…
If we introduce a new feature to our survey platform, how would you determine if it is successful?
Approach
- Restate the decision this analysis has to support, and who acts on the answer.
- Decompose the metric into the rates that drive it, and say which one you would check first.
- State what result would change your recommendation, so the answer is falsifiable.
Follow-up
- What would you do if the primary metric and the guardrail moved in opposite directions?
- Which segment would you cut first, and what would that rule out?
What are the trade-offs between a frequentist and Bayesian approach to…
What are the trade-offs between a frequentist and Bayesian approach to A/B testing in your projects?
Approach
- Name the guardrails that would stop a launch even on a positive primary result.
- Say whether units interfere with each other, and switch design if they do.
- Name the randomisation unit first; it decides the variance and what the test can detect.
Follow-up
- How would you handle interference between treated and control units?
- What would you conclude if the result is positive but the test is underpowered?
How do you ensure statistical significance when dealing with small sam…
How do you ensure statistical significance when dealing with small sample sizes common in social science research?
Approach
- Say whether units interfere with each other, and switch design if they do.
- State the primary metric and the minimum effect worth shipping, then size the test.
- Decide the analysis before seeing data, including how long it runs and when you look.
Follow-up
- What would you do if you could not randomise at all?
- What would you conclude if the result is positive but the test is underpowered?
Separate a novelty spike from a durable engagement gain
A section-randomised test of a redesigned practice interface ran four calendar weeks. Weekly active learning minutes per active learner came in at +18%, +9%, +4%, then +2%. Sections entered the test as their instructors opted in, so exposure start dates are spread across three of those weeks. Using fct_lesson_activity, dim_term and the assignment log, decide whether to ship. Say what you would recompute on exposure time rather than calendar time, and what evidence would let you treat the week-four number as the steady-state effect.
Approach
- Re-index every section to weeks-since-first-exposure. The calendar series mixes sections in their first exposed week with sections in their fourth, so part of the apparent decay is a composition change rather than decay.
- Plot the effect by exposure week with section-clustered intervals. Novelty decays monotonically toward an asymptote; primacy starts negative and recovers as learners relearn a changed interface. The two have opposite shapes and opposite ship decisions.
- Read the new-learner stratum on its own. A learner whose first_activity_at falls after their section's exposure date has no prior interface to be surprised by or to relearn, so their effect estimates the steady state free of both artefacts.
- Check whether calendar week one is term-week one in dim_term. Term start moves minutes far more than any interface change, and opt-in timing correlates with the instructor's calendar, so the +18% may be partly the academic calendar.
- Decide on the exposure-week-four interval, not the point estimate. A test sized to detect +18% will usually be unable to resolve +2%, and saying so is the answer.
Worked solution 20 min
- Join the assignment log to fct_lesson_activity and derive exposure_week = floor((activity_date - section_exposure_date) / 7).
- Recompute the treatment effect on weekly minutes per active learner within each exposure_week bucket, with section-clustered intervals.
- Split the exposure-week-four estimate by whether the learner's first_activity_at precedes or follows their section's exposure date.
- Overlay the dim_term term-week index to confirm calendar week one is not term-week one for most sections.
- State the ship decision against the exposure-week-four interval and its width.
Follow-up
- Exposure week four reads +2% with a 95% interval of [-1.5%, +5.5%]. What do you ship, and what do you run next?
- How long a holdback would you keep, and what would you measure on it that a four-week test cannot answer?
- Minutes rose but weekly verified mastery events per active learner did not move. Which metric do you act on, and why?
Activation flatlines across four hundred accounts in one term
Seat activation for the current term reads 0.12 across 400 institutional accounts against 0.58 last term, and dim_learner.first_activity_at is NULL for most learners created this term. Tables: dim_learner (learner_id, org_id, created_at, first_activity_at, age_gated) and fct_lesson_activity partitioned on activity_date (activity_id, learner_id, started_at, device_type, session_id). Before anyone contacts a single account, establish whether learners stopped showing up or rows stopped arriving. Deliverable: the ordered checks and the specific evidence each one produces.
Approach
- Separate absence of behaviour from absence of data as the first move: count rows and distinct learner_id in fct_lesson_activity per activity_date for sixty days, and compare against the same term-week a year earlier and against the trailing median. A collapse in loaded rows is a load question; normal row volume with fewer distinct learners is a behaviour question.
- For each suspect partition compute MIN and MAX started_at within it. A partition whose max started_at truncates mid-afternoon UTC while its neighbours run to end of day is an interrupted load, and that signature is not something learner behaviour produces.
- Establish how first_activity_at is maintained. If the deriving job is incremental and reads only the current partition, learners whose first row landed in a failed partition keep first_activity_at NULL permanently; re-running today fixes nothing and only a backfill plus recompute will.
- Cut the deficit by org_id and device_type to localise the fault. A client SDK or single ingestion shard failure hits a subset of orgs or one device_type; a warehouse job failure hits every org on identical dates with the same shape.
- Check the age_gated share of the affected population before concluding, since consent rules restrict behavioural logging for those learners and their absence from fct_lesson_activity is by design rather than a defect.
Follow-up
- What monitor would have caught this before the activation metric did, and what threshold would you set on it?
- The partitions are backfilled, but a downstream renewal-risk model already scored these orgs on the broken data. What do you do about the scores it produced?
Roughly 90 minutes a night on weekdays with one longer weekend block. The plan deliberately cuts scope rather than compressing everything, on the assumption that finishing one thing a night beats half-starting four.
Prepare, practise & reflect
One practical outcome each day. Spend longer where you need it.
0 / 7 done01Fix the scope and set a baseline
- Read the role description and write the three things the loop will almost certainly test, then write an explicit not-doing list for everything else and keep it visible all week.
- Take one 20-minute SQL prompt and one 10-minute metric question cold, and write the single sentence that says what blocked each attempt, since that sentence is what decides which two topics get the most evenings.
- Set the week's one rule: one problem finished to completion every night, including the night you only have 40 minutes.
Deliverable: A one-page scope with an explicit not-doing list and two cold attempts, each carrying one sentence on what blocked it.
Practice prompt ↗Practice prompt ↗Practice prompt ↗Worked solution ↗02One query pattern, written three times
- Choose the single pattern most likely to appear (a cohort retention grid, or a funnel counted by user) and write it three times from a blank file rather than editing the previous attempt.
- On the third attempt, write the grain of every CTE as a comment before writing its body.
- Stop at 90 minutes even if the third version is imperfect, and write the one thing you would fix with another hour.
Deliverable: Three independent versions of the same query plus a note on what changed between them.
Practice prompt ↗Practice prompt ↗Practice prompt ↗03Only the statistics you will be asked to defend
- Write, in under 200 words, how you would decide whether a difference between two groups is real: the test, its assumptions, and what you would switch to when an assumption fails.
- Compute a 95 percent confidence interval for a difference in proportions by hand on realistic numbers, then write in one sentence what changes if the two samples are paired rather than independent.
- Write your answer to "what does a p-value mean", check it against a definition, and delete the version that describes it as the probability the hypothesis is true.
Deliverable: A 200-word written answer and one hand-computed interval you can reproduce under pressure.
Practice prompt ↗Practice prompt ↗Practice prompt ↗04One case, and the assumptions holding it up
- Answer one product case aloud in 20 minutes with a recording running, then listen back with a pen and mark every claim you asserted without saying what it rested on: an assumed user behaviour, an assumed data source, an assumed baseline rate, an assumed grain.
- Pick the three assumptions the recommendation actually depends on, write how you would check each one against data, and say which one being wrong would flip the recommendation rather than merely weaken it.
- Write the four-step structure you used onto a card small enough to hold in working memory when you are nervous.
Deliverable: One recording, three load-bearing assumptions each with a written check, and a four-step structure card.
Practice prompt ↗Practice prompt ↗Worked solution ↗05Your own work, timed
- Write a 90-second version and a four-minute version of your main project, and time both out loud rather than reading them.
- Prepare answers to the two follow-ups that always come: what you would do differently, and how you knew it worked.
- Put one number in the first sentence and be able to say exactly where that number came from and what it excludes.
Deliverable: Two timed narratives with one defensible number in the opening line.
Practice prompt ↗Practice prompt ↗06The one full rehearsal, in a longer weekend block
- Run a 60-minute mock covering query work, a case and a behavioural question in a single sitting with no breaks, because sustained attention is the thing evenings have not trained.
- Immediately afterwards, and before hearing any feedback, write the three moments you lost the thread.
- Spend the rest of the block only on those three moments, and on nothing you merely feel shaky about.
Deliverable: Mock notes naming three failure moments with a specific fix written under each.
Practice prompt ↗Practice prompt ↗07Taper
- Write the 20-minute warm-up you will actually do on the morning of the interview: one query you can already write from a blank file, one metric you can define out loud, and nothing you have never seen before.
- Re-read only your own notes from this week, and open no new material.
- Write down the logistics: the tool you will be asked to work in, whether lookups are allowed, and the sentence you will use when you do not know something.
Deliverable: A one-page card holding the case structure, the project numbers, and the logistics.
Practice prompt ↗Practice prompt ↗Worked solution ↗Expand any day for tasks and deliverables. Your progress is saved on this device.
Nearly every data role forces a trade between the analysis you want and the one that fits the decision window. Prepare a case where you deliberately shipped something less rigorous, named the weakness to the person relying on it, and said what would change your answer. The naming is the part interviewers listen for.
Describe a situation where you disagreed with a team member’s methodol…
Describe a situation where you disagreed with a team member’s methodology; how did you resolve it?
Approach
- Quantify the outcome, including what you would not claim credit for.
- State the situation in two sentences and spend the rest on your reasoning.
- Name the disagreement or constraint, and how you resolved it with evidence.
Follow-up
- What did you decide not to do, and why?
- What would you do differently if you ran that project again?
State the impact of your last project without inflating it
Pick one project from the past year and account for its impact as if the listener could audit every number. Cover the metric you claim to have moved and its exact definition, how much of the movement is attributable to your work, what else was changing at the same time, what the counterfactual was and where it came from, and what you would have needed to measure at the start to make the claim clean. If the effect is not separable, say so and say what you would do differently. Probed: attribution discipline applied to your own work.
Approach
- Name the metric by its definition rather than its label. 'Verified mastery rate' means nothing without its numerator, denominator, window, and the reporting lag the delayed retention check forces.
- Separate the three claims usually merged into one: the metric moved, your work moved it, and the movement was worth what it cost.
- State the counterfactual explicitly and say what produced it: a holdout, matched sections, or a pre-period trend. If it was a before-and-after with no control, say that in those words.
- List the co-occurring changes with dates: a term boundary, a roster sync at scale, a pricing change, another team shipping into the same surface. In this domain the academic calendar alone routinely dwarfs a product effect.
- Convert the shortfall into a specific design artefact: the holdout you would have reserved, the instrumentation you would have shipped before the change, the analysis plan you would have written first.
Follow-up
- What fraction of the observed movement would you defend under cross-examination, and on what evidence?
- Your project shipped in term-week one. How does that change what you can claim?
- If the effect is genuinely not separable, was the project worth doing?
Scope a vague request to investigate falling engagement
A curriculum director messages you: engagement is down in middle school, can you dig into it. You have dim_learner, fct_lesson_activity, fct_assessment_response, fct_enrollment and dim_term. Before writing any query you get one fifteen-minute conversation. Deliverable: the questions you ask in that conversation, the written scope you send afterwards, and the first three things you compute, with the reason each comes before the others. Probed: whether you convert an ambiguous request into a decision and a single metric before touching data, rather than after.
Approach
- Find the decision behind the request first: what the director would do differently depending on the answer, and by when. That sets how much of this analysis is worth doing at all.
- Pin the four ambiguities that make 'engagement is down' uncomputable: which population (grade_band = '6_8', which orgs), which metric (active_seconds, activity days, scored submissions), which comparison window, and against which baseline.
- Ask where the belief came from. 'Down' is almost always a chart someone saw, and reproducing that exact chart is faster than building a new one and then arguing about why the two differ.
- Check the calendar before believing the shape at all: a decline spanning a term boundary is the default explanation in this domain, so term-week alignment through dim_term is the first computation, not a later refinement.
- Send the scope back in writing with an explicit out-of-scope list and a time box, so follow-on questions arrive as new requests instead of as scope creep.
Follow-up
- The director cannot name the decision this feeds. Does that change whether you do the work, and how much of it?
- Your term-aligned version shows no decline at all. What do you send, and to whom first?
- 01
Describe a situation where you disagreed with a team member’s methodology; how did you resolve it?
- 02
Pick one project from the past year and account for its impact as if the listener could audit every number. Cover the metric you claim to have moved and its exact definition, how much of the movement is attributable to your work, what else was changing at the same time, what the counterfactual was and where it came from, and what you would have needed to measure at the start to make the claim clean. If the effect is not separable, say so and say what you would do differently. Probed: attribution discipline applied to your own work.
- 03
A curriculum director messages you: engagement is down in middle school, can you dig into it. You have dim_learner, fct_lesson_activity, fct_assessment_response, fct_enrollment and dim_term. Before writing any query you get one fifteen-minute conversation. Deliverable: the questions you ask in that conversation, the written scope you send afterwards, and the first three things you compute, with the reason each comes before the others. Probed: whether you convert an ambiguous request into a decision and a single metric before touching data, rather than after.
Is this an official NORC at the University of Chicago interview guide?
No. It is PracHub's own research and practice material for the Data Scientist role at NORC at the University of Chicago. Rounds and questions reflect what candidates have reported, not a process NORC at the University of Chicago has published, and they change over time. Confirm the current format and scope with your recruiter.
PracHub interview research ↗How long does the interview process typically take?
The timeline from initial application to offer generally spans 4 to 8 weeks, depending on the volume of applicants and the specific team's hiring urgency.
PracHub interview research ↗Is there a coding assessment?
Yes, you should expect a technical evaluation of your SQL and programming skills, often conducted via a live coding session or a take-home assignment.
PracHub interview research ↗What is the work environment like?
NORC at the University of Chicago maintains a professional, collaborative environment that values intellectual rigor and the pursuit of objective truth in data.
PracHub interview research ↗How can I best prepare for the behavioral rounds?
Use the STAR method (Situation, Task, Action, Result) to structure your answers, and ensure your examples highlight your collaborative spirit and your commitment to data ethics.
PracHub interview research ↗Sources & methodology 3 sources ↗
No official company page is cited. Rounds and questions come from candidate reports and PracHub editorial material; each source shows the date it was read.
- 01PracHub interview research ↗
PracHub editorial research into this company and role, maintained with this guide. Candidate-reported, not an employer publication.
PracHub page · Accessed 2026-09-22 - 02PracHub Data Scientist practice ↗
Cross-company practice questions for this role.
PracHub page · Accessed 2026-09-22 - 03PracHub interview preparation framework ↗
The framework the preparation plan follows.
PracHub page · Accessed 2026-09-22
Explore related engineering environments and compare the preparation focus.
Yale University Data Scientist Interview Questions & Guide 2026
Timed practice for Yale University data scientist candidates who freeze under a clock: short SQL and probability sets run end to end, then reviewed against a written rubric.
Universitätsklinikum Tübingen Data Scientist Interview Questions & Guide 2026
A prep plan for Universitätsklinikum Tübingen data scientist candidates built on spaced review: write a full answer before opening any solution, keep a running list of the assumptions you failed to state, and repeat every missed exercise th
University of Michigan Data Scientist Interview Questions & Guide 2026
Metric-diagnosis practice for University of Michigan data scientist candidates: a number moves week over week and you work out why, splitting by segment, ruling out a logging or release change, and separating a mix shift from a real change
University of Florida Data Scientist Interview Questions & Guide 2026
Causal-inference exercises for University of Florida data scientist prep: difference-in-differences, matching, instrumental variables, and knowing when observational data simply cannot answer the question asked.