Explain

W08 · 5 h 30 min

From observation to identification

Essential question

Can the change be attributed to the intervention?

What you should be able to do

  1. Define an estimand.
  2. Draw a causal diagram and identify open paths.
  3. Distinguish design, adjustment, and untestable assumptions.
PrerequisitesW07 metric card.Bridge note on potential outcomes, controls, and randomization.
Builds on

W07 metric card. · Bridge note on potential outcomes, controls, and randomization.

Investigates

Can the change be attributed to the intervention?

Feeds

Pre-analysis plan clinic feeding L05/L06.

Arrive with a prepared artifact

Reading route
Core PAPER-10/PAPER-23; audit PAPER-32 for benchmark-size, test-set selection, and run-uncertainty gaps; inspect V03 as a design case.
Viewing route
Open the complete text-first lecture packageEight-minute estimand and DAG walkthrough. The current equivalent is notes, slide script, worked case, and no-video transcript; no recording is claimed.
Readiness check
Write treatment, control, unit, outcome, and target population for one proposed test.
Bring
Draft pre-analysis plan without outcome data.

Concepts, assumptions, and boundary

An effect requires a comparison world

An intervention effect is a contrast between potential outcomes under treatment and control for a defined unit and population. Before seeing outcomes, the analyst states the estimand, treatment assignment, control condition, observation schedule, exclusion rules, and analysis. Simple before/after changes are vulnerable to platform releases, query drift, seasonality, concurrent edits, and regression to the mean. Randomization, blocking, matched controls, staggered timing, or interrupted-series designs can reduce particular threats but cannot make every assumption observable.

Validity is a boundary, not a badge

A causal diagram makes assumed relationships and adjustment choices visible. Interference matters when one page or source ecosystem affects another unit. Sequential peeking inflates false-positive risk unless stopping rules are planned. Power depends on variance, clustering, treatment size, and feasible sample size—not just the number of screenshots. Even a strong internal design may generalize only to the tested systems, queries, dates, locales, and interventions.

Inspect the mechanism or evidence structure

Conceptual modelCausal diagram and validity boundary
Causal diagram and validity boundaryDirected paths identify confounding, mediation, and drift. Adjustment candidates and unobserved variables use different line styles. A boundary box states exactly where the estimand applies.InterventionContent qualityQuery mixPlatform stateExposureOutcome
Figure design. A DAG shows intervention, content quality, query mix, platform state, exposure, and outcome; a shaded frame encloses the tested population and period. The graph records assumptions; it does not prove them.
Long description

Directed paths identify confounding, mediation, and drift. Adjustment candidates and unobserved variables use different line styles. A boundary box states exactly where the estimand applies.

One action, one feedback state

Action

Toggle confounding and interference paths, then choose a design response.

Feedback

Feedback distinguishes blocking, randomization, measurement, adjustment, sensitivity analysis, and assumptions that remain untestable.

Accessible alternative

A decision table presents every graph state and defensible response.

Open the repeated-measurement explorer

Produce a reviewable intermediate file

Task
Critique two field designs and register a bounded intervention plan.
Inputs
DAG template, pre-analysis schema, synthetic design cases, power worksheet.
Timebox
75 minutes
Intermediate file
preanalysis-v1.md and dag-v1.svg
Stop condition
No outcome inspection before hypotheses, exclusions, stop rules, and primary analysis are frozen.

Why each controlled source is here

Complete, retrieve, and revise

Checkpoint
Pre-analysis plan clinic feeding L05/L06.
Reflection
Name the strongest remaining threat to identification.
Revision
Add a deviation log and explicit external-validity boundary.
Low-compute route
Use the supplied cases, DAG template, and analytic power table; no live experiment is required.

Five retrieval questions

01What is an estimand?

The precisely defined target contrast, including treatment, control, unit, outcome, population, and time.

02Why can before/after be misleading?

Time-varying platform state, query mix, concurrent edits, seasonality, and regression can explain the change.

03What does a DAG not prove?

It records causal assumptions and implications; it does not validate those assumptions by itself.

04What is the evidence boundary for W08?

PAPER-32 and vendor cases are used to audit reporting and design, not to admit unsupported headline effects into the course.

05What must you submit or revise after this week?

Pre-analysis plan clinic feeding L05/L06. Add a deviation log and explicit external-validity boundary.

Continue in the practice package