Time Series Cross Section Analysis

Compare many units across many moments so change over time is not confused with stable differences between units.

Solution archetype #

1080

Problem family

Uncertainty, Evidence & Inference Failure

Problem subfamily

Temporal Process, Nonstationarity & Trend Inference

The Diagnostic Story

Symptom: Before-and-after claims ignore that comparable units changed in the same direction at the same time. Cross-sectional rankings treat historically advantaged units as if they started from the same baseline. An apparent program effect may be a shared trend, a seasonal pattern, a cohort shift, or a measurement revision. When analysts pool all observations together, they erase the difference between within-unit change and between-unit variation.

Pivot: Construct an explicit unit-time comparison frame before interpreting any difference or change. Define comparable units, repeated observation periods, outcomes, exposures, baselines, shared time contexts, missingness rules, and comparison contrasts. Keep within-unit change and between-unit difference as distinct analytical dimensions throughout.

Resolution: Temporal trend, cross-sectional difference, and differential change are separated rather than conflated. Claims about policy, program, or organizational change carry clearer comparative grounding. Uncertainty about causal, statistical, and interpretive limits is named rather than concealed inside a tidy number.

Reach for this when you hear…

[public health researcher] “Every city improved after the intervention, but so did every city that did not get it — we cannot claim credit for a trend that was already underway everywhere.”

[corporate strategy analyst] “The ranking puts us third but it compares us to companies that entered the market ten years earlier — we need a within-cohort comparison, not a raw cross-section.”

[education policy evaluator] “Test scores went up this year, but every district's scores went up this year — I need to know whether ours improved more than comparable districts, not just more than last year.”

When This Archetype Applies

Partial catalog groundingSome structural conditions are represented by existing abstractions, but no sufficient condition set is fully represented.

Diagnostic problem

A comparison must explain change or difference across cases, but the evidence is vulnerable to conflating temporal trends, stable unit characteristics, shared period shocks, measurement changes, and true intervention or exposure effects.

Show the applicability expression

Applicability expression5 distinct conditions

Repeated multi-unit observationsandRelative unit changeandStable unit confoundingandShared temporal shockandAggregate-hidden trajectories

Algebraic1∧2∧3∧4∧5

groundedpartly groundedopen

5 conditions, all required.

5Required in every casenumbered 1–5

These hold no matter which pattern applies.

1

Repeated multi-unit observations · grounded · any one of 2

Multiple cases, units, places, organizations, cohorts, assets, or populations can be observed at repeated moments. more

and — also required

2

Relative unit change · open

The question asks whether one unit changed differently from others, not merely whether it changed at all. more

and — also required

3

Stable unit confounding · open

Cross-sectional differences may be caused by stable unit traits rather than the exposure under study. more

and — also required

4

Shared temporal shock · grounded · any one of 2

A shared time shock, cycle, season, policy environment, or historical event could affect all units at once. more

and — also required

5

Aggregate-hidden trajectories · open

A trend visible in aggregate could hide diverging trajectories among subgroups or cases. more

Other requirements and context (2)

Why these sit outside the expression

Deployment constraint — it constrains how the intervention must be deployed, not the situation that calls for it.

Goal — a goal states an intended outcome or evaluation criterion, not a pre-existing situation that independently summons the archetype.

• Deployment constraintData availability is uneven across units or periods and the design must state what comparisons remain valid. more

• GoalDecision makers need a comparative basis for claims about policy, performance, intervention, risk, or trajectory.

Coverage

2 of 5 conditions grounded · 3 open.

Read the methodology · Download the trigger-logic data

Mechanisms / Implementations

• Balanced-Panel Completeness Check: Assesses whether units have observations across required periods and where missingness threatens comparison.
• Difference-in-Differences Design: Compares differential before-after change between exposed and comparison units.
• Event-Study Panel Plot: Shows trajectories around an event, treatment, or adoption time.
• Fixed-Effects Panel Model: Controls for stable unit effects and/or shared time effects in repeated unit-time data.
• Lagged Panel Regression: Models delayed relationships between exposures and outcomes across units and periods.
• Peer-Trajectory Benchmarking: Compares a focal unit to selected peers across shared time windows.
• Unit-Time Dashboard: Displays repeated observations by unit and period while retaining trajectory context.

Related Abstractions

Abstractions this archetype builds on — directly (a source ingredient) or as a related pattern. Links follow the typed catalog namespace.

Built directly on (1)

• Comparative Method: Systematically juxtaposing selected cases so that their similarities and differences do the causal-inference work that controlled experiments cannot.

Also references 22 related abstractions

• Aggregation: Deliberately collapsing many items into a single summary, choosing which information to discard to gain tractability.
• Blocking (In Experimental Design): Group similar units.
• Causality: Cause-effect relationships.
• Confounding: Hidden variable interference.
• Counterfactual Reasoning: Hypothetical alternatives.
• Counterfactuals: Alternate hypothetical scenarios.
• Data Integrity: Accuracy and consistency preserved.
• Decision: Committing to one alternative from a set under uncertainty and trade-off, collapsing open deliberation into a chosen path and foreclosing the others.
• Decomposition: Breaking a whole into parts that can be analyzed independently and recombined to reconstitute the whole, making complexity tractable through divide-and-conquer.
• Effect Size: Magnitude of effect.

▸ Show 12 more

Variants

Narrower or domain-specific specializations that share this archetype's core structure. Recognized variants are established; candidate variants are provisional.

Descriptive Panel Benchmarking · domain variant · recognized

Track comparable units over time to show relative trajectories without making strong causal claims.

Differential Change Evaluation · implementation variant · recognized

Use the unit-time frame to ask whether one group or unit changed differently from another over the same period.

Staggered Adoption Panel Analysis · temporal variant · candidate

Exploit variation in when units adopt a change to compare trajectories before, during, and after adoption.

Editorial Notes

Problem Classification

Classification: Uncertainty, Evidence & Inference Failure → Temporal Process, Nonstationarity & Trend Inference

Problem kernel: temporal change is confounded with stable unit differences

Rationale: Earliest causal condition: A comparison must explain change or difference across cases, but the evidence is vulnerable to conflating temporal trends, stable unit characteristics, shared period shocks, measurement changes, and true intervention or exposure effects.

Independent corroboration: The earliest necessary condition in the frozen evidence is: A comparison must explain change or difference across cases, but the evidence is vulnerable to conflating temporal trends, stable unit characteristics, shared period shocks, measurement changes, and true intervention or exposure effects. That is a temporal process nonstationarity and trend inference problem because Historical and sequential evidence is treated as stable, deterministic, or self-explanatory despite drift, dependence, trends, survival conditioning, and time ordering.

Review outcome: Independent reviewer agreement; high confidence.