Skip to main content

Atlas Blog

Confounding Variable: Spot It Before Trusting a Causal Claim

Learn what a confounding variable is with a simple example, then check how published studies report adjustment, limitations, and uncertain causal claims.

Constant Comparative Method: Compare Cases and Revise Categories

Constant comparative method, explained with a worked example: compare new incidents with earlier cases, check contradictions, and document category revisions.

Contribution Analysis: Test a Causal Story Against Evidence

Learn contribution analysis through a worked program example. Test each link in a theory of change, weigh rival causes, and write a bounded final finding.

Convergent Mixed Methods Design With a Worked Joint Display

Plan a convergent mixed methods design, merge separately analyzed findings in a joint display, and explain agreement, difference, and unresolved results.

Evaluation Plan: A Question-to-Evidence Template

Build an evaluation plan that links decisions, questions, criteria, data sources, methods, timing, and reporting. See a worked question-to-evidence matrix.

Evaluation Questions: How to Choose and Test a Useful Set

Write evaluation questions tied to a program and its users. Use a worked question-rationale matrix to check purpose, evidence gaps, and who approves the set.

Exit Interview Analysis: Trace Reasons to Evidence

Analyze exit interviews by coding reasons, checking contrary accounts, and building a source-linked evidence table before recommending a retention fix.

External Validity: Test Whether a Finding Fits Your Setting

External validity asks whether a study finding can apply beyond its sample. Use a source-backed matrix to compare populations, settings, outcomes, and time.

Framework Analysis: Build a Case-by-Theme Matrix

Learn the five stages of framework analysis, build a case-by-theme matrix, and check every summary against its source transcript before drawing conclusions.