# A Success Collection Generates Hypotheses Before It Tests Causes ## Claim A volume assembled from cases chosen because they succeeded can suggest what to investigate but cannot show what caused the success, because the selection already fixed the outcome. <!-- atomicity: 1 idea ✓ --> ## Scope & conditions - The defect is selection on the outcome variable. Without the churches that adopted the same model and shrank or closed, the comparison that would license a causal claim does not exist in the book. - Compounded when contributors write their own chapters: the reporter, the participant, and the beneficiary are the same person, and no failure case has an author. - Not a reason to discard the cases. Hypothesis generation is real work — the error is treating the collection as though it had already done the testing. - Generalizes directly to secular genres built the same way: customer success-story pages, "how we scaled" conference talks, curated case-study decks, best-practice roundups. ## Evidence *The New Apostolic Churches* collects chapters by pastors of thriving churches describing their own models. The review's PN-04 names the structure: a success collection generates hypotheses before it tests causes. > [!quote] Wagner, *The New Apostolic Churches*, p. 20 / P-0022 > "We are seeing a transition from bureaucratic authority to personal authority" That framing is the book's thesis, and the volume's evidence for it is the assembled successes — which is exactly the inference the selection cannot support. ## Links - [[3. Resources/MOCs/NAR and Charismatic Authority Critique MOC]] — the methodological reason the movement's showcase volume cannot function as the evidence base it is usually cited as - [[3. Resources/Permanent Notes/Institutional Success Is Not Evidence of Moral Legitimacy]] — the sibling error on a different axis: that note says growth doesn't establish trustworthiness, this one says a curated set of growers doesn't establish what produced the growth - [[3. Resources/Permanent Notes/An Unaudited Felt-Effectiveness Standard Quietly Replaces a Groups Stated Standard]] — extends: with no failure cases in view, felt effectiveness is the only standard left operating - [[3. Resources/Permanent Notes/Separate the Datum From Its Explanation When Evaluating Claims]] — applies-in: each chapter's growth is the datum; the model credited for it is the explanation, supplied by the beneficiary - [[3. Resources/Permanent Notes/Adoption and Satisfaction Are Not Evidence That a Teaching Method Works]] — the same failure in instructional design, reached from an unrelated source; two independent arrivals at one inference rule ## Tests / falsifiers - Locate churches that adopted the same apostolic model and declined or dissolved. Their absence from the literature is the prediction this claim makes; finding a published set of them would show the movement did do the comparison. - If contributor chapters routinely reported their own failed campuses alongside successes, the selection problem would be much weaker than stated. ## Source - [[4. Archives/Book reviews/The New Apostolic Churches#PROMPT 10 — Notes pipeline|The New Apostolic Churches — full review]], PN-04 (added to vault 2026-09-04). Do not attribute the methodological critique to Wagner or the contributors.