Filtered dossier

base rates

  • 4 / 4
  • No visual on file
    Statistical Illusions Entry #0428

    Why a ninety-nine per cent accurate test is usually wrong

    A test that is right ninety-nine times in a hundred can still be wrong about most of the cases it flags, and simple arithmetic shows why.

    Novice
    The hidden part #0428

    Accuracy is a property of the test. Whether an alert is right also depends on how rare the thing is.

    Open file
  • No visual on file
    Statistical Illusions Entry #0442

    Why a cluster is expected in random data

    Randomness produces clumps. A pattern spread out evenly is the unlikely arrangement, so finding a cluster is not yet finding a cause.

    Adept
    The hidden part #0442

    Independent events arrive in bursts, so a cluster is the expected texture rather than a signal. Evenness is what would need explaining.

    Open file
  • No visual on file
    Framing Effects Entry #0519

    Why a comparison class is chosen before a claim

    Best in class, safest, fastest growing. Each is a true statement about a set, and the set was selected after the numbers were known.

    Adept
    The hidden part #0519

    Choosing the comparison class is not a step before the analysis. It is the analysis, and it can be run backwards from the answer.

    Open file
  • No visual on file
    Attention & Memory Entry #0198

    Why you remember the exception and forget the rule

    Memory does not store what happens most often. It stores what stood apart from its neighbours, which is why the archive over-represents the unusual.

    Adept
    The hidden part #0198

    Distinctiveness is a contrast, not a property. Memory keeps whatever failed to resemble its neighbours, then lets you estimate frequency from it.

    Open file