January 2, 2015

Maybe not a representative sample

The Dominion Post asked motorists why they thought the road toll had climbed, and what should be done about it.


Interestingly, three of the five(middle-aged, white, male ,Wellington area) motorists attributed it to random variation. That’s actually possible: the evidence for a real change in risk nationally is pretty modest (and the Wellington region toll is down on last year).

(via @anderschri5 on Twitter)


Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »


  • avatar

    I would love to know the wording of the question that was posed. I find it remarkable that 4 of the 5 men included the words “statistical”, “variation”, or “random” in their answers.

    3 years ago

    • avatar
      Thomas Lumley

      I would have thought, based on the headline, that they asked “Why do you think the road toll has climbed?” I agree it would be good to know for sure.

      3 years ago

  • avatar
    Frank Collette

    Non measurable occult factors like earth magnetism or what dowsers often call geopathy, may have some effect.
    I would put some fatalities down to All Black, Black Cap and Rugby League Kiwi’s dominance creating a ten foot tall syndrome among some kiwis.
    In addition to your cancer article, and cancer’s generally increasing, cell coverage intensification, wifi proliferation and general electromagnetic stress increase’s, many also be contributing to negative health and well being statistics in ways that are not yet apparent or clear.
    Traditional chinese medicine [TCM] states: 2014 was a year of conflicts and misunderstandings due to the dominant presence of the fire element.
    Which may also be a contributing factor.

    3 years ago

  • avatar
    Frank Collette

    Comets [most not all] have traditionally been associated with bad luck and accidents.
    Moons and lunacy have very strong correlation, possibly causation for some.
    Drownings for this holiday period have also nearly doubled from last year, when the non specified Kariotahi Beach surfers drowning is included. At present the cause of death is not specified.

    Much of the variance within statistics has reasons beyond the grasp of modern empirical thinking.

    Look for ancient [or modern alternative]reasonings to explain baffling variance.

    Alternative archaeology places vikings in america before columbus, and early celtic and possibly chaldean evidence in some ancient nz sites.

    Things are not always as they seem.

    There are more things in heaven and earth, Horatio, Than are dreamt of in your philosophy. – Hamlet (1.5.167-8)

    3 years ago