[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: RE: mlogit and IV? polychotomous logistic model and endogenous explanatory variable
The OED works on descriptive, not prescriptive, principles,
so does not even purport to judge on correctness.
> If it is good enough for the Oxford English Dictionary,
> it is good enough for me:
> Divided, or involving division, into many
> (or more than two) parts, sections, groups, or branches:
> = POLYTOMOUS
> So polychotomy, division into more than two
> parts or groups, as in classification: = POLYTOMY
> *1858* MAYNE
> /Expos. Lex./, /Polychotomus/, applied to a body that is
> divided into numerous articulations..: polychotomous. *
> 1887* /Amer. Naturalist/ Oct. 915 Polychotomy is probably never more
> than provisional, and all classification will eventually be
> So until we eventually reach the dichotomy where some of us
> are right and some of us are wrong, let's allow polychotomous.
> Nick Cox wrote:
> >My only advice is marginal to your main question.
> >The term "polychotomous", although common in the
> >literature, is malformed and based on a misparsing
> >of the word "dichotomous", whose elements
> >are "dicho" and "tomous". The term "polytomous",
> >also common in the literature, is more nearly correct.
> >Help stamp out this linguistic monstrosity!
> >N.B. this is a different kind of argument from
> >those in favour of "heteroskedasticity" rather than
> >"heteroscedasticity". In the latter case, there are
> >plenty of precedents for rendering the Greek letter
> >kappa into the English letter c, so one could be
> >sceptical about that argument.
> >"polychotomous" just got into the literature because someone
> >didn't understand the etymology of "dichotomy" and other people
> >copied that mistake. It's still wrong.
* For searches and help try: