Apropos confusing quantile language: how about this chart from last week's
Economist:
http://www.economist.com/world/britain/displaystory.cfm?story_id=12480828&CFID=28295143&CFTOKEN=53908777
The right hand chart plots "decile averages" and the median!?
maja.
2008/10/14 John Sorkin <[log in to unmask]>
> Doug,
> I am not sure what you are advocating. It appears that you don't think
> quartile is jargon and you don't think the term should be shunned, but it
> also appears that you don't like the usage upper quartile or upper quintile.
> If so, we are in agreement. Use of quartile to describe the point separating
> one quarter of a distribution from the next is, I believe quite correct.
> However I believe that referring to the upper 10% of a distribution as upper
> decile is jargon and should be avoided.
>
> Do we agree?
> John
>
> John David Sorkin M.D., Ph.D.
> Chief, Biostatistics and Informatics
> University of Maryland School of Medicine Division of Gerontology
> Baltimore VA Medical Center
> 10 North Greene Street
> GRECC (BT/18/GR)
> Baltimore, MD 21201-1524
> (Phone) 410-605-7119
> (Fax) 410-605-7913 (Please call phone number above prior to faxing)
>
> >>> Doug Altman <[log in to unmask]> 10/14/2008 5:05 PM >>>
> I disagree.
>
> Of course we should choose language according to the audience but
> redefining terms is not the solution. Nor is accepting the wrong
> usage instead of the technical (ie correct) definition just because
> it is common in the wider world. We don't do that, for example, with
> "random" to take one key example, and nor should we with quartile (etc).
>
> I know that this usage is common in my field of medical research [1]
> and also even by statisticians. Does that make it OK?
>
> I believe that statisticians should use terms correctly in
> professional discussions and in general outside too. (In the right
> context technical terms are not jargon.) Thus there are 3 quartiles
> (one being the median), 4 quintiles, and so on. These are specific
> examples of quantiles. Growth charts show multiple centiles against
> which individuals are judged. One might be at (or above) the 70th
> centile for example. It is common to describe the distribution of a
> skewed variable using the median and (lower and upper) quartiles -
> clearly these are specific values. The term interquartile range
> becomes meaningless if we redefine the quartiles as the quarters.
> Likewise quantile regression.
>
> Confusion will surely arise if we use one word to mean two different
> things, and it seems crazy when we have two different words already.
> What is vague about the "top quarter"? Or perhaps the "top 25%"? If
> there is no simple way to describe those between the 3rd and 4th
> deciles, then call them bands or some other term. That is not a good
> reason to redefine what is a perfectly good word. One possibility is
> to call them "decile groups" - "groups of the population defined by
> the decile points". See http://www.dwp.gov.uk/asd/hbai/glossary.html
>
> Doug
>
> Reference:
> [1] Altman DG, Bland JM. Quartiles, quintiles, centiles and other
> quantiles. BMJ 1994;309:996.
>
>
>
> At 21:06 14/10/2008, John Sorkin wrote:
> >Perhaps better usage would entail using the term upper half, third,
> >quarter, fifth, sixth, etc. Quartile constitutes jargon, something
> >we all use too often, but should strive to avoid.
> >John
> >
> >John David Sorkin M.D., Ph.D.
> >Chief, Biostatistics and Informatics
> >University of Maryland School of Medicine Division of Gerontology
> >Baltimore VA Medical Center
> >10 North Greene Street
> >GRECC (BT/18/GR)
> >Baltimore, MD 21201-1524
> >(Phone) 410-605-7119
> >(Fax) 410-605-7913 (Please call phone number above prior to faxing)
> >
> > >>> Murray Jorgensen <[log in to unmask]> 10/14/2008 3:46 PM >>>
> >Here in NZ all schools are classified into 10 deciles and we often hear
> >talk of a "decile 3" or a "decile 10" school. I have never heard of
> >anyone being pulled up over this usage.
> >
> >Murray Jorgensen
> >
> >Allan Reese (Cefas) wrote:
> > > I was pulled up recently for using the phrase "top quartile" to define
> a
> > > quarter of a sample. Statistical dictionaries appear to support my
> > > critic, defining the quartile as the point estimate of the percentile.
> > > However, I feel it's a long-established extension of meaning to refer
> to
> > > the parts of the sample or population. For example, this from the
> > > Guardian 2003:
> > >
> > > "... They quote Alison Wolf, of the London School of Economics, to show
> > > that when only 10% of the population have degrees, someone in the top
> > > 25% of the ability range will still have plenty of job opportunities
> and
> > > it is a rational decision not to bother with university. But once more
> > > than 50% of the population holds a higher education qualification, the
> > > same employer that was targeting the top quartile of the population in
> > > terms of ability will now make the assumption that they require a
> > > graduate."
> > >
> > > Do you agree with my usage, or think it sloppy and wrong? If the
> > > latter, what word does describe the subset? The "top quarter" seems to
> > > me vague and ambiguous.
> > >
> > >
> > >
> > > Comments to me, and I'll summarize later.
> > >
> > > Allan
> > >
> > >
> > >
> > >
> >
> ***********************************************************************************
> > > This email and any attachments are intended for the named
> > recipient only. Its unauthorised use, distribution, disclosure,
> > storage or copying is not permitted. If you have received it in
> > error, please destroy all copies and notify the sender. In
> > messages of a non-business nature, the views and opinions expressed
> > are the author's own and do not necessarily reflect those of the
> > organisation from which it is sent. All emails may be subject to
> monitoring.
> > >
> >
> ***********************************************************************************
> >
> >
> >--
> >Dr Murray Jorgensen http://www.stats.waikato.ac.nz/Staff/maj.html
> >Department of Statistics, University of Waikato, Hamilton, New Zealand
> >Email: [log in to unmask] Fax 7 838 4155
> >Phone +64 7 838 4773 wk Home +64 7 825 0441 Mobile 021 1395 862
> >
> >Confidentiality Statement:
> >This email message, including any attachments, is for the sole use
> >of the intended recipient(s) and may contain confidential and
> >privileged information. Any unauthorized use, disclosure or
> >distribution is prohibited. If you are not the intended recipient,
> >please contact the sender by reply email and destroy all copies of
> >the original message.
>
> _____________________________________________________
>
> PLEASE NOTE NEW EMAIL ADDRESS BELOW
>
> Doug Altman
> Professor of Statistics in Medicine
> Centre for Statistics in Medicine
> University of Oxford
> Wolfson College Annexe
> Linton Road
> Oxford OX2 6UD
>
> email: [log in to unmask]
> Tel: 01865 284400 (direct line 01865 284401)
> Fax: 01865 284424
> www: http://www.csm-oxford.org.uk/
>
> EQUATOR Network - resources for reporting research
> www: <http://www.equator-network.org/>http://www.equator-network.org/
>
> Confidentiality Statement:
> This email message, including any attachments, is for the sole use of the
> intended recipient(s) and may contain confidential and privileged
> information. Any unauthorized use, disclosure or distribution is
> prohibited. If you are not the intended recipient, please contact the
> sender by reply email and destroy all copies of the original message.
>
|