One sees axis breaks in epidemiology, though not all that often. For
example, one might be interested in graphing the weekly incidence of
influenza over several seasons. One may want to show that Influenza B
strains were more common during the inter-epidemic periods, while Influenza
A strains were more common during the epidemic periods. If there were no
axis break, then the axis required to show the high incidence rate during
the epidemic periods would make it difficult to see the difference between
the A and B strains during the low-incidence inter-epidemic periods. There
are other graphical solutions, to be sure, but the axis break is a good
solution for this sort of thing. For clarity, it is essential to break the
bar as well as the axis.
Lee
Lee Sieswerda, Epidemiologist
Thunder Bay District Health Unit
999 Balmoral Street
Thunder Bay, Ontario
Canada P7B 6E7
Tel: +1 (807) 625-5957
Fax: +1 (807) 623-2369
[email protected]
www.tbdhu.com
> -----Original Message-----
> From: Nick Cox [SMTP:[email protected]]
> Sent: Wednesday, May 14, 2003 1:04 PM
> To: [email protected]
> Subject: st: RE: broken axis symbol?
>
> Ernest Berkhout
>
> > so far I was not able to detect an option in Stata 8 graphs
> > to display an
> > axis with the sort of layout that shows that is does not
> > start 'naturally'
> > at the zero-point. Visually, I mean something like this:
> >
> > |
> > 6
> > |
> > 5
> > |
> > >
> > <
> > |
> > +-----
> >
> > Is it possible but did I overlook that? Or is it not an
> > option yet? In the
> > second case, are the Stata people willing to introduce
> > something like that
> > in the future? From the 'How to lie with statistics'-line
> > of thought my
> > opinion is that this would make a lot of sense for people
> > when submitting
> > graphs to scientific papers.
>
> Clearly you want this, and your main questions are
>
> (1) can you do it in Stata 8? (I guess not)
> (2) is it on Stata Corp to do list? (don't know)
>
> Nevertheless I want to discuss the presumption here.
>
> As I recall, Darrell Huff in his 1954 book implied that axes
> which don't start at zero are deceitful. And I seem
> to remember my high-school science teachers saying the
> same thing in the 1960s and recommending a scale break.
>
> But another point of view is this: if people can't be
> bothered to look at the axes and read the labels,
> then they won't understand the graph, and whose fault
> is that?
>
> William S. Cleveland in his book "The elements of
> graphing data" (full reference in manuals) includes
> the following advice:
>
> Do not insist that zero always be included on a scale
> showing magnitude. (p.92)
>
> Following that, he clearly and firmly discusses
> and dismisses the Huff dictum.
>
> He also says
>
> Use a scale break only when necessary.
> If a break cannot be avoided, use a full scale break.
> Do not connect numerical values on two sides of a
> break. Taking logs can cure the need for a break. (p.104)
>
> A full scale break is more than a squiggle on the axis.
> It is a division into separate panels.
>
> I don't think any time series graph for recent decades suffers
> in the least from not having a scale break because it doesn't
> go all the way back to a time origin, whether BC/AD
> or the origin of the Earth or whatever.
>
> My question, therefore:
> are there literatures in which this is still standard?
> Just curious.
>
> P.S. There is a tip at
>
> http://www.stata.com/support/faqs/graphics/scbreak.html
>
> The following has been added, not by the author:
>
> Note: This FAQ is relevant for users of releases prior to Stata 8
>
> but the _idea_ about what to do is equally applicable
> to Stata 8. All that has changed is the syntax to implement
> it.
>
> Nick
> [email protected]
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
