On Tue, 14 Jan 2014 04:08:52 -0800, Rainer Scheuchenpflug wrote:
Dear Tipsters,
one of my students pointed me to a tweet by JK Keller which
contains an excellent example of a misleading diagram:
https://twitter.com/jk_keller/status/410498080765919232
Could be useful in a research methods class.
A few points:
(1) What is completely obvious to one person may be completely
opaque to another person if there is not sufficient "common
ground" in the communication. In a case like the one referred
to above, one needs to be explicit about which rules of
graphical representation are being violated as well as why
a "corrected" figure is better than an "misleading" diagram.
(2) It could just be me but what exactly is misleading in the
"uncorrected" figures. Wikipedia has an entry on
misleading graphs that provides (a) specific types of misleading
techniques and (b) Tufte's rules for measuring distortion in
figures. Which violations of good practice are present in
the "uncorrected" and "corrected" figures? See:
http://en.wikipedia.org/wiki/Misleading_graph
NOTE: For good examples of misleading figures in the
Wiki entry, the use of pie charts should be emphasized as
well as why they are misleading. I suggest that a good
homework exercise for classes is for students to check
the reports and new releases put out by one's college
and determine what percentage contain pie charts.
(3) One point that there should be common ground for all
researchers is to examine original sources and not rely upon
secondary sources since one does not know whether the
secondary source is an accurate representation of the primary
source (think Freud's iceberg). The "Tweet" refers to the
following Reuter's news article -- NOTE that it has been
updated to take into account the above Tweet; see:
http://blogs.reuters.com/data-dive/2013/12/10/gms-new-ceo-is-another-small-crack-in-the-glass-ceiling/
(4) The point of the Reuters article appears to be summarized
in the following quote from article:
|Women are still far from reaching parity in upper reaches of big
|American corporations.
Question 1: How does the uncorrected figures mislead the reader
about this point? For example, how does it imply that the percentage
of female CEOs in the Fortune 500 is much greater than 4%,
under the assumption one can read the y-axis?
Question 2: How does the mindless use of "rituals" (the usual favorite
example is null hypothesis testing) lead one to be consist with the
rituals but not necessarily clear and understandable communication?
Does the corrected figures really provide better information? What
will the editor say about all that unused space in the figure that
will be used on the expensive printed page?
Question 3: How do responses to the tweets seem to be a good
example of conformity as shown by, say, Asch's experiments.
See the Wikipedia entry on Asch's research program:
http://en.wikipedia.org/wiki/Asch_conformity_experiments
(5) I cover APA style I point out that it is continually evolving
to take into account new forms of communication, such as social
media. I have pointed out that this includes Tweets and though
I could not think of a instance of when one would want to cite a
tweet in a paper it is possible that such an instance could occur.
The APA website apastyle and the APA guide to electronic resources
would have to be consulted in order to (a) determine how to cite
the tweet and (b) whether one needs to provide a reference for
the tweet (in contrast to a "personal communication" though this
doesn't seem appropriate given that it is a "public communication").
I now am happy to have an example of a tweet that one might
want to reference in a manuscript. ;-)
-Mike Palij
New York University
[email protected]
P.S. Is the second figure (B) correct in the following:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3361017/figure/F3/
P.P.S. There is the issue of sampling bias in using the Fortune 500
data which indicates that only about 4% of the CEOs are female.
In contrast, the Dow-Jones Industrial Average, which consists of
the biggest 30 U.S. corporations, has three female CEOs (in IBM,
HP, and Dupont) or 10% or more than twice the number based
on the Fortune 500. Are there other samples that can be used?
Yes (e.g., the universe of all companies, the set of S&P 500, etc.).
Of course, all of this is really besides the point since we should
expect CEOs of companies to be 50% female, more if you think
women make better decisions than men; see:
http://www.degroote.mcmaster.ca/articles/women-make-better-decisions-than-men/
---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here:
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=32716
or send a blank email to
leave-32716-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
