Mathematics and Education

A slow blog

Avoiding Bad Statistics

leave a comment »

After reading Damned Lies and Statistics and Stat-spotting,[i] I suspect that factors contributing to bad statistics include ignorance of how statistics are produced, innumeracy, and selection of the most dramatic statistics.

These factors may explain why The American didn’t post a correction or a source—apparently no one thought there was a mistake and no one thought a source should be cited. (This is rather sad. The American is published by the American Enterprise Institute which claims to pursue its ideals “through independent thinking, open debate, reasoned argument, and the highest standards of research and exposition.”) The garbling and the absence of a source for the statistics in The Mathematics of Sex may have occurred for similar reasons, exacerbated by hurried production. (That garbled statistics and various typographical errors went unnoticed is also rather sad. The Mathematics of Sex was published by the venerable Oxford University Press which has as its mission “to publish works that further Oxford University’s objectives, including its objectives of excellence in research, scholarship, and education.”)

Although noticing mistakes may require numerical sophistication or knowledge of particular fields, accurate reporting of names, dates, and sources of statistics does not take much skill. At the very least, authors and research assistants can copy categories and sources as well as numbers. Editors can (and should) ask for sources.

Sources can be indicated in a variety of ways. As the Chicago Manual of Style puts it, “Whichever system is chosen, the primary criterion is sufficient information to lead readers to the sources used.”[ii] In scholarly work, citations in the text indicate the sources that provide evidence for claims and the sources themselves are listed in the bibliography. Journals in psychology often require that authors follow very specific guidelines from the American Psychological Association.[iii] In popular books, sources are often given in endnotes (as I have done here). Newspaper articles frequently give the source for information that is presented graphically, but tend not to give citations in articles. However, they often include enough information to allow readers to find sources.

Absence of any documentation for statistics suggests authors and editors do not consider statistics very important. That may be the case in some instances. However, publications such as The Mathematics of Sex and The Science on Women and Science begin with the premise that there are few women in some fields of science. It seems very strange to make the effort to write about a phenomenon without accurately documenting its existence.

For readers who notice, the absence of documentation and the presence of statistical blunders suggest that more subtle errors are lurking. For those who do not—beware! Bad statistics may not suck your blood, but they can keep you in the dark.

Next


[i] J. Best, Damned Lies and Statistics: Untangling Numbers from the Media, Politicians, and Activists, University of California Press, 2001; J. Best, Stat-spotting, University of California Press, 2008.

[ii] Chicago Manual of Style (15th Edition), University of Chicago Press, 2003, p. 594.

[iii] Publication Manual of the American Psychological Association (6th Edition)
, American Psychological Association, 2009.

Advertisements

Written by CK

March 9, 2010 at 4:46 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: