Becoming a Teacher of Statistics
Portfolio
by
Susan Sexton
Media Resource
Probability and Statistics for Secondary
Teachers
STAT 6070
University of Georgia
Fall 2007
Instructor: Christine Franklin
Berenson, A. & Carey, B. (2007, September 14). Experts question study on youth suicide rates. The New York Times, pp. A14.
URL: http://www.nytimes.com/2007/09/14/us/14suicide.html
This
New York Times (Times) article is a
reaction to a study, published in The American Journal of Psychiatry,
associating the effects of lower prescription rates of antidepressants for
children and adolescents on the youth suicide rates. While the Times article provides details of ÒexpertsÓ who critique
the validity of the journal article and a response from the journal articleÕs
lead author, it does not provide enough detail for the general reader (who does
not have prior information about the journal article) to make intelligent
decisions on whether or not the ÒexpertsÓ are making valid and reasonably
statistical critiques.
The
major critique discussed in the Times
article discusses the importance of considering lurking variables and the
feasibility of running short-term trial (observational) studies on suicide
which is considered to be Òrare and uniquely personal events that can be driven
by many factors.Ó However, the lurking variables, identified as confounding
variables, leave the reader to wonder whether or not the actual study under
discussion includes the lurking variables in their analyses (although, in
reality, the actual study does not).
While
the journal article does claim that as the prescriptions rates decreased, the
suicide rates increased, the authors do make a disclaimer that, ÒBecause the
period of observation in this study was brief, our findings must be considered
preliminaryÓ (Gibbons, Brown, Hur, Marcus, Bhaumik, Erkens, Herings, Mann,
2007, p. 1359). However, the authorsÕ use of data available from the trial
observations is problematic, as pointed out by the ÒexpertsÓ cited in the Times article, due to the fact that part of their claims
are made on data that has not yet been collected for 2005.
As stated earlier,
without prior knowledge provided by the journal article, the reader can only
use the information provided by the newspaper article in an attempt to make a
decision about the results of the study. Thus the TimesÕ article provides a unique opportunity to expose
students to an actual published study, make their own statistical arguments for
their opinions on the analyses and results of the study and compare their
opinions to those of the ÒexpertsÓ cited in the newspaper article.
The
journal article is six pages of content and uses many of the statistical ideas
presented at the secondary level. Therefore an AP statistics student, with a
fair amount of coursework completed, should be able to read the article and
make intelligent statistical decisions about the analyses and results of the
study. Which article is presented first to the students is a matter of
preference. Reading the Times article
first and having lingering questions to be addressed by reading the journal
article would be one way to motivate the discussion. On the other hand, having
students read the journal article first, react, and then read the reactions
presented in the Times article, is an alternative method for motivating
discussion.
Regardless of the method chosen, the
statistical information, analyses, and discussions provided in both articles
are a great way for students to engage in statistical reading, arguments, and
discussions to enhance their statistical understanding of both the material and
how it is used in the professional community.
Reference
Gibbons, R.,
Brown, C., Hur, K., Marcus, S., Bhaumik, D., Erkens, J., Herings, R., &
Mann, J. (2007). Early evidence on the effects of regulatorsÕ suicidality
warnings on SSRI prescriptions and suicide in children and adolescents. The
American Journal of Psychiatry, 164(9),
1356–1363.