Statistics in the Newspapers: Extra Credit
Statistics are pervasive in our society.
H.G. Wells foresaw the essential role of math and statistics a century ago.
He wrote, "The great body of physical,
...
financial science,
and endless social and political problems are only accessible and only
thinkable to those who have had a sound training in mathematical
analysis, and the time may not be very remote when it will be
understood that
...
it is as necessary to be able to compute, to think
in averages and maxima and minima, as it is now to be able to read
and write."
(Mankind in the Making, p. 192, Scribner's, New York, 1904)
Your assignment is to find newspaper articles related to five specific
statistical topics, and to submit them with an interpretive report.
The articles must be from actual newspapers, not online versions.
Avoid standard sports statistics.
For each article, include
- the actual article or a photocopy and,
- on one page,
- a complete reference for the article:
name of newspaper, date, page number(s)
- a specification of the statistical topic (from below)
- your typed report, giving an explanation and critique
of the statistics used.
Make sure that your critique is informed by your understanding
of the appropriate section of the text.
You should find one article for each topic listed here:
- Numerical summaries: measures of center (mean or average, median)
or spread (standard deviation, IQR, range) or percentiles, quartiles, etc.
Interpret the numerical summary or summaries in the context of the article.
Tell whether you think the measure used is appropriate for the given
situation, and why.
- Graphical summary or summaries of data.
Interpret the graphical depiction used in the article in its context.
Is the type of graph chosen an appropriate one? Why?
Would another type of graph be better?
Are there any potentially misleading features of the graphical display?
Does the graph depict the distribution of one variable,
or a relationship between variables?
- An observational study.
Explain why the study is observational.
Identify the individuals and the variables, including the type of variable
(categorical or quantitative
[and perhaps whether it is in an ordinal, interval, or ratio scale])
and the units of measurement.
Why, in your view, was the study conducted observationally rather than
experimentally?
- An experimental study.
Explain why the study is experimental.
Identify the experimental units, the explanatory variable(s), and
the response variable(s). How were subjects assigned to treatments?
- Poll. Describe the sampling method and the use of randomness.
Interpret the margin of error in the given context.
Discuss the appropriateness of the sample size (too large, about right,
too small).
Discuss the wording of the of the question(s) asked in the poll,
pointing out any ambiguities or biases that may affect the responses.
You should not do a report on polls until you have learned to understand
confidence intervals.
You should summarize the parts of the article that are relevant for your report.
Don't confine your reporting to just the questions and suggestions noted above.
You may also wish to critique the adequacy of the reporting in the article.
What should the writer have told the intelligent reader that was omitted?
Do not report on an article that is grossly lacking in supporting details.
You should turn in at most one article a week.
This means, in particular, that you cannot do five articles in
the last month of the semester.