Common Errors in Statistics (and How to Avoid Them), 4th Edition

TO LEARN MORE

Godino, Batanero, and Gutiérrez-Jaimez [2001] report on errors found in the use of statistics in a sample of mathematics education doctoral theses in Spain. Fanelli [2009] and Martinson, Anderson, and Devries [2005] report on the prevalence of fraud. Durtschi et al. [2004] report on the use of Benford’s Law to detect fraud.

CRITICIZING REPORTS

Commenting on an article by Rice and Griffin [2004], David Hershey (http://www.fortunecity.com/greenfield/clearstreets/84/hornworm.htm) cites the following flaws:

1. Using the arithmetic average (linear interpolation) of two values that do not fall on a straight line.

2. Plotting curves without plotting the corresponding confidence intervals.

3. Failure to match treatment groups based on baseline data. As a result, such factors as the weight of the subject were confounded with treatment.

4. No explanation provided for the missing data (occasioned by the deaths of the experimental organisms).

5. No breakdown of missing data by treatment.

6. Too many significant figures in tables and equations.

7. Extrapolation leading to a physiologically impossible end point.

8. Concluding that detecting a significant difference provided confirmation of the validity of the experimental method.

Notes

¹  Eagle Iron Works, 424 F. Supp, 240 (S.D. Ia. 1946).

²  185 F.2d 258 (4th Cir. 1950).

³  Amstar Corp. v. Domino’s Pizza, Inc., 205 U.S.P.Q 128 (N.D. Ga. 1979), rev’d, 615 F. 2d 252 (5th Cir. 1980).

⁴  487 F.Supp 389 (N.D. Tex 1980). ...