In my (Dex Whittinghill) tenure as editor for the Datasets and Stories column of JSE, I have realized that the occasional clarification of the column is in order. Last January the JSE editor John Gabrosek and I instituted some clarifications and modifications in the type and format of submissions. We believe that we did the right thing, but have come to the realization that B. below, under Expanding the field of submission, needs to be tweaked.

Potential authors should start by reading these guidelines and then the paper by Lock and Arnold (1993) (from which some quotes are drawn in the discussion below).

A 'dataset' submission is mostly that, a dataset to be added to the Data Archive. The submission must include a 'flat' ASCII text file with the data, and an ASCII 'doc' file, a template for which is included in Lock and Arnold (1993). Note that the 'doc' file includes details of the format of the text file, such as variable names and units of measurement. The 'doc' file also includes "sufficient narrative to put the data in context, and suggest some interesting questions to pursue." [An Office Excel 97-2003 file for the data can be included in addition to, but not as a substitute for, the ASCII data file.]

A 'dataset and story' submission includes the submission of the data file and the 'doc' file as described under the 'dataset' submission, as well as a significant "expansion of the narrative that is found in the 'doc' file." This expansion is a submission to JSE, and "should follow the general guidelines for any JSE article and will be subject to a similar review process." For more details see Lock and Arnold (1993).

A little more structure. From the Lock and Arnold (1993) article I want to quote the following:

"Authors are encouraged to emphasize the "story" aspect of this section by elaborating on the circumstances and questions which led to the collection of the data. We also encourage descriptions of creative ways the data might be used in teaching statistics, particularly those that are based on actual experiences."

Because the implication of the quote is that a Datasets and Stories article was originally conceived to include "teaching notes" to help the reader see how the dataset can be used in their class, and because inclusion of such material will help potential authors write an article beneficial to our readers, we are formalizing the concept of "teaching notes."

Hereafter, new Datasets and Stories submissions must include "teaching notes" that walk the reader through how to use the dataset (or simulation/program; see below) in the classroom. These notes could include recommendations such as Helpful Hints, Potential Pitfalls, and Alternative Applications from the author's use of the data (or simulation/program) in his or her own teaching. These teaching notes should be incorporated throughout the paper at the place where the teacher would apply them, and should be indented and italicized. Example teaching notes are:

Helpful Hint: I have found that the best way to utilize the applet is to first have the students perform a small-scale tactile simulation of the activity using playing cards.

Potential Pitfall: When you do this you might find that students tend to confuse the standard deviation of the sample data with the standard deviation of the sampling distribution of the sample mean.

Alternative Application: Although I have used the prices of used sedans for my activity, you could substitute the prices of used cross-over/utility-vehicles.

[While we believe that these three forms of notes cover the range of potential teaching notes, creative authors could convince us differently. Authors that cannot envision substantial teaching notes probably do not have a potential dataset and story article, but only a dataset. The rationale for indenting teaching notes is that a reader can initially read the article without the notes, and then if he or she believes that the dataset will be useful can reread the article taking the notes into account.]

Expanding the field of submissions. In the fall of 2010 JSE received two articles that looked like they would be interesting to JSE readers, but they did not quite fit in with the Datasets and Stories column or the more general articles found in JSE. The first used a dataset, but it was not a static dataset. The second used a dataset, but the primary focus of the paper was Monte Carlo simulation. We had made two additions in our guidelines, with the second (B.) now exhibiting the changes the Datasets and Stories guidelines needed. The guidelines are:

A. Articles based on a "continuously changing" dataset. Most Datasets and Stories datasets have been fixed and unchanging. This allows readers to investigate the datasets and follow the discussion given by the author. However, some data are not static over time. (For example, weather data, ozone data, and economic data that is readily available on the web.) To bridge the gap between the changing dataset and the Dataset and Story submission the author should take a snapshot of the data at a point in time and analyze this cross-section of the data. In parallel with the analysis of the cross-section of the data the author could discuss the changing nature of the data by keeping a running commentary on the potential changes that might arise through the use of "teaching notes" as discussed above.

B. Articles where the dataset’s utility is illustrated primarily with computing and/or simulation. Authors of these articles should follow the guidelines below in addition to the general guidelines described above for contributors to the Datasets and Stories column in JSE, always keeping in mind that it is the dataset that is the centerpiece of the article:

1. Program code, such as command script or macros from SPSS, JMP, SAS, Minitab, R, or S, etc., needs to be included in a ‘flat’ .txt “command” file so that it is readily accessible to the reader. The code should be described in an appendix to the paper, but actually presented in this separate .txt file so that it can be copied and pasted into the appropriate software. Hence, an author of this type of paper has one more file to submit. [Rationale: The Datasets and Stories section of JSE is meant to provide materials that readers can use in the classroom. Without the code the simulations and programs cannot be easily used by readers.]

2. Any applets or other interactive computing features should be readily available online. A link should be included to the interactive computing feature with illustration (perhaps through screen captures) of the computing feature in use.

3. When appropriate the paper must include adequate "teaching notes" that walk the reader through how to use the simulation/program to work with the data. These could occur throughout the narrative, especially when the simulation/program could be used in the classroom (for an example see Albert, November 2010 ) or in an appendix at the end of the paper, when inclusion would detract from the classroom. In either case instructions would be included in the .txt “command file.”

4. Should the simulation/programming give rise to intermediate datasets and/or files, a sufficient subset of the intermediate output, datasets and/or files should be included as well. What is sufficient will be decided by the Editor and referees. [Rationale: Analogous to the rationale for #1, inclusion of intermediate output or datasets will provide the reader with "touchstones" to verify his or her understanding.]

JSE Copyright and Usage Policy

Unlike other American Statistical Association journals, the Journal of Statistics Education (JSE) does not require authors to transfer copyright for the published material to JSE. Please click on the Guidelines for Readers/Data Users link in the menu bar at the left to access the policy.  

Department Editor

Copyright © 2013 American Statistical Association. All rights reserved.