|
Table. Editorial criteria for mathematical, economic, and statistical papers |
|
|
Criterion |
|
|
Overall content |
-
Must provide information that our audience (public health officials) is likely to find of "immediate and practical" value.
|
-
Must reflect the realities of public health.*
|
|
Writing style |
-
Should be written in a style most likely to appeal to most of our intended audience.
|
-
In the main text, equations should be kept to a minimum, and those that are presented should preferably be written out in words rather than mathematical notation.
|
-
In the main text, and in diagrams and tables associated with the main text, mathematical notations should be kept to a minimum.
|
|
Technical aspects |
-
Where possible, mathematical, economic, and statistical articles should include a simple schematic diagram outlining the elements in the model(s) and how they are connected.
|
-
Models should contain detailed sensitivity analyses. Univariate (1 variable at a time) sensitivity analyses are generally considered inadequate. One goal of sensitivity analyses should be to define which inputs are, within the model, relatively most important.
|
-
All articles should contain sufficient description of the methods to allow independent replication of results by another researcher with suitable skills and interest.
|
-
Papers should contain a Table 1 which lists all the inputs, the values and ranges, and identifies data sources.
|
-
Figures should be as simple as possible. The use of color should be kept to a minimum.
|
-
It is insufficient to only report p values as evidence of statistical significance. Authors must also report some measure of dispersion (e.g., standard deviations, confidence intervals).
|
-
For statistical models, a table of results should provide the results of all the variables used in the model, the statistical significance of each variable, and a measure of goodness-of-fit of the entire model.
|
|
|
*Public health officials have to be practical. Purely conceptual modeling papers, for example, are unlikely to be of immediate and practical value to our intended audience. |
