Survey on typeset models

In seeking to enrich our offerings to the community, and to ensure your research is clearly communicated, we asked for feedback on typeset models. We conducted a survey, between 4th and 18th December 2012, and sought feedback relating to typeset models’ readability of text, appearance of formulae, and general spacing on a page.

11,300 recent contributors to Elsevier mathematics and statistics journals were invited to complete a survey rating 5 different typeset models 1,316 authors completed the survey, giving a response rate of 12%.


  • Sample 1 Comments:

    “Cramped letter spacing.  Math font and italics don't work well with  upright text.  Can't tell the difference  between a and alpha.”

    “Type is too small and dense. Bad for those middle-aged and older who need reading glasses!”

    “Although far better than Computer Modern, I would not say that the math font is extremely aesthetically pleasing.”

    View Sample (PDF)

  • Sample 2 Comments:

    “Good old Computer Modern font looks kind of boring, yet has the distinct advantage that if using a properly implmented LaTeX class one can detect all the possible problems when writing the article, not when submitting it. Formulas are much more readable than in Example 1 and the spacing is better.”


    "The text is a little too spread out, but I like the spacious appearance. Words appear as a lot of letters put together rather than a single entity. The font could be a little more stylish. However, this is very good."

    “I can't help it, but something about this typeset model feels more like "preprint" than "final paper" to me. Perhaps it's the larger spacing between lines. Although I find this typeset model more readable than the previous one, there's something about the previous one that I find "looks better".

    View Sample (PDF)

  • Sample 3 Comments:

    “Nice and clean. A bit poor in "personality", it is similar to basic LaTeX style. It's a choice on the safe side.”

    “I like this the best so far, in that words now look like a single entity. The lines are too close, especially in places with many mathematical symbols.”

    “I like the spacing between the text-only lines. But the distance between the lines containing inline formulas should be bigger.”

    View Sample (PDF)

  • Sample 4 Comments:

    “This is quite good. The thing I really don't like is the variable spacing of lines to make space for mathematical symbols. I would prefer a slightly more spacious appearance overall, and perhaps a more stylish font.”

    “The formulae are not spaced very well.”

    “The font is nice, easy to read, yet I still have difficulties with the readability of formulae embedded inline, even though the spacing is good in formulae.”

    View Sample (PDF)

  • Sample 5 Comments:

    “I think I prefer larger margins, it is somewhat easier to read. I am looking at the examples on a small laptop, as I am travelling. Perhaps I'll change my mind on a large screen.”

    “My opinion is that this variant is the best one: it is readable, the lines are long, and it is nice for long formulas, it allows to economy the paper (this is very important point for our ecology!)”

    “Do not like this style, because it has the same weak points as style #1 (not excellent overall readability), but missing the strong points (nice individuality and characterisation). It is not overall bad, but one can play better with the parameters at disposal.”

    “Lines are much too long - it makes the paper hard to read. And there is not enough spacing between words.” 

    View Sample (PDF)

The survey results have provided a good understanding of the community’s preferences for typeset models. Overall all, there is a broad appeal for models using Computer Modern and Times Fonts. And most importantly, we are able to examine preference for typeset model by subject areas. For example Times Fonts are favoured by statistics and probability disciplines whilst the algebra disciplines favour a Computer Modern model.

We would like to extend our thanks to all those who took part in the survey.  We are now working to adjust current models to suit these preferences and also create a new typeset model for use in mathematics and statistics journals.

More detailed results of the survey.