Archives

Categories

Ad Lagendijk Ad Lagendijk 15 August 2008

Giving your new results away too soon

Posted in Getting published, Tips

A scientific paper has a dull structure: Title, List of Authors, Abstract, Introduction, Results, Conclusions, References are the headings of sections to be found in many papers. However exciting and new the results of your paper are, do not experiment by inventing a new, original structure that will surprise and confuse your readers.

But given this rigid structure where do you announce your results first: in the title? In the abstract? In the introduction? Or, in the results paragraph? If you wait to long your paper will become a whodunit and readers will get bored and stop reading your paper. If the clue of your paper is already in the title you might fear that many of your readers will only read your title and will then go on to read the next paper.

In a number of cases the results of your paper will be so complicated that they cannot be summarized in one sentence, so that the title-option will not be applicable. I think it is bad practice to withhold information in the abstract with the intention to force your readers to read on. Any reader should be important to you. Also those readers that only read your abstract. Or only look at one of your figures with its caption. Help these people by making your text as modular as possible.

How about the Introduction? My idea is that in the introduction you describe a field. You discuss prior, relevant, results. You make clear what additional information, theory, or observations are wanted for to advance this field. Then you announce in the introduction that these badly needed results will be supplied by you in exactly this paper. But you should not already present in the introduction your (main) results. The introduction is about what other people have done before. A number of my Ph. D. students do not agree. They are eager to present immediately their new results in the Introduction.

- - - - - -
If you like this post why don't you email subscribe to our new posts. Or subscribe to our RSS feed.
  1. Allard Mosk

    20 Aug 2008 18:08, Allard Mosk

    By the end of the introduction, there should at least be a fairly accurate description of what _type_ of result is going to follow. I think we all agree on that. My idea is, if the result can be captured in a single, simple sentence, there is no harm in stating it (but, usually, neither is there a _need_ to do so, so it remains a matter of taste).

    Some examples:

    Too vague (to my taste):
    In this letter we present measurements of Y waves and compare them to our model.

    About right (to my taste):
    In this Letter we present measurements of the power spectrum of Y waves in the 1-6 MHz region, and find reasonable agrement with an Y-wave model.

    Too much information: (a real killer, not a matter of taste)
    We present measurements of the power spectrum of Y waves in the 1-6 Mc region and find that it is constant at a level of 37.4(3) dBm in the 1-3 Mc band, and increases in the high-frequency region. We present a model that reproduces the Y-wave increase correctly but consistently shows an amplitude that is 2.6 dB higher.