24

I am writing a review paper and I came across a literature in which measurement of force is given in pounds and likewise for other quantities. In all the other papers, only the SI system of units is followed. Should I use the units used by the author or should I convert the measurement values to SI units?

  • 9
    I would convert the measures to SI units. But details could be domain specific – Basile Starynkevitch Apr 2 '18 at 8:21
  • 1
    Would the answer change for different countries, or even different faculties ? – Criggie Apr 2 '18 at 11:12
  • 10
    If you do convert the measures to SI units, please indicate such in the paper. If you're converting someone else's data (i.e., from another paper), please also include a note that you've converted their data, and include a relevant citation as to where the original data (in non-SI units) can be accessed. – tonysdg Apr 2 '18 at 15:46
  • 5
    To expound on @tonysdg 's comment, converting units involves multiplication by a conversion factor, generally followed by a rounding. You should make a note (a footnote would probably be the best place) discussing what the original units were and what you're taking the number of significant digits to be, and thus what sort of rounding you're doing. – Acccumulation Apr 2 '18 at 15:58
  • Yes, please always use SI units. Here in Norway unfortunately calorie is still commonly used for food energy measurement, while our Danish neighbours have sensibly switched to most commonly use kilojoules. Please contribute to making the world a better place by using SI units. If you must use non-standard units to better target some audience, always also present the corresponding SI value. – hlovdal Apr 3 '18 at 17:16
49

To avoid confusion and to ensure the comparability of the results, you should use uniform and consistent units throughout the paper. You can write a note explaining that in the original paper the authors used pounds, furlong per fortnight etc.

And of course, yes, use the SI units. Nowadays all non-SI units are defined in terms of SI units. However, don't trust random sources for the conversion factors: an up-to-date and recommended source for conversion factors is the NIST Guide to the SI, Appendix B: Conversion Factors.

  • 10
    And not specifically that if not using S.I. units, the units may be ambiguous, e.g. the term "gallon" can refer to the ~3.8-litre US gallon or the ~4.3-litre Imperial gallon. – Jim MacKenzie Apr 2 '18 at 17:43
  • 2
    There are plenty of fields which use non-SI units, often for good reasons. For example, Astronomy generally uses CGS. A good rule of thumb is to use the most customary units in your field, which is not always SI even within the natural sciences. – KAI Apr 2 '18 at 22:44
  • 7
    @TonyStewartEEsince1975 This is linear. – BartoszKP Apr 3 '18 at 7:49
  • 2
    @BartoszKP: it is an affine map, not a linear one (due to the offset). – Andreas H. Apr 3 '18 at 8:01
  • 1
    @TonyStewartEEsince1975 and BartoszKP: Celsius and Fahrenheit temperatures (note that they are two different physical quantities, not two units for the same quantity, see physics.stackexchange.com/a/320193/111969) are both related to the thermodynamic temperature through an affine map. In this sense, the Fahrenheit temperature is no worse than the Celsius one, but the latter has the advantage that the magnitude of the degree Celsius equals the magnitude of the kelvin. – Massimo Ortolano Apr 3 '18 at 8:22
13

I echo Massimo Ortolano's answer that the most important thing is to be consistent: if you normally use SI units in your article, then you should definitely convert the units from articles that use other systems to the same system as the rest of your article.

However, although this wasn't your actual question, you should not assume that you must always use SI units by default. Although that is indeed the almost universal standard for scientific journals, it is not necessarily the case for non-scientific disciplines (and you didn't specify which discipline you are talking about). In particular, some American journals (probably even some scientific ones) seem to permit non-SI units (for example, the published article you found with non-SI units, though that might possibly be an older article). So, the question of which units to use actually depends on the standards of the specific journal in which you are publishing. You should consult the guide for authors or ask the editor if you have any doubt on this point.

  • I'd venture a guess that many journals permit non-SI units. For example, in my field it's very standard to use non-SI units for some quantities, and I've never known any of the popular journals in the field to have a problem with that. – David Z Apr 2 '18 at 9:22
  • 10
    @DavidZ Many journals actually require the authors to use SI units.Some traditional, field-related, non-SI units are usually tolerated though. How strictly certain requirements are enforced depends also on the reviewers. We once used "ppm" (I know it's deprecated but it's quite common in everyday discussions) in a paper and the reviewer required us to remove it. – Massimo Ortolano Apr 2 '18 at 10:46
  • 6
    "The proof of this? Obviously, the published article you found with non-SI units." That only proves that some time in the past, articles with non-SI units were accepted. It doesn't imply that this is still the case. – Abigail Apr 2 '18 at 21:47
  • @Abigail Good point; I've revised my answer accordingly. – Tripartio Apr 3 '18 at 7:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.