59

A paper I am reviewing very clearly originates from the US. The authors don't use SI units. As a European, I am always put off by non-metric units and I would like to request to change the units to SI units, not only because I am more familiar with them, but also because it would make the paper more accesible to researchers worldwide living in countries where metric units are used (almost all countries).

Is this request to ask to change the units unreasonable? In a way it feels like it, because the authors did not do anything wrong or incorrect?

Additional info: the journal doesn't provide a guideline for unit use, the units in the paper are non-SI non-metric (of the pounds per square inch quality) and the discipline is specialized engineering subfield.

Sursula
  • 20,540
  • 8
  • 62
  • 121

7 Answers7

60

The International System of Units has a universal value, and if the paper is written for an international audience then, yes, ask the authors to change the paper to conform to the SI.

Check also the journal guidelines: most journals explicitly require the usage of the SI. Should this be the case, you can refer the authors to the guidelines.

You can also point out that nowadays the US customary units are exactly defined in terms of SI units. Therefore, also those who use the US customary units, from a scientific point of view, are actually using the SI, because it's the SI that specifies the definitions of those units and their realizations. For instance, 1 in = 2.54 cm, where the centimetre is the SI one (for more about the connection between US units and SI units, and for a comprehensive list of conversion factors, see this publication from NIST).

Massimo Ortolano
  • 55,428
  • 19
  • 166
  • 207
  • 48
    I don't think this third paragraph is on-subject. Particularly, I worry if OP actually pointed it out to the authors, it would be perceived as condescending. – ShapeOfMatter Mar 05 '22 at 02:03
  • 14
    @ShapeOfMatter In my experience, reviews and response to reviewers are exchanges were people does not make much effort to avoid condescension. So, if that gives the opportunity for the authors to learn something, as reviewer, I'd accept to be perceived as condescending. – Massimo Ortolano Mar 05 '22 at 10:15
  • 12
    @MassimoOrtolano Am I understanding correctly, that you are essentially suggesting to jump the bandwaggon and just make no effort to be non-condescending because nobody does it? If so, that is precisely the reason why I'm so put off by academia. – infinitezero Mar 07 '22 at 08:24
  • 1
    @infinitezero The key point of a review is to help the authors to provide correct information according to certain standards. If, in doing so, one is perceived as condescending, I think it can be tolerated. And if you're put off by the academic environment, I can say that I have also experience in industry and, in my experience, the industrial environment is no less condescending than the academic one. – Massimo Ortolano Mar 07 '22 at 18:16
  • 2
    @infinitezero while there is a risk to be perceived condescending, I find the fact that they did not use SI in the first place a significantly larger problem. There are standards, and for good reasons. It's not that I would write my paper in Latin or German instead of English. They're action probably stems from lack of knowledge (that SI is the standard in most sciences) and therefore also likely don't know the latter point. Why else would they not use SI? – Mayou36 Mar 07 '22 at 22:35
  • 1
    @Mayou36 To make it clear: I absolutely agree to request the authors to convert to SI units. There's no way to tell why they wouldn't use SI and it doesn't matter. The problem is "[making] no effort to be non-condescending". Even if they were to lack the knowledge that SI is the standard, telling them that "their units" are derived from SI and they're already using SI units anyway, bears no relevance. Make the world a better place, not a more toxic one. – infinitezero Mar 08 '22 at 07:49
47

If it's in fields I am familiar with, using metric is mandatory and no reviewer should let a paper through that does not use metric.

However, some metric units are not SI and may be acceptable. In many fields it is widely encouraged to use eV instead of J, amu instead of g, angstroms instead of meters, etc. Feet instead of meters or F instead of C is not okay.

You need to know the customs of your field.

Anonymous Physicist
  • 98,828
  • 24
  • 203
  • 351
  • 27
    Note that the units you mentioned (eV, amu, angstrom) are officially recognised as "Units accepted for use with the SI", alongside the minute, the litre, the decibel, the astronomical unit, mmHg, and some others. – Glen O Mar 05 '22 at 04:53
  • 5
    It very much depends on the field. For example – as far as I’m aware – aviation commonly uses feet and knots all over the world. Of course it would still be prudent to convert such units to SI units before doing any kind of calculation with them. – Michael Mar 06 '22 at 08:04
  • 2
    @DanRomik: They use feet for altitude, nautical miles for distance and knots for speed as far as I’m aware. – Michael Mar 06 '22 at 17:27
  • 1
    @Michael Feet is for height above the ground. Pressure altitude (flight level) is in units of 100 feet. In some parts of Asia, flight level is in meters. – Anonymous Physicist Mar 06 '22 at 17:45
  • @Glen O the amu, ångstrøm and mmHg are not currently accepted for use with the SI: https://en.wikipedia.org/wiki/Non-SI_units_mentioned_in_the_SI – glaux Sep 16 '22 at 11:38
19

If the journal’s intended readership extends beyond the US, Liberia and Myanmar, the only three countries in the world that officially use a non-metric system of units, then yes, it is not only reasonable but in fact your responsibility to the journal and its readers to point out this feature of the paper that makes it less accessible to the journal’s international audience, and ask the authors to address this by using standard (SI) units.

This is assuming that the journal is a science journal. If the journal publishes, say, creative writing or some other non-scientific content that just happens to mention someone walking two miles or describes a hot weather day by saying it was a hundred degrees out, it’s appropriate to let the author keep their units.

Dan Romik
  • 189,176
  • 42
  • 427
  • 636
  • 6
    The US officially uses SI units. But it’s also a free country, and NIST is a non-regulatory agency. – Gilbert Mar 05 '22 at 04:08
  • 9
    @Gilbert interesting. I researched this a bit, the real story is apparently quite a bit more complicated than what either of us wrote. Regardless, imperial units are the de facto standard in the US for almost all non-scientific uses. The claim that SI is the official system of units has a kernel of truth to it, but is misleading in the extreme IMO. – Dan Romik Mar 05 '22 at 06:06
  • 3
    oh I agree that it is misleading to say that the US uses metric. I’m just leaning heavily (and admittedly, pedantically) on the word “officially.” It’s indisputable that NIST is the official national metrology institute, and that’s all I mean. NIST is all-in for SI. For example, as a matter of policy, all NIST publications are mandated to exclusively use SI units, and the duty of all calibrations is traceability to the SI. NIST advocates for SI usage in the country and represents the US at the meetings of the General Conference of Weights and Measures. – Gilbert Mar 05 '22 at 13:24
  • 2
    All of the grocery scales and gas pumps (among other things) in the country are calibrated as a matter of state law, and the state metrology labs all ultimately rely on NIST’s SI-based primary calibrations for interstate and international consistency. In a real way, SI is the bones of commerce in the US, even if Imperial is the skin. – Gilbert Mar 05 '22 at 13:36
  • 3
    @Gilbert SI may be the official in the US but there are a lot of laws/regulations based on non-metric. Gasoline is taxed at per gallon rate. Property taxes on on sq. feet or acres. Milk subsidies at per hundredweight. Federal/state/local governments have not caught on that we're officially SI. – doneal24 Mar 06 '22 at 16:05
  • @doneal24 I agree! And speed limits are in MPH. But nowhere are these units actually defined except by NIST and in terms of the SI. – Gilbert Mar 06 '22 at 16:37
  • 1
    @doneal24 The US (liquid) gallon is defined as exactly 231 cubic inches, and the inch is defined as exactly 25.4 millimetres. Acres are defined in terms of feet, which again go back to millimetres. A hundredweight is 100 pounds, with an avoirdupois pound being defined as exactly 0.45359237 kilograms. Dig deep enough and it all goes back to SI. – GB supports the mod strike Mar 07 '22 at 12:18
  • @DanRomik fyi, US customary units are not Imperial units. Imperial units are the system used in the commonwealth prior to various states becoming metric, and whilst most of the units share the same names, and many the same values, some don't (e.g. a US pint is < 0.5 l, whilst an Imperial pint is > 0.5 l). There are also some units only in one (e.g. cups are US customary but don't exist in Imperial, and afaict, stone are Imperial but don't exist in US customary) – Tristan Mar 07 '22 at 13:58
  • @GeoffreyBrent I know the customary units are defined in terms of SI units. My point was more towards why are milk subsidies defined in terms of hundedweight instead of kg.? If SI is official shouldn't the government be using it? – doneal24 Mar 07 '22 at 19:01
  • @doneal24 the federal government, at least, should be using it (though not all lawmakers have caught on, and there’s a need to balance principle with practicality). The states and localities are free to do as they please. – Gilbert Mar 07 '22 at 19:29
  • @Gilbert It's not the lawmakers that are setting the standards. USDA, Commence, FDA, HHS, etc., all need to get on-board. – doneal24 Mar 07 '22 at 19:33
  • 1
    I will give up my furlongs-per-fortnight when you pry them from my cold dead hands. – Tony Ennis Mar 07 '22 at 19:49
  • @doneal24 suffice to say, the Metric Conversion Act of 1975 remains a work in progress (although it’s rather sad if even the Department of Commerce as a whole hasn’t embraced SI; that’s NIST’s department!). – Gilbert Mar 08 '22 at 03:09
14

If the units are crucial for the understanding of this paper, particularly for replicating results, reusing the methodology and so on, I would point it out. "Take 0.18 ounces of potassium hydrogenphthalate and dissolve in 3 cubic inches of distilled water" probably would not do.

If, on the other hand, the values themselves seem less important for the wide scientific audience to examine or less unit conversion is needed in the first place (e.g. climate studies and temperatures in Fahrenheit) - I would not raise this issue.

Lodinn
  • 8,379
  • 11
  • 45
10

You can suggest. I'd avoid it, personally. It would be wrong to vote to reject based on that alone. The editor probably has a view and the journal may have some standards (required or suggested) that cover it.

But a statement that a switch would increase accessibility to a non-US audience is benign. And in some fields in might be more important than in others.

Buffy
  • 363,966
  • 84
  • 956
  • 1,406
  • 13
    That is not matter of chauvinism, but a matter of standards in the field. Everyday life and science may not use the same ones. I also don't insist on writing my papers in German. That being said, I agree with the rest of your answer (+1) – Severin Schraven Mar 06 '22 at 18:59
  • @SeverinSchraven I think you should write your papers in German if it makes you happy. Seriously. – Tony Ennis Mar 07 '22 at 19:47
  • 1
    @TonyEnnis I would have a hard time doing so, given that I know most of the terminology only in English (because this is the standard) :) It is was just a lucky coincidence that my mother tongue is some obscure language to make my point. In fact, I really dislike people whining about being unable to publish in German (or French for that matter) anymore. This is the new standard, just roll with it. If everybody used their own language, I wouldn't be able to read half the research papers... not a world I want to live in. – Severin Schraven Mar 07 '22 at 20:31
10

I have reviewed a paper covering both my field (spectroscopy) and another (diamonds). The latter uses some units that aren't really used anywhere else - carats - even in scientific writing.

The paper was in a materials journal and therefore aimed at non-specialists, so my review made a strong recommendation to provide SI conversions whenever field-specific units were used. That seemed acceptable to all parties.

Chris H
  • 8,576
  • 22
  • 36
  • 4
    This. The paper should feel free to use Imperial units, but should also have the equivalent SI alongside for the wider audience. Going through and adding the equivalent values is an easy task to do before publishing. – Dúthomhas Mar 06 '22 at 14:52
  • 1
    @Dúthomhas my preference would be SI first, but including SI is a necessity (and I said roughly that in my review) – Chris H Mar 06 '22 at 14:58
  • 3
    The paper should present the most precise value first, then the conversion, IMHO. – Pablo H Mar 07 '22 at 17:40
  • 1
    If I were reading about diamonds, then carats might be the most appropriate unit. – Tony Ennis Mar 07 '22 at 19:42
  • @TonyEnnis about 50% of the target readership would agree. The other 50% wouldn't have a clue unless they'd been buying jewellery. That's based on my coffee-break straw poll at the time; even some of the diamond growers preferred grams – Chris H Mar 07 '22 at 21:17
-1

I agree that the US Government and citizens should migrate to predominantly SI Units by choice and judgement and therefore requesting this of US citizens is eminently reasonable, but I’d add ‘please’, ‘you know it makes sense’ and ‘thanks for your consideration’.

I wholeheartedly agree with concepts of freedom of choice and the like, but in the units case, this must start with schools, universities, teachers and textbooks used by students, so that students firstly become familiar with SI Units from the start and have to ‘mentally convert’ or calculate to use Imperial or non- SI units. The impetus should come from a judgement that academics should choose to make in this field (or should no longer stand back from or defer).

I agree with international practices in fields such as Aviation.

The UK had similar, prevailing non SI units prior to 1970’s but the Educational system with the help of legislation was able turn it about.

Another field in which there could be world-wide conformity is in the dating system, in which dd/mm/yyyy seems more consistent or logical than mm/dd/yyyy.

Amoolya Prasad
  • 11
  • 1
  • 1
    If we're changing the way we write dates, we may as well all use the objectively superior yyyy/mm/dd (or yyyy-mm-dd), which is the only date order that will make your digital photo albums automatically sort by date. – A. R. Sep 16 '22 at 18:07
  • Although yyyy/mm/dd is objectively superior, I think that superiority is for machines, calculators and computers. For humans, we think innately in weekdays, date in month and then by month. We know the year but don’t usually mention it in conversation, unless it is a different year - even then we favour last year or next year if those are the relevant periods. – Amoolya Prasad Feb 03 '24 at 18:48
  • If you asked me the date, I could tell you without looking that it's 2024, it's Monday, and it's February. I had to check the calendar to see that it was the fifth, because I think very little about specific dates. – A. R. Feb 05 '24 at 13:11