9
$\begingroup$

I would like to know if there is a preferred typographic form for the term "z score".

A search on Google scholar for "z score" shows six forms

  1. no markup: z score
  2. hyphenated: z-score
  3. capitalized: Z-score

and 4, 5, and 6: the above forms with z italicized, e.g., z-score.

I don't see any favorites with a quick scan of the first three pages.

$\endgroup$
1
  • 2
    $\begingroup$ I prefer z̖̜̑̇̌̏̎̏̕̕-s̥̭̝̤̏̃̓̏̋̚có̢̪̮̙̤̔̍̆̚ȓ̡̢̗̭̮̙ȩ̭̫̄̋̇. ;) Welcome to CV.SE. $\endgroup$
    – Alexis
    Commented Jan 28 at 3:11

3 Answers 3

12
$\begingroup$

It depends on where you intend to publish.

The APA publication manual (6th edition) recommends using "z score" (italicized, without a hyphen). (Note that this rule seems different from the version of the APA manual that Peter Flom mentions in his answer – so you should check which version of APA you're asked to follow!).

The NCBI style guide uses an italicized, hyphenated "z".

The ASM guidelines say:

Z score or Z-score (with italic or roman Z)

There are probably many other guidelines out there, saying different things.

Now, if you intend to self-publish something, the best course of action to take is probably to follow already existing guidelines that people in your field generally use. But if you don't want to do that, the least you can do for your readers is to stay consistent in the choice you'll make.

$\endgroup$
1
  • 3
    $\begingroup$ (+1) Be consistent within your own paper and follow any established convention in your field are perhaps the top meta-rules in this territory. I'd add small votes for (1) noting that you may be able to avoid any notation by using an explicit word-definition such as "standardizing using (value $-$ mean) / SD", which may help given a lack of standardization[!] on what "standardizing" means (2) noting that especial care is needed if you also refer to the Fisher $z$ or atanh transformation for correlations. But if most or all papers in a sub-sub-field use some notation, that is decisive. $\endgroup$
    – Nick Cox
    Commented Jan 27 at 9:55
8
$\begingroup$

Preferred by whom?

This is the sort of thing that is entirely style-dependent. If you are submitting to a journal, it ought to have style sheet. Many journals follow the American Psychological Association manual, because it is pretty comprehensive and covers a lot of things relevant to academic publishing. They suggest a Roman type letter with a hyphen: z-score.

I couldn't find an explicit recommendation in the Chicago Manual of Style, but they generally say to use italic for all math symbols, unless there is some special reason not to (e.g., in some cases, the same letter may mean different things in different fonts: $\text{x} \ne x$ but that doesn't seem to apply here).

If it is dissertation, consult someone in your department (there is often someone who knows and enforces every jot and tittle of the style).

$\endgroup$
1
$\begingroup$

The question arises in a draft of an IEEE standard and the IEEE style guide does not address this.

The NIST/SEMATECH e-Handbook of Statistical Methods http://www.itl.nist.gov/div898/handbook/

Uses Z-score, (Z capitalized and italic, hyphenated), so that's definitive for my purposes.

Thanks for the suggestions.

$\endgroup$
1
  • $\begingroup$ You probably already thought about it, but it might covered by a more general rule in the IEEE style guide, that does not mention specifically the Z-score. Otherwise, having a look at a couple of recent IEEE standards that use mathematical/statistical notation might also give you a general idea of how to proceed. Nethertheless, following the NIST/SEMATECH e-Handbook rules is a good idea if it's what is used in your field and you can't really find anything about the IEEE rules. $\endgroup$
    – J-J-J
    Commented Jan 27 at 18:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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