I realize this is pedantic and trite, but as a researcher in a field outside of statistics, with limited formal education in statistics, I always wonder if I'm writing "p-value" correctly. Specifically:

  1. Is the "p" supposed to be capitalized?
  2. Is the "p" supposed to be italicized? (Or in mathematical font, in TeX?)
  3. Is there supposed to be a hyphen between "p" and "value"?
  4. Alternatively, is there no "proper" way of writing "p-value" at all, and any dolt will understand what I mean if I just place "p" next to "value" in some permutation of these options?
  • 1
    $\begingroup$ See the meta thread: meta.stats.stackexchange.com/questions/213/… where this question is proposed to be closed. $\endgroup$
    – user28
    Jul 30 '10 at 2:00
  • 1
    $\begingroup$ Here's what I want to know: If one is using a lower-case "p" in "p-value" or "p value", should the "p" be capitalized if it's at the beginning of a sentence or a section heading? (I realize this is rare, but it can occur when one is discussing p-values as such.) $\endgroup$
    – Mars
    Jul 13 '18 at 17:56

There do not appear to be "standards". For example:

  • The Nature style guide refers to "P value"
  • This APA style guide refers to "p value"
  • The Blood style guide says:
    • Capitalize and italicize the P that introduces a P value
    • Italicize the p that represents the Spearman rank correlation test
  • Wikipedia uses "p-value" (with hyphen and italicized "p")

My brief, unscientific survey suggests that the most common combination is lower-case, italicized p without a hyphen.

  • $\begingroup$ Wow! I didn't expect this to vary. BTW, has the Nature style guide changed since you posted this answer? It's now pointing to "Life sciences reporting guidelines". $\endgroup$ Oct 30 '13 at 1:35
  • 6
    $\begingroup$ It becomes more and more common to write "p-value" or "$p$-value": small italicized "p" with a hyphen. Earlier this year American Statistical Association (ASA) issued a statement on $p$-values; they used this spelling. Wikipedia uses this spelling too. I think your conclusion about what is the most common choice might be a bit outdated by now. $\endgroup$
    – amoeba
    Dec 9 '16 at 20:56
  • $\begingroup$ The Blood style guide is particularly odd since that's not actually supposed to be a "p" for the Spearman rank correlation test, but a lowercase rho (ρ)... $\endgroup$
    – Patrick B.
    Jan 23 '17 at 23:25

This seems to be a style issue with different journals and publishers adopting different conventions (or allowing a mixed muddle of styles depending on authors' preferences). My own preference, for what it's worth, is p-value, hyphenated with no italics and no capitalization.


The ASA House Style seems to recommend italicizing the p with hyphen: p-value. A google scholar search shows varied spellings.


P value from theoretical point of view is some realization of random variable. There is some standard (in probability) to use upper case letters for random variables and lower case for realizations. In table headers we should use P (maybe italicize), in text together with its value p=0.0012 and in text describing for example methodology p-value .

  • $\begingroup$ So in a table header, something like "Student's $T$" would be used for the statistic and "$P$" for the significance? I'm a bit worried that nobody will get this subtlety and people will just assume a stupid typo ... $\endgroup$
    – Christian
    Feb 27 '13 at 11:57

Omitting the hyphen can sometimes change the meaning of sentences or at least they can become ambiguous. This can occur especially in papers that describe statistical tests or introduce algorithms to evaluate p-values, but one may also describe methods that have nothing to do with statistics, and still calculate p values from t tests (but not the p-values using statistical t-tests). In this kind of context, the hyphens would really be necessary, even if writers usually try to avoid notations that could get easily confused.

Example (with a bad choice of notations): We would like to find a set of strong association patterns and evaluate the probability that the result would have occurred by chance. In the first phase, we search for the z best patterns with some goodness score. So, after the search phase, we will have z scores (but the z-scores). Then we evaluate the best patterns with a randomization test. We generate t random data sets and evaluate the score of the z:th best pattern in each data set. So, we perform t tests (but not the t-tests) and output the score of the z:th best pattern. We find out that p values (but not the p-values) of all t score values are better then the original z:th best pattern had. Therefore, we could estimate that the probability of getting z so good patterns by chance is p/t.


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.