Best way to report p =1.000

KevinvSchie

Hey all,

Sometimes JASP (NHST analyses) reports that p = 1.000. I believe there is no way of finding out what the exact number is, but I assume it is incredibly close to 1.000. What would be the best way to report this?

hbreitsohl

Independent of the software being used, I would assume that most manuscript guidelines have rules on how many decimals to report. At the same time, most software packages will round numbers.

When JASP gives you "P = 1.000", this really means "when rounding to three decimals, the result is 1.000".

Thus, when your manuscript guidelines tell you to report two decimals, you report "p = 1.00", implying that this is the result of rounding the actual value to two decimals.

The only point where there might be trouble is when the guidelines ask for more decimals than JASP will give you. I would then simply note that JASP only gives you three decimals, and I think that most readers would be okay with that.

Unless, there is a way of changing that setting in JASP, of course.

EJ

1. Yes, you can change the number of decimals that JASP reports in the Preferences -> Results section.
2. When p=1, sometimes it is exactly 1. This happens, for instance, when an effect size in a t-test is 0 exactly. More commonly, it will happen in a contingency table: for instance, if H0 says we have a fair coin, and we observe 50/100 hits, then all data are more extreme than the data observed, so p=1.

E.J.