Guidelines for Reporting Statistics

JMIR Publications employs professional external copyeditors (see What are the steps during copyediting? and What are the authors' responsibilities during copyediting?) to bring articles into editorial style, so authors do not have to worry too much about the style on initial submission, although it does shorten the production process after acceptance if the article is roughly aligned with our guidelines (see also JMIR's editorial guidelines), and it helps us to find reviewers if the references are properly formatted and have PMIDs.

While pointing out issues in statistical reporting is also a responsibility of the editor/section editor—who should point out missing statistics and incorrectly reported statistics already during the review process (and not rely on external peer-reviewers)—copyeditors act as the "second line of defense" and must enforce reporting in line with generally accepted guidelines. The recent SAMPL Reporting Guideline is a useful resource that should be known and enforced by all copyeditors.

General notes:

• When reporting multiple statistics within brackets, separate each statistic with a semicolon. For example: (P=.50; OR 2.72, 95% CI 0.45-2.6).

• Do not use possessives for the name of any statistical test (see Use of possessives with eponyms).

 

 

Omission of Leading Zero

We follow the guidelines set out by AMA (Section 19.7.1)

AMA states that a zero should be placed before the decimal point for numbers less than 1, except when expressing the 3 values related to probability: P, α, and β. These values cannot equal 1, except when rounding. Because they appear frequently, eliminating the zero can save substantial space in tables and text. (Although other statistical values also may never equal 1, their use is less frequent; to simplify usage, the zero before the decimal point is included.)

Examples:

P=.16

1 − β=.80

Our predetermined α level was .05.

BUT: κ=0.87

 

 

Percentages and Decimal Places

• If N < 100, there is no decimal point in the percentage.
• If N is 100 to 999, 1 decimal point is reported.
• If N ≥ 1000, 2 decimal points are reported.
• BUT: If a table contains mixed denominators, be consistent and use, for example, 1 decimal point consistently even if some denominators are less than 100.
• See AMA 2.13.9, 17.3.1, and 17.3.2.

N=87, 45%

N=356, 45.1%

N=1024, 45.13%

Note: Do not use n=488/550, 88.7%; instead, it should be only 488/550 (88.7%).

 

 

Percentages Within a Sentence

Preferred JMIR style is to always make clear what the numerators and the denominators are.

In expressing a series of proportions or percentages drawn from the same sample, the denominator need be provided only once.

Of the 200 patients, 6 (3%) died, 18 (9%) experienced an adverse event, and 22 (11%) were lost to follow-up.

Example I:

"the majority of participants (59%) felt that..."

should be revised to

"the majority of participants (59/100, 59.0%) felt that...." NOT “the majority of participants (59.0%, 59/100) felt that...."

or if the percentage is used in the sentence,

"…, where 59.0% (59/100) of the participants felt that…".

Example II:

"a vast majority (n=488, 88.7%) of participants"

should be changed as follows:

"a vast majority (488/550, 88.7%)" [Note: The "n" has been dropped and the "N" value (ie, 550) has been added]. When reporting multiple statistics within brackets, separate each statistic with a semicolon.

Example III:

“a vast majority (488/550, 88.7%; P=.002)”

Do not use square brackets within round brackets for statistics. Separate values with a comma if statistics are linked (ie, % and numerator/denominator); separate values with a semicolon if statistics are unlinked (ie, % and odds ratio).

WRONG:

"The most common functions among studies that involved children with special needs were consultation (8 studies [73%]) and diagnosis (7 studies [64%]). "

 

 

Chi-square test

• Include the degrees of freedom (subscript).
• Authors tend to not include degrees of freedom; query the author if it is not provided.
• Degrees of freedom are subscripted.
• Chi-square value should be reported to only 1 decimal place.
• When reporting Chi-square values in a table, include the degrees of freedom in brackets; eg, Chi-square (df).
• The abbreviation for degrees of freedom (df) does not need to be expanded; it should be italicized (AMA glossary of statistical terms 20.1).

Example:

 

 

Mean, standard deviation, standard error, and range

• Equal signs are not used; separate the value from the statistic with a space.

Examples

mean 4.71 (SD 0.47)

range 4-5

SE 2.55

When reporting multiple statistics in a sentence, use a semicolon to separate the terms.

Example: “The mean age of the participants (mean 4.71, SD 0.47; range 4-5)...”

For mean (SD), we prefer not to use the +/- sign. Instead, an expression like 1.11 ± 2.33 should be formatted as "1.11 (SD 2.33)." It is also wise to add a query notifying the author(s) of this change, in case there are also other statistics that have also been formatted by authors with a +/- sign (eg, SE) and they now need to specify the expression.

When reporting a value that is calculated from a mean and SD value, report it in the following manner: Mean+SD=1.6

Note: When reporting mean and SD in a table, include the mean and SD in the same column with the heading “mean (SD)”.

As of December 4, 2013, AMA no longer requires the expansion of "SD" or "SE" in the text (Section 20.9, page 894 in the print) — See Which abbreviations don't need to be expanded?

 

Odds ratio and confidence interval

• ORs should always be presented with CIs
• Include 0 before the decimal point
• If one value in the CI range is negative, then “to” should be used rather than a hyphen

OR 1.2 (95% CI 0.9-2.4)

OR 1.2% (95% CI 0.8%-1.6%)

95% CI –0.1 to 0.8

• Note: For consistency, use the "x to y" format for all CIs when some include a negative number; ie, use the "to" construction for positive ranges as well if it needs to be used for a negative. This is most important in tables.
• Avoid brackets within brackets. If brackets within brackets are necessary, use square brackets We never use round brackets within round brackets. Either avoid the brackets inside of brackets altogether, i.e. (OR 2.92 (2.36 - 3.62)) should be rewritten as (OR 2.92, 95% CI 2.36-3.62), or use square brackets inside of round brackets (OR 2.92 [95% CI 2.36-3.62]). The former is preferred.
• CI does not need to be expanded (Section 14.11, page 504 in the print; see Which abbreviations don't need to be expanded? ).
• When defining “OR” within parentheses, use square brackets. Example: (odds ratio [OR] 2.92, 95% CI 2.36-3.62). Note that OR needs to be defined in tables (through a footnote or in the caption).

Confidence limit

Upper and lower boundaries of the confidence interval, expressed with a comma separating the 2 values.

Example: The mean (95% confidence limits) was 30% (28%, 32%).

 

Interquartile range

• Include 0 before the decimal point.
• Should be formatted as: IQR 5 (ie, no equal sign).
• At first mention in the text, the expanded form should be used alongside the acronym; IQR should be added to the Abbreviations List.
• In tables, a footnote should be added to explain the acronym.

 

P value

From our instructions for authors:

 

• (Again, this is the primary responsibility of the academic editor, but the copyeditor acts as a second line of defense if this has been overlooked by the editor/section editor)
• Note for copyeditor: point author to the relevant section in the Instructions for Authors, if P values are missing (ie, "no significant differences were found..." without stating the P-level), incorrectly reported, or replaced by statements of inequality (or in Tables * / ** footnotes) such as P<.05.
• The actual P value should be expressed rather than a statement of inequality (P=...), unless P<.001 or P>.99 or P=0 (which should be changed to P<.001) or P=1 (which should be changed to P>.99). In other words, for expressions like P<.05, authors should be queried to provide the actual P value.
• P values cannot be 0 or 1—change to <.001 or >.99, respectively. Note for copyeditor: add a comment to the manuscript in copyediting saying something like “Because P values theoretically cannot reach 1, AMA Manual of Style guidelines are that the highest P value to report is ‘P>.99’ so I have changed them accordingly” (and similar for P=0).
• P values less than .001 (including 0.000) are not allowed and are to be converted to the expression P<.001. In other words, for example, P<.0001 or P=.0005 must be rewritten as P<.001.
• P is italicized and capitalized (Note: Our typesetting scripts are actually intelligent enough to convert the P to italics if the copyeditor forgets this).
• DO NOT use zero before the decimal point.
• The actual value of P should be expressed to two digits, whether or not it is significant.
• If P<.01, P should be expressed to three digits.
• When rounding, 3 digits are acceptable if rounding would change the significance of a value (eg, P=.049 rounded to .05).
• No blanks in expressions. Our typesetting scripts now automatically remove these blanks to our preferred format without blanks. So please do not spend time inserting or removing blanks. The same is true for other equality and inequality expressions (eg, P < .001 is changed automatically during typesetting to P<.001 and n = 12 is corrected to n=12 automatically during typesetting.
• In a table, the column heading should be “P value” not “P”. Note that "value" should not be italicized.

Examples:

P=.60

P=.37

P=.02

P=.008

P=.049

P<.001

 

NOTE for P values for very large sample sizes (according to AMA guidelines)

Excerpt from AMA: “Though our style manual recommends (Section 20.9, page 888 in the print) that "[expressing] P to more than 3 significant digits does not add useful information to P<.001," in certain types of studies (particularly GWAS [genome-wide association studies] and other studies in which there are adjustments for multiple comparisons, such as Bonferroni correction, and the definition of level of significance is substantially less than P<.05) it may be important to express P values to more significant digits. For example, if the threshold of significance is P<.0004, then by definition the P value must be expressed to at least 4 digits to indicate whether a result is statistically significant. GWAS express P values to very small numbers, using scientific notation. If a manuscript you are editing defines statistical significance as a P value substantially less than .05, possibly even using scientific notation to express P values to very small numbers, it is best to retain the values as the author presents them.”

If a study has a very large sample size, it may be necessary to report P values to a value smaller than P<.001 in order to show statistical significance. This will be up to editorial discretion.

 

N and n

• N designates the entire population under study.
• n designates a sample of the population under study.
• Do not insert spaces before and after the sign. Our typesetting scripts now automatically remove these blanks. So please do not spend time inserting or removing blanks. 

Examples:

N=468

n=234

 

 

F test

• Degrees of freedom are subscripted in the text.
• In a table, degrees of freedom are included in brackets after the number; ie, “F test (df)”

Example:

 

t test

• t is italicized.
• Include the degrees of freedom (subscript).
• Include 0 before the decimal point.
• Authors must include whether the test is 1-tailed or 2-tailed
• Authors tend to not include degrees of freedom; query the author if it is not provided.

Example:

 

Effect size

• Include 0 before the decimal point.

...an effect size of 0.277 standard deviation units.

 

Cronbach alpha

• DO NOT use “Cronbach’s” alpha.
• Spell out “alpha”; DO NOT use the Greek letter.
• There is no zero before the decimal point.

Example:

Cronbach alpha=.78

 

Cohen d

• DO NOT use “Cohen’s” d with the possessive
• Include a zero before the decimal point if the value is less than 1.

Example:

Cohen d=0.29

Cohen d=1.45

 

beta level

• Spell out “beta”; DO NOT use the Greek letter.
• There is no zero before the decimal point.

Example:

beta=.2

 

Spearman rank correlation

• The symbol is ρ (rho).
• Include a zero before the decimal point if the value is less than 1.

Example:

ρ=0.67

 

Kappa statistic

• The symbol is κ (kappa).
• Preferred to spell out in the text.
• Include a zero before the decimal point if the value is less than 1.

Example:

κ=0.50

The kappa indicating inter-observer reliability was 0.5.

 

Equal and “inequality” signs

• DO NOT insert spaces before and after the sign.
• Our typesetting scripts now automatically remove these blanks to our preferred format without blanks. So please do not spend time inserting or removing blanks. The same is true for other equality and inequality expressions, e.g. (P < .001) is changed automatically during typesetting to (P<.001) and (n = 12) is corrected to (n=12), automatically during typesetting.
• For greater than or equal to and less than or equal to signs, insert a single symbol using MS Word (Mac shortcuts ≥: option+> ; ≤: option+<; Windows shortcuts ≥: ALT+8805 ; ≤: ALT+8804)
• DO NOT use an underlined greater than / less than symbol.

Examples:

x=15

n>2

y≤0

 

Greek letters in text

• Greek letters are spelled out wherever possible.

Examples:

beta-thalassemia

Cronbach alpha

 

Currency

• Specify all dollar currencies.
• DO NOT use zeros after whole numbers of currency.

US $99

Can $125.35

€40

Aus $100 (Note: AMA says "A$" for Australian dollars, but since we use a space before the $, this would be confusing as "A $100"; therefore, we'll abbreviate to "Aus $")

 

Complex equations

Multiple numbered simple equations that are kept in-text should be left-aligned with numbering (number in bold) in parenthesis after the equation itself.

Whenever possible, the equations should be inserted in the text using characters, not formatted as inline figures.

Example:

yi = Ci - ci (1)

Do NOT use the equation editor/tools in Word. The special formatting will not be picked up by our scripts.

Use spaces between all mathematical operators in complex equations (including “=”). Note in this case, “=” is functioning as an operator, it is not a statement of equality. In the case of “n=2,” no spaces are used as it is a statement of equality.

  

 

"Trending" towards significance

We do not use "trending towards significance" or other variants.

For example, "There was a trend (P=.06) showing that...was significant".

It's fair to say that there was a trend in something, but this must be followed up with "...but these results were not statistically significant."

Alternatively, it's best to simply state the results' significance and not use variations of "trend".