Numbers in Journalism
Numbers crop up all the time in journalism so it is important to know how to use them.
It is also important to remember that numbers are just words, and as such carry meaning. Just as a journalist you take care with the words you use, you should also take are with the numbers too.
Numbers one to nine are written as words, and 10 and more are written as digits.
When writing numbers greater than 999 split the digits into groups of three separated by a comma (not a space) such as 1,999, or 25,000, or 128,282,597. The exception is street numbers and years that do not take the comma separator.
Some news organisations do not use the comma separator on four digit numbers but this could lead to some confusion, especially with years. The Ultimate Guide's advice is to be consistent and use the comma separator in all cases greater than 999.
In text media write the number to as many significant digits as possible while preserving some readability, using any rounding sparingly and with caution.
In broadcast write out a number as you would say it. So 10.62 becomes "ten point six two", or 6,000,000 becomes "six-million", or 500,000 becomes "five hundred thousand" or even "half a million".
If a number is used at the start of a sentence then spell it out in words.
Most numbers in journalism need a unit or quanitfier to help give the number meaning.
Australia uses the International System of Units (SI) or metric system of measurement of kilograms, metres and seconds, although some non-SI units are also accepted.
Details on how those units are applied are governed by the National Measurement Institute.
The US and some other countries still used pounds, feet and gallons so it is important to know how to convert between the various units.
Writing large numbers
When it comes to writing a large number in journalism it is best to make it both as simple and as accurate as possible.
The table below may help you express your number using either the scientific notation or prefix notation accepted in Australia, and the accepted English word version of some large numbers.
It used to be the case that a billion was a million-million but then the US came along and downgraded a billion to just a thousand-million. That expression has gained acceptance now in Australia.
A trillion is now a million-million (or a thousand-billion), and is cropping up more in news stories, usually in government budget figures and computer memory.
Expressing numbers greater than a trillion can be tricky so it is best to stick to factors of a million, billion or trillion. Beyond that and you could opt for the scientific notation where one billion becomes 1x109.
Care should be taken with some British figures, especially in older documents, where the older British definitions of a billion and trillion etc may still be used. Again, that is why the scientific notation is sometimes the best to avoid any ambiguity.
|Number in full||Scientific Notation||Prefix||Symbol||English|
|1... (continues for googol '0's)||10googol||-||-||googolplex|
|10 ... (continues for 100 '0's)||10100||-||-||googol|
|0.000 000 001||10-9||nano||n||billionth|
|0.000 000 000 001||10-12||pico||p||trillonth|
|0.000 000 000 000 001||10-15||femto||f||quadrillionth|
|0.000 000 000 000 000 001||10-18||atto||a|
|0.000 000 000 000 000 000 001||10-21||zepto||z|
|0.000 000 000 000 000 000 000 001||10-24||yocto||y|
When doing calculations using a computer it is important to bear in mind accuracy.
Computers are not perfect calculating machines as they have limitations. This may not be a problem for most of the typical calculations you come across as a journalist, but it is best to be aware of those limitations.
For example, the Calculator program available on most Windows operating systems says it is only accurate to 32 digits. That means it only stores 32 digits of any number, regardless of where the decimal point occurs.
So any digits in a number beyond 32 will be lost, which includes any digits in any irrational numbers such as pi (π).
While this should not be a problem for most simple calculations, the Calculator program warns that any repeated calculations involving such numbers may lead to a loss in accuracy.
Other computer programs have their own limitations when dealing with the accuracy of number calculations and rounding errors. The thing to remember about such errors is that they add up.
It is also important to be aware of any errors (if any) involved in any measurement of a unit to be converted, especially when converting from small to large units, or vice versa.
However, the idea of having a range of units is to aid the expression of a number with a suitable unit. Expressing an astronomical distance is far better in leap years than say millimetres.
The conversion tools available on the Ultimate Guide allow you to include any (+/-) error in involved in the measurement of the original unit.
If you want to convert one unit to another, with no error in measurement involved, then go for it. For example, a road sign in Brisbane says Sydney is 986km, with no error given, so that converts to about 612.67 miles. (See rounding.)
But if you are converting a measurement you know is accurate only within a given (+/-) error then you can include that error margin in the conversion tool.
Say you have measured a distance using a metre rule that is marked only down to millimetre markings, then you can probably say at best your measurement is accurate to about +/-0.5mm.
Say you measured something 750mm +/-0.5mm and want to convert that to inches. You conversion gives 750mm as 29.527559055118 inches.
Your +/-0.5mm error margin becomes +/-0.019685039370079, or about 0.07 per cent to two decimal places.
Rounding, it would be fair to say the conversion was accurate to +/-0.02 inches so your final calculation would be to say 750mm +/-0.5mm is about 29.52 inches +/- 0.02 inches.
Rounding is often used to truncate a number to a certain number of significant digits or decimal places to make it easier to understand. Rounding can be useful to a journalist in making a number easier to read, or say in broadcast.
In the example above the number 29.527559055118 would be difficult to read in text and very hard to say in broadcast, so it is probably best expressed to just two decimal places at 29.53.
The basic rule on rounding is to look at the last significant digit you wish to keep in any number, then look at the next digit.
If that number is 5 or greater then you increase the last significant digit number by 1 (called rounding up), if the number is less than 5 then you leave the number as it is (rounding down).
So rounding 29.527559055118 to one significant decimal place it becomes 29.5, rounding to integer value only it becomes 30.
The same rule applies to large numbers and is used often in rounding large sums of money.
So given a budget of $250,842,234 then rounding to four significant digits this would become $250,800,000. Rounding to three significant digits this would become $251,000,000, or $251-million.
Be careful when rounding too much though as to say $926,768 is $1-million has suddenly added $73,232, or about 8 per cent, to the value. Why not say $929,000?