TL;DR: A basic calculator has no fraction key, so you enter a fraction as division in brackets: 1/2 becomes (1÷2). In a scientific calculator you may have a fraction template or an a b/c key. In both cases the operation you are doing matters more than the key: 1/2 + 1/3 must be entered as (1÷2)+(1÷3), because entering it left to right without brackets computes something else entirely.
Why basic calculators cannot do fractions
A four-function calculator computes one operation at a time and shows a decimal result. There is no data type for a fraction, so 1/2 is not stored as one half — it becomes 0.5 the moment you press equals.
That is not a limitation to work around cleverly; it is a design decision that makes a cheap device behave predictably. The consequence is that any fraction arithmetic you do on a basic calculator is decimal arithmetic, with the rounding that implies.
The practical rule: enter fractions as division, and add brackets so the division happens before the addition. Without brackets, a calculator that evaluates as you type will compute 1 ÷ 2 + 1 ÷ 3 in a different order from one that supports precedence, and the two devices will disagree.
The bracket method, step by step
Take 1/2 + 1/3.
Press (, then 1, ÷, 2, ), then +, then (, 1, ÷, 3, ), then equals. The result is 0.8333…, which is 5/6.
Three details make this reliable.
Put brackets around every fraction, not just the first. Most errors come from bracketing one side and leaving the other bare.
Use the ÷ key, not the / on a keyboard. On a physical calculator they are the same operation, but on software calculators the slash sometimes starts a function command instead.
Convert mixed numbers to improper fractions first. 2½ is 5/2, entered as (5÷2). Entering it as 2 + (1÷2) works only if the surrounding expression is bracketed correctly, and it usually is not.
If your calculator has a memory or parenthesis stack, write the whole expression once rather than chaining partial results. Chained partial results are where rounding accumulates: rounding 1/3 to 0.33 before adding introduces an error larger than any display precision issue.
Scientific calculators with a fraction key
Scientific models usually have either a dedicated fraction key or a template that draws a stacked fraction on the display.
The a b/c key enters mixed numbers as three parts: whole number, numerator, denominator. Pressing it between numbers moves the cursor between the parts. It also toggles the display between decimal and fraction after a result, which is the fastest way to see that 0.8333… is 5/6.
The n/d key is the two-part version: numerator, then denominator, with no whole number field. Use it for proper fractions and convert mixed numbers yourself.
Template keys draw a fraction placeholder and let you tab between numerator and denominator. These are the least error-prone, because the display shows the structure you are entering rather than relying on key order.
One caveat with fraction modes: some models set a maximum denominator, and results that exceed it are displayed as decimals instead of fractions. If a result looks wrong, check that setting before assuming an input error.
The four errors that produce wrong answers
These account for almost every “my calculator is broken” report.
Missing brackets. 1÷2+1÷3 evaluated as (1÷2)+1, then ÷3, gives 0.5 rather than 0.833. Order of operations is not universal across devices.
Fraction as a mixed number without conversion. 3½ entered as 3 × 1 ÷ 2 gives 1.5 instead of 3.5.
Degrees versus radians. Not a fraction issue, but the most common cause of a wildly wrong trigonometric result, and it looks like a fraction error because the answer is a strange decimal.
Rounding mid-calculation. Writing down 0.33 and re-entering it introduces an error of 0.0033 per occurrence, which compounds fast across several fractions.
Keeping the full expression in one input, with brackets, eliminates the first and the last.
Converting a decimal answer back to a fraction
Often the result is a decimal and you need the fraction. Three approaches, in order of reliability.
Use the fraction toggle if the calculator has one. It converts the display exactly, within the device’s maximum denominator.
Recognise common values. 0.5 is 1/2, 0.333… is 1/3, 0.666… is 2/3, 0.25 is 1/4, 0.125 is 1/8. These cover most practical cases and cost nothing.
Multiply by the denominator. For a repeating decimal, multiply by 9, 99, or 999 depending on the repeating block length, then simplify. 0.666… × 9 = 6, so 6/9 = 2/3. This is the exact method rather than an approximation, and it works for any repeating decimal.
The third method is worth knowing because it explains why the decimal and fraction are the same number, which is what most people actually want to understand.
When a plain calculator is the wrong tool
If the answer must stay exact — algebra, ratio work, probability, or anything where a decimal would lose information — a four-function calculator cannot get you there, because it has already discarded the fraction.
For exact fraction arithmetic, including simplification and mixed-number output, a fraction calculator keeps the numerator and denominator as integers throughout, which is the only way to guarantee the result is exact rather than approximately right.
The short version
Enter fractions as bracketed division: (1÷2). Bracket every fraction, convert mixed numbers to improper fractions, and keep the whole expression in one input. Scientific models add a/b/c and n/d keys that toggle between decimal and fraction, and a repeating decimal can be converted exactly by multiplying by 9, 99, or 999.
