Rounding Decimals: Rules and Examples

Decimal numbers often contain more digits than you actually need. For example, 6.784396 can be rounded to 6.78, 6.784, or 6.8 depending on the precision you need. The key is knowing which decimal digit to keep and which digit to check next.

Rounding decimals means replacing a decimal number with a nearby value that has fewer decimal digits while keeping the value as close as possible to the original under the selected rounding rule.

For example, 6.784396 rounded to two decimal places = 6.78.

• Target digit = 8
• Deciding digit = 4
• 4 is less than 5
• Keep 8 unchanged
• Remove the remaining digits
• Result = 6.78

A decimal number consists of a whole-number part and a fractional part separated by a decimal point.

Examples include 4.5, 12.75, 0.0048, and 125.6789. The digits to the right of the decimal point represent fractions of a whole. The first place is the tenths, followed by the hundredths, thousandths, and ten-thousandths.

Rounding reduces the number of decimal digits while keeping the result close to the original value.

Examples for the number 6.784396:

• 6.784396 \rightarrow 6.78 (two decimal places)
• 6.784396 \rightarrow 6.784 (three decimal places)
• 6.784396 \rightarrow 6.8 (one decimal place)

All three results can be correct because they use different target precisions based on what is needed.

The procedure is simple:

1. Decide which decimal place you want to keep.
2. Look at the digit immediately to its right.
3. If that digit is 0 to 4, leave the target digit unchanged.
4. If that digit is 5 to 9, increase the target digit by 1.
5. Remove the digits after the target place.

Each position after the decimal point has a specific name.

• Tenths: First digit after the decimal point. Example: 4.7
• Hundredths: Second digit after the decimal point. Example: 4.73
• Thousandths: Third digit after the decimal point. Example: 4.736
• Ten-thousandths: Fourth digit after the decimal point. Example: 4.7365

Here is a place-value breakdown for the number 12.73645:

Use this reusable five-step process:

• Step 1: Choose the decimal place.
• Step 2: Identify the target digit.
• Step 3: Look at the digit immediately to the right.
• Step 4: Apply the rounding rule.
• Step 5: Remove the remaining digits.

Example: Round 8.3764 to two decimal places.

• Target = hundredths digit 7
• Deciding digit = 6
• 6 is greater than 5
• Increase 7 to 8
• Result = 8.38

When rounding to the nearest tenth, exactly one digit after the decimal point remains.

Examples:

• 4.23 \rightarrow 4.2
• 4.26 \rightarrow 4.3
• 7.84 \rightarrow 7.8
• 7.86 \rightarrow 7.9

For 7.86, the target digit is 8 and the deciding digit is 6. Because 6 is five or greater, the 8 increases to 9, resulting in 7.9.

Rounding to the nearest hundredth means two decimal digits remain. The third decimal digit acts as the deciding digit.

Examples:

• 5.674 \rightarrow 5.67
• 5.678 \rightarrow 5.68
• 12.341 \rightarrow 12.34
• 12.347 \rightarrow 12.35

For the nearest thousandth, three decimal places are kept. The fourth decimal digit determines the result.

Examples:

• 3.14159 \rightarrow 3.142
• 7.2344 \rightarrow 7.234
• 7.2346 \rightarrow 7.235

You can round numbers to 4, 5, 6, or even more decimal places.

Using the number 6.784396:

• 4 decimal places \rightarrow 6.7844
• 5 decimal places \rightarrow 6.78440
• 6 decimal places \rightarrow 6.784396

Trailing zeros, such as the zero in 6.78440, can be important because they communicate the exact requested number of decimal places.

When dealing with numbers smaller than 1, you often see leading zeros immediately after the decimal point. These zeros are placeholders, but they still count as decimal places.

Examples:

• 0.0047 \rightarrow 0.00 to two decimal places, or 0.005 to three decimal places.
• 0.0365 \rightarrow 0.04 to two decimal places.
• 0.1256 \rightarrow 0.13 to two decimal places.
• 0.9996 \rightarrow 1.000 to three decimal places.

Leading zeros immediately after the decimal point are not the same as meaningful decimal digits when calculating magnitude, but they strictly occupy a decimal position.

Rounding works identically regardless of the whole numbers positioned to the left of the decimal point.

• 12.348 \rightarrow 12.3 (one decimal place)
• 125.6789 \rightarrow 125.68 (two decimal places)
• 1,234.5678 \rightarrow 1,234.568 (three decimal places)

To round a decimal to the nearest whole number, use the tenths digit as your deciding digit. If it is 5 or higher, increase the ones digit by 1.

Examples:

• 4.2 \rightarrow 4
• 4.6 \rightarrow 5
• 12.3 \rightarrow 12
• 12.8 \rightarrow 13

The tenths digit completely determines the result when rounding to the nearest whole number.

Exact midpoint cases like 2.25, 3.15, and 4.35 require a specific tie-breaking rule.

The result depends on the rounding method you are using. Common methods include Half Up, Half Down, Half Even, and Half Away From Zero. Different methods can produce different answers for these midpoints. You can learn more about how Half Even balances these cases in our guide to Banker’s Rounding.

The standard rounding method used across most of RoundSolver is Round Half Up. Under this rule, exact midpoints always move in the positive direction toward positive infinity.

Examples to one decimal place:

• 2.24 \rightarrow 2.2
• 2.25 \rightarrow 2.3 under Half Up
• 2.26 \rightarrow 2.3

Examples to two decimal places:

• 3.144 \rightarrow 3.14
• 3.145 \rightarrow 3.15 under Half Up
• 3.146 \rightarrow 3.15

Banker’s Rounding, officially known as Round Half to Even, resolves exact midpoints by looking at the target digit. If the target digit is even, it stays the same. If it is odd, it increases to become even.

Examples:

• 2.25 \rightarrow 2.2 (the target 2 is already even)
• 2.35 \rightarrow 2.4 (the target 3 is odd, so it increases)
• 2.45 \rightarrow 2.4
• 2.55 \rightarrow 2.6

The retained digit determines whether the midpoint rounds to a higher or lower value. For more details, explore our full Banker’s Rounding guide.

Rounding negative decimals requires thinking about the number line.

Examples:

• -2.4 \rightarrow -2
• -2.6 \rightarrow -3
• -7.3 \rightarrow -7
• -7.8 \rightarrow -8

Always distinguish standard nearest-value rounding from other mathematical functions. Ceiling always moves toward positive infinity. Floor always moves toward negative infinity. Truncation strictly cuts off the decimal toward zero. Away From Zero strictly pushes the value outward on the number line.

When the target digit is 9 and needs to be increased, it becomes 0 and carries 1 to the next digit to the left.

Examples:

• 2.99 \rightarrow 3.0 when rounding to one decimal place.
• 9.999 \rightarrow 10.00 when rounding to two decimal places.
• 1.9996 \rightarrow 2.000 when rounding to three decimal places.

The carry behaves exactly like standard addition.

Consider the values 2.5, 2.50, and 2.500.

The numerical value is the same in all three cases. However, the displayed decimal places communicate different precision or formatting requirements. Writing 2.50 explicitly shows that the number is formatted to the hundredths place, whereas 2.5 only shows precision to the tenths place.

These are two different ways to measure precision. Decimal places count every digit after the decimal point. Significant figures count meaningful digits starting from the first non-zero digit.

Example using 0.004567:

• To 2 decimal places: 0.00
• To 2 significant figures: 0.0046

This is an important distinction. Learn more with our significant figures counter or our decimal places calculator.

Significant figures ignore leading zeros.

Examples:

• 45.678 \rightarrow 46 to 2 significant figures
• 0.004567 \rightarrow 0.0046 to 2 significant figures
• 123.456 \rightarrow 123 to 3 significant figures

Significant figures are not the same as decimal places because they track total accuracy magnitude rather than a fixed column position.

Rounding changes the original value, creating a mathematical discrepancy.

Example: 6.786 \rightarrow 6.79

• Signed difference: 6.79 - 6.786 = 0.004
• Absolute error: 0.004

You can read a detailed explanation of this topic in our article on rounding error.

For nearest-value rounding, the maximum absolute error is half the rounding increment.

Examples:

• Nearest tenth \rightarrow 0.05
• Nearest hundredth \rightarrow 0.005
• Nearest thousandth \rightarrow 0.0005
• Nearest ten-thousandth \rightarrow 0.00005

This strict rule specifically applies to nearest-value rounding.

Financial systems often require rounding to two decimal places.

For instance, a tax calculation of \14.994 usually becomes \14.99 under a method that rounds this value downward. A midpoint example such as \14.995 becomes \15.00 under Half Up.

Currency systems can define their own specific rounding rules, dictating whether intermediate calculations or only final totals are rounded.

Decimal precision matters deeply in measurements, experiments, engineering, and scientific reports.

You must distinguish mathematical rounding, measurement uncertainty, and significant figures. Rounding is a math rule, whereas measurement uncertainty is a physical reality of the instruments used. Rounding does not create or determine measurement uncertainty, but it is used to correctly format the reported results.

Programming languages may use different rounding behavior for their default rounding functions. Some default to Half Even, while others use Half Up or Half Away From Zero.

Additionally, binary floating-point representation means that seemingly simple decimal values such as 0.1 cannot be stored perfectly in memory. Clearly distinguish these intentional decimal rounding rules from underlying binary floating-point representation quirks.

• Checking the wrong digit: Always evaluate the digit immediately to the right.
• Rounding the deciding digit: The deciding digit is dropped, only the target changes.
• Forgetting to carry: Increasing a 9 requires carrying a 1 to the left.
• Confusing decimal places and significant figures: These follow entirely different counting rules.
• Removing important trailing zeros incorrectly: Sometimes formatting requires preserving them.
• Mishandling negative values: Look at the number line carefully.
• Assuming every 5 rounds upward under every method: Different rules handle midpoints differently.
• Rounding intermediate values unnecessarily: Keep precision until the final step.
• Double rounding: Rounding in stages creates compounding errors.
• Assuming software uses the same rounding method: Languages behave differently.

Rounding a number multiple times in succession is called double rounding.

Example using 3.449:

• Directly to one decimal place: 3.4
• First to two decimal places: 3.45
• Then to one decimal place: 3.5

The two processes can produce different results. Always round directly from the exact original value.

Our decimal places calculator makes precision tasks simple.

To use it:

1. Enter the decimal number.
2. Choose the number of decimal places.
3. Select the rounding method if available.
4. Calculate the result.
5. Review the rounded value.

You can also test these concepts using our main rounding calculator.