Significant figures rounding, explained without the textbook
Significant figures are the digits in a number that carry actual meaning. They communicate how precisely you measured something. Saying "the cup contains 250 mL of water" is different from "the cup contains 250.0 mL of water" — the second sentence claims you measured to the nearest tenth of a milliliter. Wrong sig fig handling is the most common avoidable mistake in chemistry homework, physics labs, engineering measurements, and finance reporting.
The four rules in 90 seconds
Rule 1: All non-zero digits are significant. The number 235 has three sig figs. The number 4287 has four. Easy.
Rule 2: Zeros between non-zero digits are significant. The number 1002 has four sig figs. The number 50.04 has four. These zeros are "trapped" between meaningful digits, so they count.
Rule 3: Leading zeros are NOT significant. The number 0.0042 has only two sig figs (the 4 and the 2). The leading zeros are just placeholders telling you where the decimal point sits. They don't add precision.
Rule 4: Trailing zeros count only if there's a decimal point. The number 305.040 has six sig figs (the trailing zero after the decimal is significant). The number 305040 (no decimal) is ambiguous — it could have 5 or 6 sig figs, and the convention depends on context. To avoid ambiguity, write it as 3.0504 × 10⁵ (5 sig figs) or 3.05040 × 10⁵ (6 sig figs).
How rounding actually works
Once you know how many sig figs to keep, the rounding rule is simple: look at the digit just to the right of where you're cutting off. If it is 5 or more, round up. If it is 4 or less, round down. Replace any digits to the right with zeros if they're to the left of the decimal point, or drop them if they're to the right. For example, 305.459 rounded to three sig figs becomes 305 (the 4 rounds down). Rounded to four sig figs it becomes 305.5 (the 5 rounds up).
Why this calculator has seven modes (most have one)
The half-up rule (round 0.5 up) is the most common but it's not the only one. Half-up has a small upward bias: if you average a million rounded numbers, the average will be slightly higher than the true average because 0.5 always rounded up. For most homework that doesn't matter. For finance and statistics, it does.
Banker's rounding (half-to-even) fixes this. When the digit is exactly 0.5, you round to the nearest even number instead of always up. So 2.5 → 2 (down to even), 3.5 → 4 (up to even), 4.5 → 4 (down to even). Over many numbers the rounding errors cancel out. This is the IEEE 754 default and what Python's round() function does. Many financial systems and scientific instruments use it.
Half-away-from-zero is what most school textbooks teach for negative numbers. -2.5 rounds to -3, not to -2. It feels more intuitive when working with absolute values.
Half-down is the opposite of half-up: 0.5 always rounds down. Rare but occasionally used.
Ceiling always rounds up regardless of the next digit. 2.1 → 3 if you're going to 1 sig fig. Useful for "worst case" estimates.
Floor always rounds down. Useful for "guaranteed minimum" estimates.
Truncate just chops off the digits without rounding at all. 2.999 truncated to 1 sig fig is just 2. This is technically NOT rounding, but it's a common operation in low-level programming.
Sig figs vs decimal places
People confuse these constantly. Sig figs care about the precision of the entire number. Decimal places care about position relative to the decimal point. The number 305.459 rounded to 3 sig figs is 305 (we kept three meaningful digits). The same number rounded to 3 decimal places is 305.459 (we kept three places after the decimal — which happens to be all of them). For 0.0042195: 3 sig figs gives 0.00422; 3 decimal places gives 0.004. Very different answers.
Scientific notation makes sig figs explicit
One of the cleanest ways to communicate sig figs is to write the number in scientific notation. 4500 is ambiguous: it could have 2, 3, or 4 sig figs depending on whether the trailing zeros are precise or just placeholders. But 4.5 × 10³ is unambiguously 2 sig figs. 4.50 × 10³ is 3 sig figs. 4.500 × 10³ is 4 sig figs. Every textbook recommends scientific notation when sig figs matter, which is why this calculator outputs all three formats: standard, scientific with × 10ⁿ, and e-notation.
The most common student mistakes
Mistake 1: Rounding intermediate results in a multi-step calculation. Always carry extra sig figs through every step and only round at the end. If you round each step, the errors compound.
Mistake 2: Confusing sig figs with decimal places (see above).
Mistake 3: Not knowing that the number 100 has only 1 sig fig (technically), not 3. To indicate that the trailing zeros ARE significant, write it as 100. (with the decimal point) or 1.00 × 10² (scientific notation).
Mistake 4: Treating exact numbers like counts the same as measured numbers. The number of students in a classroom (counted: 23) has infinite sig figs and never rounds. The measured height of those students absolutely needs sig figs.
Pairing with other JustDownSize tools
After rounding your data, the marks to percentage calculator handles grade calculations to the right precision. The GPA calculator and CGPA to percentage tools cover the academic-grade math that benefits from consistent sig fig handling. For physical science conversions, the height converter uses sig fig appropriate rounding by default.