Calculadora de error porcentual
Calcula el error porcentual entre un valor experimental y el valor real, o la diferencia porcentual entre dos medidas, con el desarrollo paso a paso.
Cómo usar esta calculadora
- 1Choose percent error if one value is a known accepted figure, or percent difference if you are comparing two equally valid measurements.
- 2Enter your measured value and the true or comparison value.
- 3For percent error, choose signed output if you need to know whether you over- or under-measured.
Cómo funciona
Error porcentual y diferencia porcentual
Error porcentual = |medido − real| ÷ |real| × 100 Error con signo = (medido − real) ÷ |real| × 100 Diferencia porcentual = |a − b| ÷ ((|a + b|) ÷ 2) × 100 Error relativo = |medido − real| ÷ |real|
El error porcentual expresa cuánto se aleja una medida del valor real aceptado, como porcentaje de ese valor. Cuando no hay un valor de referencia y solo comparas dos medidas, se usa la diferencia porcentual, que las divide por su media. El error relativo es la misma idea sin multiplicar por 100.
Ejemplo resuelto
Si mides 9,8 y el valor real es 10, el error porcentual es |9,8 − 10| ÷ |10| × 100 = 2 %. El error absoluto es 0,2 y la exactitud, un 98 %.
Calculadora de error porcentual: la guía completa
Percent error, percent difference, and percent change
These three are constantly mixed up, and they answer different questions. Percent error asks how far a measurement is from a known correct value, so it divides by the true value. Percent difference asks how far apart two equally credible measurements are, so it divides by their average. Percent change asks how much a quantity moved over time, so it divides by the starting value and keeps its sign.
Pick the denominator that matches the question. Using the measured value as the denominator for percent error is a common mistake that makes large errors look smaller than they are.
What counts as an acceptable error
There is no universal threshold — it depends entirely on the precision of the instrument and the tolerance of the application. In an introductory physics lab, under 5% is usually considered a good result and under 1% is excellent. Analytical chemistry often demands better than 0.5%. Manufacturing tolerances can be tighter still.
What matters more than the number is whether the error is random or systematic. Random error scatters around the true value and shrinks as you average repeated trials. Systematic error — a miscalibrated scale, a consistently late stopwatch trigger — pushes every trial the same direction and averaging will not fix it. A signed percent error that stays the same sign across many trials is the classic signature of a systematic problem.
Why the true value cannot be zero
Percent error divides by the true value, so a true value of zero makes the calculation undefined — any nonzero measurement would represent an infinite proportional error. This comes up with quantities that legitimately pass through zero, such as a temperature in Celsius or a net displacement.
In that situation, report the absolute error in its own units, or switch to percent difference where the denominator is the average of the two readings. This calculator flags the zero case rather than returning a misleading number.
Preguntas frecuentes
¿Cómo se calcula el error porcentual?
Resta el valor real del medido, toma el valor absoluto, divide por el valor real y multiplica por 100. Por ejemplo, medir 9,8 frente a un valor real de 10 da |9,8 − 10| ÷ 10 × 100 = 2 %.
¿En qué se diferencian el error porcentual y la diferencia porcentual?
El error porcentual compara una medida con un valor real conocido y aceptado. La diferencia porcentual compara dos medidas cuando ninguna es la referencia; en ese caso se divide por la media de ambas en lugar de por un valor verdadero.
¿Puede ser negativo el error porcentual?
En su forma habitual (con valor absoluto) siempre es positivo. Pero el error con signo conserva el sentido: negativo si mediste por debajo del valor real y positivo si mediste por encima, lo que ayuda a detectar sesgos en las medidas.