百分位数计算器 — 查找百分位排名 & 数值

输入一组数字,即时查找任意值的百分位排名,或查询给定百分位对应的精确数据值。采用中点公式计算排名,线性插值计算数值——与 Excel 和 NumPy 使用相同方法。支持逗号、空格、分号和换行符作为分隔符。

Enter Your Data

Enter numbers separated by commas, spaces, semicolons, or newlines.

Enter a value to find what percentile rank it occupies in your dataset.

Frequently Asked Questions

什么是百分位数?

百分位数是一个值,数据集中某个百分比的观测值落在该值以下。例如,若您的考试成绩在第75百分位,意味着75%的考生成绩低于您。第50百分位即中位数(数据集的中间值)。

百分位数和百分比有何不同?

百分比是某部分占总体的比例(如得了85分满分100分,得分率85%)。百分位数是相对于其他人的排名位置(如85分在第90百分位,说明超过了90%的考生)。可以有相同的百分比但不同的百分位数(取决于其他人的成绩分布)。

如何手动计算百分位数?

方法(插值法):将数据从小到大排序;计算索引 L = (P/100) × n(P为目标百分位,n为数据个数);若L为整数,百分位数 = 第L和L+1个数据的平均值;若L不是整数,向上取整,百分位数 = 第ceil(L)个数据。不同软件(Excel、R、Python)可能使用不同插值方法,结果略有差异。

什么是四分位数?

四分位数将排序数据分为四等份:Q1(第25百分位):下四分位数,25%的数据低于此值;Q2(第50百分位):中位数,数据的中间值;Q3(第75百分位):上四分位数,75%的数据低于此值;IQR(四分位距)= Q3 - Q1,是衡量数据离散程度的稳健指标。

百分位数在标准化测试中如何使用?

标准化考试(如高考、SAT、GRE)使用百分位数排名来比较考生表现:SAT 1400分(满分1600)≈ 第95百分位(超过95%的考生);GRE语文160分 ≈ 第86百分位。百分位排名比原始分数更有意义,因为它直接告诉你在考生群体中的位置,而原始分数随考试难度变化。

BMI百分位数如何用于儿童?

儿童BMI(2-19岁)不直接用数值分类,而是使用CDC年龄和性别特定的百分位数:低于第5百分位:体重过轻;第5-85百分位:正常体重;第85-95百分位:超重(风险区间);第95百分位及以上:肥胖。这种方法考虑了儿童在不同年龄和性别的正常生长规律。

百分位数在商业分析中有哪些应用?

业务指标中的百分位数应用:P50(中位数):典型用户体验(比平均值更稳健);P95/P99(第95/99百分位):"长尾"极端情况,关键SLA指标;服务器响应时间P99 < 500ms意味着99%的请求在500ms内完成;收入分析:P80的客户贡献了多少收入?薪资分析:处于行业P60意味着薪资高于60%的同类岗位。

如何在Excel中计算百分位数?

Excel函数:PERCENTILE(数组, 百分位) 或 PERCENTILE.INC(数组, 百分位):计算含端点的百分位数(0-1之间输入,如0.75表示第75百分位);PERCENTILE.EXC(数组, 百分位):排除端点计算(更严格);PERCENTRANK(数组, 值):返回某值在数据中的百分位排名。

Percentile Rank Formula

A percentile rank tells you what percentage of values in a dataset fall at or below a given score. The most widely-used formula — known as the inclusive or midpoint method — is:

Percentile Rank Formula

PR = ( (B + 0.5 × E) / N ) × 100

  • B — number of values strictly below the score
  • E — number of values exactly equal to the score
  • N — total number of values in the dataset

Example

Dataset: 15, 20, 35, 40, 50 — finding the rank of 35:

B = 2 (values 15, 20 are below) | E = 1 (value 35 equals 35) | N = 5

PR = ((2 + 0.5 × 1) / 5) × 100 = (2.5 / 5) × 100 = 50th percentile

Percentile Rank vs Percentile Score

These two terms are often confused. Understanding the difference is essential for interpreting results correctly.

TermDefinitionExample
Percentile RankThe percentage of values at or below a given scoreScore 35 has rank 50 → 50% of scores are ≤ 35
Percentile Score (Value at Percentile)The value that corresponds to a given percentile rankThe 75th percentile value is 40

Practical tip:When a test result says you scored in the “85th percentile,” it means your score (the percentile score) was higher than 85% of all test takers. The percentile rank is 85.

Linear Interpolation for Value at Percentile

When finding the value at a given percentile, the requested percentile often falls between two data points. Linear interpolation is used to estimate the exact value:

Interpolation Formula

index = (p / 100) × (N − 1)

value = data[floor(index)] + (data[ceil(index)] − data[floor(index)]) × (index − floor(index))

Example

Dataset: 10, 20, 30, 40 — finding the 50th percentile:

index = (50 / 100) × (4 − 1) = 1.5

value = 20 + (30 − 20) × 0.5 = 20 + 5 = 25

This is the same method used by NumPy's np.percentile() with the default linear interpolation setting.

Common Percentiles and Their Names

Certain percentiles have special names because of their frequent use in statistics and data analysis:

PercentileNameCommon Use
25thQ1 (First Quartile)Lower bound of the IQR
50thQ2 / MedianMiddle value, robust to outliers
75thQ3 (Third Quartile)Upper bound of the IQR
90thP90Performance benchmarks (latency, income)
95th / 99thP95 / P99Tail latency in systems engineering

Related Tools