Average Percentage Calculator
Enter up to eight percentage values and get their average immediately. Switch to weighted mode to account for different group sizes, for example when combining test scores from classes of different numbers of students. The step-by-step panel shows every calculation with your actual numbers.
Formula
Worked example
Three classes scored 72%, 85%, and 60%. Simple average: (72 + 85 + 60) / 3 = 72.33%. If the classes had 30, 25, and 20 students, weighted average: (72x30 + 85x25 + 60x20) / (30+25+20) = (2160+2125+1200)/75 = 5485/75 = 73.13%.
Simple average vs weighted average of percentages
A simple average adds all the percentages together and divides by how many there are. This is correct when each percentage applies to a group of the same size, or when you just want the arithmetic mean of a list of numbers. A weighted average multiplies each percentage by its group size before summing, then divides by the total group size. Use it whenever the groups behind each percentage are different sizes. For example, a 90% pass rate in a class of 200 students should carry more weight than an 80% pass rate in a class of 10. A simple average would treat those two numbers as equal, which overstates the contribution of the smaller group.
When does averaging percentages give a misleading answer?
The most common trap is computing the simple average of percentages that come from differently sized groups. Suppose two departments reported customer satisfaction: Department A scored 95% with 10 responses and Department B scored 55% with 90 responses. The simple average is (95 + 55) / 2 = 75%, but the weighted average is (95x10 + 55x90) / 100 = (950 + 4950) / 100 = 59%. The weighted figure is far more accurate because 90 times more customers responded to Department B. This error is sometimes called Simpsons Paradox, and it appears in grade calculations, survey aggregation, conversion-rate reporting, and sports statistics.
How to calculate a weighted average percentage
Step 1: For each percentage value, note the group size it represents. Step 2: Multiply each percentage by its group size to get a weighted contribution. Step 3: Add all the weighted contributions together. Step 4: Divide by the total of all group sizes. The result is the weighted average percentage. This is identical to asking "if I pooled all the groups into one, what percentage of the combined total would count as a success?". The formula is: weighted average = sum(p_i times w_i) divided by sum(w_i), where p_i is each percentage and w_i is each group size.
Practical uses
Average percentages come up in grade calculations (combining scores from assignments of different lengths or worth), conversion rate analysis (combining traffic sources with different visitor counts), survey aggregation (merging responses from groups of different sizes), finance (computing portfolio returns where positions have different weights), and manufacturing quality control (combining defect rates from production runs of different sizes). In all these cases, deciding between simple and weighted is the most important choice before you calculate.
Simple vs weighted average: quick reference
| Scenario | Correct method | Error if wrong method used |
|---|---|---|
| All groups have the same size | Simple average | None: results will match |
| Groups differ in size | Weighted average | Can be far off (Simpson's Paradox) |
| Combining test scores, all same max marks | Simple average | Minimal to none |
| Combining pass rates from different classes | Weighted average | Potentially large |
| Averaging survey satisfaction across regions | Weighted average | Potentially large |
| Averaging a list of percentages for reference | Simple average | None: just a summary |
Choose the method based on whether the groups behind your percentages are the same size.
Frequently asked questions
Can I average percentages by just adding and dividing?
You can, and the result is called the simple average. It is correct when all the groups or samples behind your percentages are the same size. If the groups differ in size, adding and dividing gives a misleading answer, and you should use a weighted average instead (multiply each percentage by its group size, sum those products, then divide by the total group size).
What is the difference between simple and weighted average percentage?
A simple average treats every percentage as equally important regardless of how many observations it represents. A weighted average gives each percentage a weight equal to the number of observations behind it, so larger groups influence the result more. If three classes of 30, 25, and 20 students scored 72%, 85%, and 60% respectively, the simple average is 72.33% but the weighted average is 73.13%, because the largest class happened to have the lowest score.
What is Simpson's Paradox and how does it relate to averaging percentages?
Simpson's Paradox occurs when a trend that appears in several groups reverses or disappears when the groups are combined using a simple average. It happens because simple averaging ignores group sizes. A department with a 95% success rate on 10 cases and one with 55% on 90 cases have a simple average of 75%, but the accurate combined rate is only 59%. Always use weighted averages when group sizes differ.
Can a percentage be over 100%?
Some metrics expressed as percentages can exceed 100%, for example year-over-year growth rates, click-through rates calculated differently, or conversion metrics in certain industries. This calculator accepts any numeric value, positive or negative, so you can average growth rates or other unbounded percentages freely.
How do I average percentages from a spreadsheet?
For a simple average use =AVERAGE(range) in Excel or Google Sheets. For a weighted average use =SUMPRODUCT(percentages, weights)/SUM(weights). You can also paste your values here one by one and switch to weighted mode to get the result with a step-by-step breakdown.
Does the order of the percentages affect the average?
No. Both simple and weighted averages are commutative: the result is the same regardless of the order you enter the values. Only the values themselves (and their group sizes in weighted mode) matter.