How to Calculate a Weighted Score or Index

Which job offer wins when one pays $13,000 more but adds a 45-minute commute? A weighted score answers that with one number per option. You put every criterion on a 0 to 100 scale, give each one a weight, then multiply and add. This guide builds a full scoring model from scratch, tests it, and shows how the same math powers indexes like the CPI.

What Is a Weighted Score?

A weighted score is one number that combines several criteria, each scaled to 0 to 100 and multiplied by its importance. You add the products, and the highest total wins. The method takes five steps.

First, list the criteria that matter. Second, score each option on each criterion using a common 0 to 100 scale. Third, split 100 points of weight across the criteria. Fourth, multiply each score by its weight, add the products, and divide by 100. Fifth, move the weights and confirm the winner survives.

The arithmetic in step four is a plain weighted average. The hard part is everything around it: fair scales, honest weights, and a stress test. Course grading uses the same math with syllabus weights. Our guide on how to calculate a weighted grade covers that case.

How Do You Build a Weighted Decision Matrix?

A decision matrix lists options in rows and criteria in columns, then adds a weighted total for each row. Here is a full example with three job offers and four criteria.

After converting each row to a 0 to 100 scale, the matrix looks like this. The method for that conversion comes in the next section.

Offer C scores (0 x 40 + 100 x 20 + 100 x 25 + 66.7 x 15) / 100 = 55.0. Offer B scores 52.5 and Offer A scores 40.1. To check any row fast, enter the scores and weights into the Weighted Average Calculator.

How Do You Normalize Criteria to a 0-100 Scale?

Use min-max scaling: score = 100 x (value – worst) / (best – worst). The best option gets 100, the worst gets 0, and the rest fall proportionally between them.

For salary, the range runs from $72,000 to $85,000. Offer A sits $6,000 above the bottom of a $13,000 range, so it scores 100 x 6 / 13 = 46.2. Growth ratings run from 6 to 8, so a rating of 7 scores 50.

How Do You Handle Lower-Is-Better Criteria?

For price, commute, or risk, the smallest value is the best. Swap the ends: score = 100 x (worst – value) / (worst – best). A 45-minute commute scores 100 x (60 – 45) / (60 – 15) = 33.3.

Skipping normalization breaks the model. Multiply raw salaries by weights, and $85,000 swamps a rating of 7. The raw totals crown Offer B for no real reason.

One more trap: min-max depends on the options in the set. Add a fourth offer with a higher salary, and every salary score shifts. For a model you reuse, fix the range in advance, such as 0 to 90 minutes for commute.

How Should You Choose the Weights?

Split 100 points across the criteria in proportion to how much each one matters to you. Weights that sum to 100 read directly as percentages, so salary at 40 means 40 percent of the decision.

Points allocation works best with 3 to 6 criteria. Start from an even split, then move points toward what you would refuse to give up. Write the weights down before you score any option, or the numbers drift toward the option you already like.

What Is a Quick Rank-Based Method?

Rank the criteria, then use the rank-sum rule. With 4 criteria, ranks 1 to 4 earn 4, 3, 2, and 1 points out of 10. That gives weights of 40, 30, 20, and 10.

Ranking salary first, growth second, commute third, and benefits fourth yields weights of 40, 30, 20, and 10. Offer C then scores 56.7, Offer B scores 55.0, and Offer A scores 35.1. The winner stays the same, though the gap narrows to 1.7 points.

Watch for double counting. Salary and bonus, or square footage and bedroom count, measure nearly the same thing. Scoring both gives that single factor twice its intended weight, so merge overlapping criteria into one.

Does the Winner Change When a Weight Moves?

Test it by moving 10 points from one criterion to another and recalculating. When the winner changes under a 10-point shift, the result is fragile and the decision needs more thought.

The job-offer model is fragile. Move 10 points from commute to salary, making the weights 50, 10, 25, and 15. Offer B jumps to 62.5 and Offer C drops to 45.0, so B now wins.

In fact, a 2-point shift is enough. At weights of 42 for salary and 18 for commute, Offer B scores 54.5 and Offer C scores 53.0. Across all 12 possible 10-point moves, Offer C wins 7 times and Offer B wins 5 times.

That pattern tells you something useful. The real question is how much you value salary against commute time, and the matrix has isolated that trade-off. A close total of 55.0 versus 52.5 is a signal to talk it through, not a verdict.

Avoid false precision as well. Growth and benefits ratings are judgment calls, so round totals to whole points.

How Does a Weighted Index Work?

A weighted index tracks a combined value over time relative to a base period set to 100. Each component’s change is multiplied by its weight, then the results are added.

Take a simple basket with three parts. Food carries 50 percent of the weight and its prices rise to 104. Housing carries 30 percent and rises to 106. Transport carries 20 percent and falls to 97.

The index equals 0.5 x 104 + 0.3 x 106 + 0.2 x 97 = 103.2. That means the basket costs 3.2 percent more than in the base period. The same base-100 idea works for a scoring model: a score that climbs from 55 to 60.5 becomes an index of 110.

The Consumer Price Index follows this pattern at a much larger scale. The BLS sets the 1982 to 1984 average price level equal to 100. It weights each item by its share of consumer spending, called relative importance. For the full mechanics, read our explainer on how the CPI is calculated.

FAQs About Weighted Scores

Sources

This article is general math and decision-making education. The job-offer figures are an illustrative example, not market data. Reviewed for accuracy by Prof. Dr. Khalil Mudassar, PhD. Last updated September 26, 2026.