We propose a simple, model-free way to measure price selection and its impact on inflation. Price selection exists when prices that change in response to aggregate shocks are not representative of the overall population of prices. Due to selection, increases (decreases) in inflation can be amplified because adjusting prices tend to originate from levels far below (above) the average. Using detailed micro-level consumer price data for the United Kingdom, the United States and Canada, we find robust evidence of strong price selection across goods and services. At a disaggregate level, price selection accounts for around 36% of inflation variance in the United Kingdom and the United States, and 28% in Canada. Price selection is stronger for goods with less frequent price changes or with larger average price changes. Aggregation largely washes out price selection for regular price changes, but not for changes associated with price discounts. This evidence favors multi-sector sticky-price models with strong price selection at a sector level.