Median vs Mean: Reading Salary Figures Honestly
Every data page on this site reports both a median and a mean, and leads with the median. The reason is the shape of wage data. Salaries are bounded below — no filing in this dataset can lawfully sit below the annualized minimum-wage floor our outlier rule uses — but they are effectively unbounded above, where executive, specialist and placeholder-style entries stretch the tail. In a distribution shaped like that, the mean, which adds every value and divides by the count, is pulled toward the tail. A handful of very large offers can lift an employer's mean thousands of dollars above the wage its typical filing actually carried, without a single typical offer changing.
The median is immune to that pull because it is a positional measure: line the filings up from lowest to highest offered wage, and the median is the one in the middle. Half the filings offered more, half offered less. Whether the largest offer is three hundred thousand dollars or three million, the median does not move. That robustness is why official statistics on pay are usually reported as medians, and why this site treats the median as the headline figure and the mean as a diagnostic: when a group's mean sits far above its median, the gap itself is information — it tells you the group's distribution has a heavy upper tail, often because it mixes junior and very senior roles.
The percentiles complete the picture. The 25th percentile is the wage below which a quarter of filings fall; the 75th percentile is the wage above which a quarter of filings sit. The band between them — the interquartile range, labelled 'middle half' on this site — is the most honest short description of what a group typically offers, because it excludes both tails by construction. When you compare two employers or two cities, comparing their middle halves usually tells you more than comparing any single number: overlapping bands mean similar typical offers even if the medians differ slightly, while barely-touching bands signal a real difference in typical pricing.
None of these measures can rescue a badly defined group. A median computed over a mixture of occupations, levels and cities describes the mixture, not any job within it — the central editorial principle behind this site's structure of occupation, city, state and employer pages. The right sequence is to narrow the group first (occupation, then place, then employer) and only then read its median. Readers who reverse that sequence, starting from a national all-occupation median, end up with a precise number answering a question nobody actually has.
A worked habit to finish with. Suppose an employer's page shows a median below the national figure. Before concluding anything, check its top occupations: if its filings concentrate in lower-paid occupations, its overall median mostly reflects mix. Then check the occupation page for your role and find where that employer's typical segment sits in the occupation's range. Then check the city page for local pricing. Three medians, read in that order, will support a conclusion. One median, read alone, will support almost any conclusion — which is exactly why it should not be trusted.
How These Figures Were Computed
Every figure on this page was computed by our script directly from the U.S. Department of Labor OFLC LCA disclosure files for FY2024 Q4 and FY2025 Q1–Q4, retrieved in October 2026. We kept only cases whose CASE_STATUS is exactly “Certified” — denied, withdrawn and certified-withdrawn cases are excluded. Headline medians use filings whose wage unit is a year, because annual salaries compare cleanly; hourly, weekly and monthly filings are counted in case totals but are not converted into the headline median, and our methodology page explains that choice in full. Wages below $15,080 a year (the annualized federal minimum wage at $7.25 an hour) or above $1,000,000 a year are excluded from medians as likely data-entry placeholders, and the number excluded is reported in the manifest that accompanies this site. Medians are shown alongside means because a handful of very large offers can drag a mean far above what a typical filing actually offered.
Frequently Asked Questions
Where does the data behind this guide come from?
All figures referenced in this guide come from the U.S. Department of Labor OFLC LCA disclosure files for FY2024 Q4 and FY2025 Q1–Q4, retrieved in October 2026, filtered to 662,198 cases with a status of exactly Certified. The methodology page documents the full cleaning process.