Free tools Windows power users keep installed
One-click scans. No signup required.
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Use the median as the clearest single summary of a typical sale when property prices are skewed. Use the arithmetic mean when you want the average transaction value, and consider a trimmed mean when you need a mean-like summary that limits the influence of unusually low and high sales. Always define the area, period, property group and number of sales: these measures describe transactions observed, not necessarily the value of every home or like-for-like price change.
What each measure tells you
Arithmetic mean: the average transaction price
Add the sale prices and divide by the number of sales. Every transaction contributes, so the mean answers, “What was the average value of a sale?” In a market with a long high-price tail, a small number of expensive properties can pull it above the price of a typical sale. The Office for National Statistics (ONS) recommends considering the mean alongside the median to understand the distribution and the influence of very high-value transactions. Mean prices and sales counts can also help combine smaller areas into custom geographies, a task that cannot be done by simply averaging their medians. ONS, House Price Statistics for Small Areas QMI
Median: the midpoint sale
Sort the sale prices and take the middle one. If there is an even number of sales, a common convention is to average the two central prices. The median is less affected by extreme sales than the mean, making it a useful summary of the midpoint transaction in skewed housing markets. ONS describes it as the most appropriate average for its skewed house-price data. ONS, House Price Statistics for Small Areas QMI
Trimmed mean: an average after trimming both tails
Sort the sale prices, remove a specified share from the low end and a specified share from the high end, then calculate the mean of the remaining sales. It reduces the influence of tail values while retaining more of the observations’ detail than the median. But a trimmed mean has no single meaning until you state the trimming rule: for example, a 5% trim from each tail removes 5% of observations at the low end and 5% at the high end. NIST describes that as a common choice in general location analysis, not as a standard for property prices. A 50% trimmed mean is a much more aggressive statistic: it is the mean between the lower and upper quartiles. NIST, Measures of Location
#1 Best Overall
Choose the measure for the question
| Question | Useful summary | Why and what to qualify |
|---|---|---|
| What does a typical sold home cost here? | Median | It is resistant to a small number of extreme prices. Define the geography, period, property type and sales population. ONS guidance |
| What is the average transaction value? | Arithmetic mean, usually reported with the median | Every sale counts, but a few high-value transactions can make the mean unlike a typical sale price. ONS guidance |
| What is a mean-like summary with less influence from the tails? | Trimmed mean, alongside mean and median as checks | State the share removed from each tail and the number or proportion of sales excluded. Reviewed official sources do not establish a standard trim percentage for property prices. NIST, Measures of Location |
| How has a comparable home or the market changed in value? | A suitable mix-adjusted house-price index | Raw sale summaries can change when the types of properties sold change, even if like-for-like values do not. DLUHC, English Housing Survey 2024 to 2025: Market Value Survey |
| What price level matters near the lower end of the market? | Lower quartile or 10th percentile, with appropriate income measures | ONS identifies these percentiles as useful for affordability analysis; the 10th percentile represents the cheaper end of the market. ONS, House Price Statistics for Small Areas QMI |
Why one expensive sale changes the mean
A UK government example uses five sales: four properties at £100,000 and one at £1 million. The arithmetic mean is £280,000, while the median is £100,000; the example also gives a geometric mean of £158,000. These are the figures in the government’s simplified example, not an estimate for a particular market. DLUHC, Market Value Survey
In an actual dataset, a mean above the median is consistent with a high-price upper tail, but the difference alone does not prove what caused it. Check the sale distribution and the mix of properties sold before explaining the gap.
Rank #2
How to use a trimmed mean responsibly
- Choose a rule before interpreting results. Specify whether the trim is symmetric and the percentage removed from each tail. An asymmetric trim needs a substantive reason.
- Apply it consistently. Do not remove transactions simply because their prices make the result inconvenient. A genuine luxury sale may be unusual and still economically relevant.
- Investigate data errors directly. If a transaction is erroneous, document the data-quality issue and handle it under a stated rule rather than silently treating it as an outlier.
- Show the sensitivity. Compare the median, ordinary mean and one clearly specified trimmed mean if valid extreme sales might affect the interpretation. Report the number or proportion of observations removed.
Call the result a “trimmed mean” and give its trimming rule; do not present it simply as “the average.”
Limits of sale-price summaries
They describe homes that sold, not all homes
A completed-sale statistic represents the transactions observed during a period. It may differ from the value distribution of all homes, including homes that did not sell. For example, the UK government’s 2024 Market Value Survey estimates values across tenures and explicitly distinguishes those estimates from completed-sale house-price measures. DLUHC, Market Value Survey
The sold-property mix can change
If one period has more large detached homes sold and another has more flats, a raw mean or median can move even if like-for-like prices have not. ONS notes that its small-area sale-price statistics are not mix-adjusted; a house-price index that accounts for changing property characteristics is more suitable for measuring property-value inflation. ONS, House Price Statistics for Small Areas QMI
Small samples make local summaries fragile
ONS does not publish its small-area median, mean, lower quartile or 10th percentile for an area-year with fewer than five sales. That is a threshold for this ONS publication, not a universal rule for every dataset. ONS, House Price Statistics for Small Areas QMI
Define the population and distinguish the kind of average
State whether your figures cover all residential sales or a narrower group such as a particular property type, new builds or a tenure group. Geography, period and coverage matter too; ONS tables report by year, geography and property type, and some combinations have limited coverage. Also distinguish the arithmetic mean from a geometric mean: the UK House Price Index uses a weighted geometric mean, which gives high values less weight and is usually closer to the median. That is part of the index method, not a substitute label for an arithmetic mean. ONS guidance DLUHC, Market Value Survey
Recommended Free Tools
How to report property-price summaries
Include the sale count, defined property group, geography and period. If you report a trimmed mean, name both tail percentages and the number or proportion of sales removed. Keep the interpretation tied to what the figures establish: transaction summaries are not automatically estimates of like-for-like price change.
Among [number] [defined property group] sales in [geography] during [period], the median sale price was [currency amount] and the arithmetic mean was [currency amount]. The [x% from each tail] trimmed mean was [currency amount], calculated after removing [number or proportion] of sales from each tail. These are transaction-price summaries, not a mix-adjusted estimate of like-for-like price change.
A separate example: values across all tenures
In the Department for Levelling Up, Housing and Communities’ 2026 report, the rounded 2024 mean and median market values for all dwellings in England were £348,000 and £275,000, respectively. The figures cover values across tenures, not completed-sale prices. For owner-occupied dwellings the report gives £398,000 mean and £320,000 median; for private rented dwellings, £289,000 and £235,000; and for social rented dwellings, £218,000 and £180,000. In that report’s Market Value Survey series, mean property value rose 41% nominally and median value rose 45% nominally from 2015 to 2024; those changes are specific to that series, not a general description of transaction-price growth or a house-price index. DLUHC, English Housing Survey 2024 to 2025: Market Value Survey
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
