Seasonally adjusted: what it means
Figures that have had regular calendar patterns removed, such as Christmas shopping or summer jobs, so that underlying movements between months or quarters can be seen.
A seasonally adjusted series has had its regular, predictable calendar effects estimated and removed. Retail sales jump every December and some jobs rise every summer; adjustment strips out these repeated patterns so that a month can be compared fairly with the month before.
Worked example
Imagine hypothetical sales of 100 in November and 130 in December. If December is normally 30% above November, the seasonally adjusted figures would show no change. If December came in at 120, the raw data show a rise but the adjusted data show a fall, because sales were weaker than the usual December boost.
How it trips up claims
- Comparing one raw (non-seasonally adjusted, often marked NSA) month with the previous month can show "rises" or "falls" that are just the season.
- Seasonal factors are re-estimated as new data arrive, so adjusted figures for past months can be revised.
- For data that are not adjusted, compare with the same month a year earlier (year-on-year).
The ONS explains its methods and labels each time series as SA or NSA in the series title.
Also searched as: SA, NSA, seasonal adjustment, what does seasonally adjusted mean, not seasonally adjusted
Key figures where it matters
- UK employment rate
- UK GDP growth (quarterly)
- UK unemployment rate
- Average weekly hours worked in the UK
- UK consumer credit growth
- UK current account balance
- UK household saving ratio
- UK job vacancies
- UK monthly GDP growth
- Mortgage approvals for house purchase
- Net additional homes in England
- NHS employment (UK)
- UK redundancy rate
- Retail sales volume growth (Great Britain)
- UK trade balance (goods and services)
Questions that use this term
Related terms
- Year-on-year change: The change between a figure for one period and the same period twelve months earlier, such as March this year compared with March last year.
- Quarter-on-quarter change: The change from one three-month period to the next, such as April to June compared with January to March. Used for GDP growth.
- Revisions: Changes to previously published figures, made when more complete data, improved methods or updated seasonal factors become available. A normal part of official statistics.
- Rolling average: An average recalculated each period over a fixed window, such as the latest three months or twelve months, to smooth out short-term ups and downs.
More measuring change over time terms: Real terms · Cash terms · Percentage point · Year-on-year change · Quarter-on-quarter change · Rolling average · Index number · Base year · Base effect · Compound annual growth rate · Discontinuity (break in series).