A weather table is a small dataset with a story attached. Mean, median, and range can describe different parts of that story, but only if you keep the units, signs, and time period visible.

Put the data in order

Imagine five daily high temperatures: 2°C, −1°C, 4°C, 3°C, and −2°C. Sort them from least to greatest: −2, −1, 2, 3, 4. Sorting makes the middle value and the spread easier to see.

Negative temperatures are less than zero. On a number line, −2 is to the left of −1, so it is colder even though 2 is larger than 1 when you ignore the signs.

Calculate the mean

Add the temperatures and divide by the number of days: (2 + (−1) + 4 + 3 + (−2)) ÷ 5 = 6 ÷ 5 = 1.2°C. The mean uses every value, so one unusually warm or cold day can pull it up or down.

Keep the degree symbol in your notes. The 1.2 is not just a number; it is the average high temperature for the five-day example.

Compare the median and range

A dataset can have a mean and median that differ. That is not an error; it is a clue that the values are not balanced evenly around the centre.

  1. The median is the middle value after sorting. Here it is 2°C.
  2. The range is maximum minus minimum: 4 − (−2) = 6°C.
  3. Use the median when you want a middle day that is less affected by an extreme value.
  4. Use the range when you want the total spread from the coldest to warmest value.

Ask a precise question

“What was the weather like?” is too broad for one calculation. Ask instead: What was the average high this week? How large was the spread? Which day was closest to the median? Clear questions help you choose the right summary.

When working with real observations, record the location, dates, and whether you are using high, low, or hourly temperatures. A summary is only as clear as the data behind it.

Keep these ideas

  • Sort signed temperatures before finding the median.
  • The mean uses every value and can be affected by extremes.
  • The range is maximum minus minimum, including the signs.
  • State the dates, location, and measurement type before interpreting a weather dataset.

Keep exploring

These references are useful places to go deeper: