Data analysis. Presenting data
The set of all objects to be examined is the population, and the objects chosen from it at random form a sample; the number of objects is the sample size n. When we cannot test every object, we draw conclusions from the sample. The terms of the sample are variants; writing them in increasing order gives the variation series. If variant x₁ occurs n₁ times, x₂ occurs n₂ times and so on, the numbers n₁, n₂, … are the frequencies and W₁ = n₁/n, W₂ = n₂/n, … are the relative frequencies; the frequencies add up to n and the relative frequencies to 1. Data are shown in tables and in a frequency polygon (a broken line joining the points).
A school wants to know how long its 400 pupils take to get to school. A random sample of 20 is asked: 5 minutes for 6 pupils, 10 minutes for 8, 15 minutes for 6. The relative frequencies are 0.3, 0.4, 0.3, adding up to 1.
A statistician puts survey results into a table, a quality inspector takes a sample from a batch, and a journalist presents figures by frequency.
Take the first 20 words of any text, write the number of letters in each and make a frequency table. Check that the relative frequencies add up to 1.
| Variant | 2 | 3 | 4 | 5 |
|---|---|---|---|---|
| Frequency | 1 | 3 | 5 | 3 |
| Relative | 1/12 | 1/4 | 5/12 | 1/4 |
“Class sample”: each pupil writes their shoe size; the class builds the variation series and the frequency and relative frequency tables.