2.2 Tables and percentages

Tables often give information in percentages. The table below indicates how the size of households in Great Britain changed over a period of nearly 30 years.

Thus, in 1961, 14% of households consisted of only one person, compared with 27% in 1991. You can see a steady rise in the percentage of smaller families and a decline in the percentage of larger families over the 30-year period.

Other information can be extracted from the table by doing simple calculations. For example, in 1991, the total number of households surveyed was 22.4 million, so the actual number of four-person households surveyed is found by calculating 16% of 22.4 million, which is 3584 000 households. (Because 22.4 million is a rounded figure, this number of households will not be exact.)

Since each column in the table should include all the households surveyed, the total of all the entries in a column should be 100; indeed, for 1991,

However, the column total is not always exactly equal to 100. All the entries have been rounded to whole numbers, and this can sometimes introduce rounding errors. For instance, the total of the 1961 column is

Rounding errors are usually very small, so the total should always be very close to 100.

Sometimes the total percentages for both rows and columns are indicated, as in the table below which shows the percentages of families in Great Britain with different numbers of dependent children (at Spring 1998).

In tables like this, the row totals and the column totals should always add up to the same number. For example, in the table above,

and

Here, the row totals and the column totals both add up to 100, but in other tables rounding errors might mean that the two totals are not exactly 100 (though they should both be the same).