Understanding Descriptive Analysis

Descriptive analysis summarises the data you have. It does not automatically claim that the same pattern exists in a wider population, and it does not explain why an event occurred.

7.3.1 Frequencies and counts

Friday's count is 60 recorded visits. Sixty visits are not necessarily sixty different people. One person may attend more than once.

7.3.2 Calculating the total

40 + 50 + 45 + 55 + 60 = 250 visits over five days. Do not use that total to describe a longer period unless you have the information for that period.

7.3.3 Calculating the mean

The mean is the sum divided by the number of values. 250 ÷ 5 = 50. The mean was 50 recorded visits per day. That does not mean every day had 50 visits.

7.3.4 Understanding the median

The median is the middle value when the observations are in order. For 40, 45, 50, 55 and 60, the median is 50. For 20, 25, 30, 35 and 140, the mean is 50 and the median is 30. The value 140 pulls the mean up. Neither measure is automatically better. Choose the one that answers the question.

7.3.5 Understanding percentages

If a hospital expected 20 reports and received 15, completeness is 15 ÷ 20 × 100 = 75%. Confirm the numerator and the denominator. Reports received divided by reports expected is completeness, not accuracy.

7.3.6 Understanding proportions and rates

If 30 of 100 expected reports were missing, the proportion missing is 30%. A rate usually names an event, a population at risk and a time period. Do not treat count, proportion, percentage and rate as the same word.

7.3.7 Understanding simple trends

Recorded visits rose from January to March and fell slightly in April. That describes the pattern. It does not show that patient health improved, that quality increased, or that an intervention caused the change.

7.3.8 Which calculation should you use?