A Guide to Statistics
The following is a brief description of two key statistics (percent and rate) used on this website and a word of caution about their use.
Percent means per 100. For example, 15% means 15 out of 100; 75% means 75 out of 100.
Percent = (Number in sub-group) ÷ (Number in whole group) x 100
Example: Percent of babies born at low birth weight (LBW), 2002
Percent = (Number LBW) ÷ (Total number of births) x 100
= 6,623 ÷ 73,250 x 100
= 9% of babies born in 2002 weighed less than 2,500 grams (5.5 pounds)
The easiest way to understand a rate is to think of a percent as a rate per 100. (Many indicators are presented as rates per 100,000 or 1,000.) In the example above, 9% of babies born at low birth weight could also be expressed as “9 babies per 100” were born at low birth weight.
Rate = (Number in sub-group) ÷ (Number in whole group) x MULTIPLIER
Example: Rate of youth (ages 10-17) arrested for violent crimes per 100,000 youth (ages 10-17)
Rate = (Number of youth ages 10-17 arrested) ÷ (Number of youth ages 10-17) x 100,000
= 3,037 ÷ 567,678 x 100,000
= 535 per 100,000 youth ages 10-17 were arrested for violent crimes in 1998
Using Percentages or Rates with Small Numbers of Incidents
Caution is necessary when using percentages and rates with small numbers of incidents. If the item to be measured has less than 5 occurrences (e.g., infant mortality in a given jurisdiction for a given year), then a percentage or rate should not be produced. One or both of the following methods can be employed to create a more stable percentage or rate:
- Multi-year averaging, which involves using a longer time period to produce the rate (e.g., using 3 or 5 years’ worth of data); or
- Enlarging the geographic area (e.g., using a region containing several jurisdictions).
Both of these methods increase the number of observed events and the stability and reliability of percentages or rates calculated.