The Oregon Health Division has a highly recommended online publication called "The Data Difference: The Data User's Guide," which addresses issues of proper usage and understanding of data. In addition, we provide some overview of basic concepts and issues here:

Everything You Always Wanted to Know About Using Data

What is a frequency?

A frequency is simply the total count of an event. For example, the total number of births in a county is the frequency.

What is a proportion?

This is usually expressed as a percentage. It is a good measurement to show the portion of the whole. Let's say you want to know the percentage of people 65 years of age and over in Oregon. This is the proportion of the population that is 65 years of age or older. To calculate this, take the number of people 65 years of age or older divided by the total population, multiplied by 100.

What is a rate?

A rate measures how frequently an event is occurring or how common an event is in a population during a specific time. The equation is as follows:    (number of events occurring during a given period) / (population at risk during same period) x 10ⁿ.

The size of 10ⁿ usually equals 1,000 or 100,000. For example, 10³ = 10 x 10 x 10 = 1,000, and 10⁵ = 10 x 10 x 10 x 10 x 10 = 100,000.

The population at risk is the population that is at risk of experiencing the event. This is the population that can have the event occur to them. For example, if you are calculating a fertility rate, the population at risk of experiencing the event consists of females who can have babies. Often the exact population at risk is not available, so the entire population of a sub-population is used. When the whole population is used in the denominator, this is referred to as a crude rate because it assumes that everyone has the same risk of experiencing the event.

When calculating rates, be sure to use:

• The same age groups in the numerator and denominator.
• The same year in the numerator and denominator.
• The same geographical area in the numerator and denominator.

Let's take pregnancies among teens (females ages 10-17) as an example. The pregnancy rates measures how frequently pregnancy occurs among the teen population. The numerator consists of the number of teens who were pregnant and the population at risk is the population of females ages 10-17. The equation looks like this: (pregnancies among females ages 10-17 during a specific time period) / (population of females ages 10-17 for the same period) x 1000.

Teen pregnancy rate is usually expressed per 1,000 teens. The teen pregnancy rate for 1995 is: 3,284 / 170,807 x 1,000 = 0.0192 = 19.2

There were 19.2 pregnancies for every 1,000 females age 10-17 in 1995.

How can the rate be per 1,000 or 100,000 people when there are not that many people in our county?

Rates give a meaningful way to compare the frequency of teen pregnancy between places with different populations. Comparing the number of teen pregnancies in Washington and Harney counties (322 vs. 9) is not meaningful: Washington County will always have a larger number simply because it has a larger teen population at risk of getting pregnant.

But when we divide the number of pregnancies by the population of the area, and multiply the rates by 1,000, we effectively negate the population difference. We're saying that if Washington and Harney counties each have 1,000 females age 10-17, the frequency of pregnancy would be 15.9 per 1,000 in Washington County and 20.9 per 1,000 in Harney County. The rate in Harney County is higher even though its number of teen pregnancies is lower.

How do I compute the change in a rate or any data from one year to another?

The best way to calculate the percentage change in the rate or in the data is to use the following equation: (new data - old data) / (old data) x 100.  For example, if you want to calculate the percentage change in the teen pregnancy rate for females age 10-17 from 1995 to 1994, set up the equation like this:

The pregnancy rate for females age 10-17 was 18.9 in 1994 and 19.2 in 1995. This computes to an increase of 1.6 percent:

(1995 - 1994) / 1994 x 100 

(19.2 - 18.9) / 18.9 = 0.3 / 18.9 = 0.016 x 100 = 1.6% 

Source: Oregon Health Division, Center for Health Statistics