5 Public Health Data
Quantitative data
Quantitative data is used to describe “what” is occurring, “who” is affected, “where” it is happening and to count “how much” there is. For example, quantitative data can be used to describe the number of cases of a disease and which towns or villages those cases live in.
There are different types of quantitative data:
Categorical data
Nominal data
Variables with no order or ranking sequence, e.g., Province
Ordinal data
Variables that have an order or ranking, e.g., educational level, age group
Binary data
Variables with only two options, e.g., True/False, Pass/Fail, Yes/No
Numerical data
Discrete data
Data that can only be certain defined values (e.g., whole numbers such as counts, number of students, number of live births)
Continuous data
Continuous data can take any value and is often data that is a measurement (e.g., weight, height, length, time)
Qualitative data
Data that explains “why” or “how” something is occurring. Qualitative data can include stories and pictures. They are words or images that describe experiences or feelings. Usually collected through interviews, group discussions, observation, and documents.
Used to explore values, attitudes, opinions, feelings, and behaviour (the “why”).
Demographic data
Demographic data are the characteristics of people that describe who is affected by a health condition. Common characteristics used in descriptive epidemiology are age, sex and occupation.
Variable
Any characteristic or attribute that can be measured, such as age, sex, village, laboratory diagnosis, blood sugar level.
Fractions
A fraction is a number that represents how many people out of a group have a condition or characteristic. It tells us what proportion of a group are affected and can be used to compare counts from group of different sizes (e.g. different sized populations). Fractions have a numerator and a denominator.
Numerator
The numerator is the number of people you are referring to with the fraction and is represented by the number of people above the line. In epidemiology, the numerator is often the number of people who meet the case definition.
Denominator
The denominator is the total number of people in the group that the people in the numerator come from. In epidemiology, the denominator is often the population at risk.
|
If eight people sat at a table and ate the food, this is the number of people (population or denominator) at risk from getting sick. If six people got sick (numerator) then 6/8 people became unwell, which we can also express as 3/4. |
Proportion
A proportion is the size of something in relation to the size of the group that it comes from. For example, the number of cases in relation to the size of the total population. It can be expressed as a fraction (1/5, a decimal (0.2), or as a percentage (20%). A decimal is based on the idea that 1 represents the whole group and in a percentage 100% represents the whole group.
A percent can be expressed as a decimal or a fraction.
A half can be written
In the fraction example above, the numerator = 1 and the denominator = 2.
Median
A measure of the centre point of a set of data, with a equal number of observations either side. The median identifies the ‘middle value’ in a set of observations.
1, 3, 5, 7, 12, 18, 25
The middle number (7) = Median
Example 1:
What is the median of these numbers?
8, 5, 14
Put the numbers in order:
5, 8, 14
The middle number is 8, so the median is 8.
Example 2:
What is the median of these numbers?
23, 12, 4, 5, 45, 22, 2, 4, 7, 7, 15, 26
Put the numbers in order:
2, 4, 4, 5, 7, 7, 12, 15, 22, 23, 26, 45
There is an even number of values (n=12), so there is no middle number. In this case, our median is between the two middle numbers (7 and 12).
Add the two middle numbers and divide by 2:
7 + 12 = 19 ÷ 2 = 9.5
The median value of these numbers is 9.5.
The median value is much less affected by outliers and skewed data and is usually the better option in these circumstances. Medians are commonly used in field epidemiology because the things we measure often have outliers or are skewed (e.g. age, number of cases per month).
Mean
A measure of the mathematical centre of a set of data. It is commonly called the average. It is calculated by adding together all the individual values in a set of observations and dividing by the number of observations.
Example 1:
What is the mean of these numbers?
8, 5, 14
Add the numbers: 8 + 5 + 14 = 27
Divide the total by the number of observations (there are 3 observations in this example):
27 ÷ 3 = 9
The mean is 9.
Example 2:
What is the mean of these numbers?
23, 12, 4, 5, 45, 22, 2, 4, 7, 7, 15, 26
Adding all of the above numbers: 172
Divide the total by the number of observations (there are 12 observations in this example):
172 ÷12 = 14.3
The mean is 14.3.
The mean value can be significantly affected by values that are much smaller or larger than the rest of the data (outliers) or if the data are unevenly distributed over the range of the data (skewed). The mean value may be misleading in these circumstances.
Outlier
Outliers are data points that are far away from most of the values in a dataset.
Outliers can change the mean a lot, so if a dataset has outliers it is better to use the median as the measure of the centre of the data. It is important to check your data is accurate and that any outliers have not been entered by mistake.
Example:
In a measles outbreak, there are 22 cases that range from 2 to 14 years of age and there is one case that is 108 years old on the linelist. Here, the outlier would be the person who is 108 years old. This could be a data entry error of the age as we do not expect to find many 108-year-old people in the community, and we also expect older people to be immune to measles. We need to check our data against what we know about the disease and population.
Example:
What is the mean and median of these numbers?
3, 78, 4, 5, 7
The mean = 3 + 78 + 4 + 5 + 7 = 97 ÷ 5 = 19.4
The median is the middle value, after putting the numbers in order:
3, 4, 5, 7, 78
The median value of these numbers is 5.
The big difference between the mean and median is because there is an outlier in the dataset. In this example, the outlier is 78.
You would need to check that 78 is correct. If it is correct, you need to report your data using the median and range.
Mode
The most common value in a set of observations (the number that appears the most times).
Range
The range is the difference between the largest and smallest values in a distribution. In common use, the span of values from smallest to largest.
Example:
What is the range of these numbers?
3, 1, 7, 5, 9, 4
Put the numbers in order: 1, 3, 4, 5, 7, 9
The range shows the minimum number and the maximum number.
The range = 1-9.
