3.3 Types of graphs and their uses
Many different types of graphs exist, and each has something different about it that makes it useful in a unique way. Here we will be looking at just two types of graph: bar graphs and line graphs.
Here is a bar graph and a line graph plotting the patient's hourly temperature data that we looked at in Table 8.
From either of these graphs, you can quickly see that the patient's temperature is gradually declining through the day, and by the seventh hour (19:00) it is at a normal level of 36.9 °C. Representing the data as a graph also allows you to estimate what the patient's temperature probably was at the fifth hour of the measurement period (17:00), when someone forgot to take a reading. If you imagine a straight line connection between the temperature values at the fourth and sixth hours (16:00 and 18:00 respectively) this line would intersect the fifth time point at about 37.3 °C.
Which type of graph is best to use? To help answer this question you can consider the list shown in Table 9.
Table 9 Choosing the best graph for your data - advantages and disadvantages of bar and line graphs
|Bar and line||trends in data can be seen clearly (how one variable affects the other)|
|Bar and line||easy to use the value of one of the variables to determine the value of the other variable|
|Bar and line||enables predictions to be made about results of data you don't have yet|
|Bar||best for 'discrete' variables (those that change in jumps, with no 'in between' values)|
|Line||best for 'continuous' variables (those that change smoothly)|
You will see from Table 9 that we have identified two different types of variable, and these are defined by the way in which their numerical values change. Discrete variables can only have specific values within any given range (e.g. 1, 2, 3). Continuous variables are not limited in this way, and can have any value within a range.
Find some examples of continuous data and discrete data in your workplace.
Examples of continuous variables could include: temperature, blood pressure and pH. Examples of discrete variables might be: blood type, numbers of patients and needle size.