A graph is a diagram showing a relationship between two variables, often sets of numbers, using a line or a series of bars, dots or other symbols. Whatever else makes up your graph, it’s impossible to create it without scales. Bar graphs have a vertical scale and a horizontal scale.

On a graph, the horizontal axis is called the x-axis. Typically, the x-axis describes a quantity that changes in a predictable fashion. For example, minutes, hours, days, months and years, or in the case of a scientific experiment, the control variable (i.e., the variable deliberately controlled by the scientist to determine the effects that change has on other variables). The scale of the x-axis changes depending on the type of data you want to record. It is usually linear, meaning one unit of length along the axis correlates to an incremental increase in the variable. If you carry out an experiment into crystal growth and want to plot your results on a graph, the horizontal axis could represent days from 0 to 14. In this case, the scale interval is one day. In some cases, there is no discernible interval on the axis; for example, if the graph represents the height of different mountain peaks or the populations of different cities.

The vertical axis on a graph, perpendicular to the horizontal axis, is called the y-axis. While the scale usually starts at 0, it does not have to. For example, if you are plotting the sales figures for a company over a period of six months, you may choose a different scale on the y-axis to give you a clearer picture of sales fluctuations. So if the figures in January, February and March are $2000, $2400 and $2800 respectively, a vertical scale from $1900 to $2900 with an interval of 200 gives a much clearer picture than a vertical scale from 0 to $3000 with an interval of 1000. In the case of a scientific experiment, the y-axis typically describes the outcome variable that is affected by the control variable on the x-axis.