A line graph (also called a time chart) is a data display used to examine trends in data over time, also known as time series data. Line graphs show time on the x-axis (for example, by month, year, or day) and the values of the variable being measured on the y-axis. Each point on the line graph summarizes the data collected at that particular time.^[1,p.127] Line graph is interpreted by looking for patterns and trends in the graph from left to right.^[1,p.127]

Tips for evaluating a line graph and spotting misleading representations are:^{[1,pp.128,130-131,339]}

• The scale of the vertical axis can make a big difference in the way the time chart looks. Large increments and/or lots of white space makes differences look less dramatic; small increments and/or a plot that totally fills the page exaggerates the differences.
• Too many points will look overwhelming and make it difficult to spot large patterns in the data. It may be appropriate to reduce a chart to longer intervals on the time axis and plot a mean or median value for all data points that fall in each interval instead.
• Conversely, too few points may oversimplify the graph and obscure smaller patterns.
• Presence of the trend in a time chart does not explain why the trend exists. Additional statistics are required to confirm any hypothesis or cause-and-effect.
• Gaps in timeline should be of the same length relative to each other. Time should be treated as intervals rather than just labels.
• Units should be appropriate for plotting of change over time. For example, plotting rates of a variable absolute values (e.g. crimes per capita) over time is more appropriate than plotting absolute number of the same variable (e.g. number of crimes).

Variability in a line graph should not be confused with the variability in a histogram. If values change over time, they are shown on a line graph as highs and lows, and many changes from high to low indicate lots of variability. A flat line on a time chart indicates no change and no variability in the values accross time. But when the top of a histogram appears flat, the data is spread out uniformly accross all groups, indicating a great deal of variability in the data.^[1,p.128]