Time Series Graphs

Construction Steps

• Axis Setup: Label the x-axis with the time units and the y-axis with the measurement units. Ensure the scale on the y-axis is consistent and starts from zero unless a break is clearly indicated.
• Plotting: Carefully mark each data point at the intersection of its time value and its measured quantity.
• Joining: Use a ruler to connect the points in chronological order with straight lines. Do not use curves, as they may imply a level of precision or a specific mathematical model that the data does not support.
• Multiple Data Sets: If comparing two different variables, use different colors or line styles (e.g., dashed vs. solid) and always include a key to identify each line.

Interpretation Techniques

• Identify Trends: Look for the 'big picture' movement—is the data generally increasing, decreasing, or remaining stable over the entire period?
• Spot Seasonal Patterns: Identify cycles that repeat at regular intervals, such as sales peaking every December or temperatures dropping every night.
• Analyze Gradients: Locate the steepest segments to find when the most significant changes occurred.

• Check the Scale: Always verify if the y-axis starts at zero. If it starts at a higher value, small changes may look much more dramatic than they actually are.

• Use a Ruler for Accuracy: When reading values from the graph, use a ruler to draw horizontal or vertical lines from the point to the axes to ensure you don't misread the coordinates.

• Describe with Numbers: When asked to describe a trend, don't just say 'it increased.' Use specific values from the graph, such as 'The temperature increased from $10^{\circ}C$ on Monday to $15^{\circ}C$ on Wednesday.'

• Context Matters: Always relate your answer back to the real-world scenario described in the question (e.g., 'The number of visitors peaked in the third quarter').

• Curving the Lines: A common mistake is to draw a 'line of best fit' or a smooth curve. In a time series graph, points must be joined directly to show the actual recorded path of the data.

• Misinterpreting Intermediate Values: Just because a line connects two points doesn't mean the value at the midpoint is exactly what the line shows. For example, if you measure temperature at 9 AM and 10 AM, the line is just an estimate of what happened at 9:30 AM.

• Confusing Bar Charts with Time Series: While bar charts can show time on the x-axis, time series graphs are preferred for continuous data to emphasize the flow and rate of change rather than individual category totals.