Welcome to Bar Charts, Line Graphs, and Frequency Diagrams

Numbers on a page can sometimes be hard to understand. Statistics is all about turning those numbers into visual stories. In this chapter, we will learn how to use bar charts, line graphs, and frequency diagrams to make data easy to read and compare. These are essential tools for any statistician!

1. Bar Charts

Bar charts are one of the most common ways to show qualitative (non-numerical) or discrete (countable) data. The height (or length) of the bar represents the frequency of that category.

Types of Bar Charts

  • Simple Bar Charts: One bar for each category. Great for showing the frequency of different types of pets or favorite colors.
  • Multiple Bar Charts (Grouped): These show two or more sets of data side-by-side. For example, you could compare the number of goals scored by Team A and Team B over several months.
  • Composite Bar Charts (Stacked): These show different parts of a total stacked on top of each other. For example, a single bar could show the total number of students in a year group, with different colored sections for those who walk, cycle, or take the bus.

Top Tip: In a standard bar chart for qualitative data, we usually leave gaps between the bars to show that the categories are separate.

Bar Line (Vertical Line) Charts

Sometimes, if the data is discrete numerical data (like the number of siblings people have), we use thin vertical lines instead of thick bars. This emphasizes that the data can only be specific whole numbers (you can't have \(1.5\) siblings!).

2. Line Graphs and Time Series

Line graphs are used to show how a variable changes, usually over time. We plot points and join them with straight lines.

Time Series Data

A time series is a specific type of line graph where the horizontal axis (\(x\)-axis) always represents time (seconds, days, months, or years).

When looking at time series, statisticians look for:

  • Trends: Is the data generally going up (upward trend) or down (downward trend) over a long period?
  • Seasonal Trends: Does the data repeat a pattern at regular intervals? (e.g., ice cream sales peaking every summer).
  • Cyclic Trends: Patterns that repeat but not necessarily at fixed calendar intervals.

Example: If you plot the temperature of a room every hour, you are creating a time series graph.

3. Frequency Polygons

A frequency polygon is a graph used to display the distribution of grouped frequency data. It helps us see the "shape" of the data set.

How to draw a Frequency Polygon:

  1. Calculate the midpoint for each class interval. Formula: \( \text{Midpoint} = \frac{\text{Lower Limit} + \text{Upper Limit}}{2} \)
  2. Plot the frequency on the \(y\)-axis against the midpoint on the \(x\)-axis.
  3. Join the plotted points with straight lines.

Common Mistake to Avoid: Do not join the first and last points to the \(x\)-axis unless there is a frequency of \(0\) at those points. Just join the points you have plotted from your table.

4. Identifying Construction Errors

Sometimes, graphs are drawn poorly—either by accident or to trick the reader. You need to be a "data detective" and spot these errors:

  • Truncated Axis: The \(y\)-axis doesn't start at \(0\). This can make small differences between bars look massive.
  • Unequal Widths: In a bar chart, all bars should be the same width. If one is wider, it might look "bigger" than it really is.
  • Missing Labels: If the axes aren't labeled or there is no title, the graph is meaningless.
  • Incorrect Scales: The gaps between numbers on the axis should be equal (e.g., \(0, 10, 20, 30...\), not \(0, 10, 50, 100\)).

Quick Review: If you see a graph where the bars look very different but the numbers are actually close together, check if the vertical axis starts at zero!

5. Comparing Data Sets

You may be asked to compare two different sets of data shown in these diagrams. To do this effectively:

  • Compare the "Average": Look at where the highest frequency is (the mode) or where the center of the graph lies.
  • Compare the "Spread": Look at the range of the data on the \(x\)-axis. Is one graph wider (more spread out) than the other?
  • Extract Values: Always use specific numbers from the scales to support your answer. Don't just say "it's higher"; say "it is \(20\) units higher."

Key Takeaways

  • Bar Charts are for categories; Multiple bars compare groups; Composite bars show totals.
  • Vertical Line Charts are used for discrete numerical data.
  • Time Series show changes over time; look for upward/downward trends.
  • Frequency Polygons are plotted using midpoints of class intervals.
  • Always check the axes for "tricks" like truncated scales or missing labels.

Note: For more advanced comparative diagrams like comparative pie charts or histograms with unequal widths, please see the specific Higher Tier chapters in this section.