Comparison of Statistical Data via Graphical Presentation

Graphs are not only a way of presenting a single data set clearly — they are one of the most effective ways of comparing two or more data sets at a glance, without a reader having to mentally process rows of numbers. This chapter's worked examples and exercises use exactly the chart types already covered above for this comparative purpose, and it is worth drawing the comparison techniques together explicitly:

• Multiple bar diagrams (Section 2) directly compare the same categories across two or more data series — which branch, product, or year performed better, and by how much, is visible bar-by-bar (Worked Example 3, Exercise Question 3).
• Multiple line graphs (Section 1) compare how two or more series trend over time on the same axes — useful when the direction of change (rising, falling, fluctuating) matters as much as the raw values.
• A histogram and its frequency polygon drawn on the same axes (Sections 4–5) let a reader compare the 'blocky' and 'smoothed' views of the identical distribution (Exercise Question 5) — and two frequency polygons for two different distributions, drawn on the same axes, let their overall shapes (where each is more spread out, where each peaks) be compared directly, which a pair of side-by-side histograms cannot do as clearly.
• A less-than ogive and a more-than ogive for the same data, drawn on the same axes, cross at exactly one point — whose X-coordinate is the distribution's median (Section 6) — and comparing two datasets' ogives side by side (or overlaid) shows which dataset's values are, overall, concentrated at the higher or lower end (Exercise Question 6).