Chart types

The essentials

Matplotlib's pyplot module provides functions for many chart types. Each takes data for the x and/or y axis and draws the chart; plt.show() displays it. A Pandas Series or DataFrame can also be plotted with df.plot(kind=...), where kind can be 'line', 'bar', 'barh', 'hist', 'pie', 'box' or 'scatter'.

Key points

• Line plot: plt.plot(x, y). Shows change or trend over time, such as monthly temperature.
• Bar graph: plt.bar(x, height) draws vertical bars; plt.barh() draws horizontal bars. Compares quantities across categories, such as marks in subjects.
• Histogram: plt.hist(data, bins=5). Shows the distribution of a continuous variable by grouping values into intervals (bins); bars touch each other.
• Pie chart: plt.pie(values, labels=...). Shows the share of each part in a whole; autopct shows percentages.
• Frequency polygon: a line graph that joins the midpoints of the tops of the bars of a histogram. It shows the shape of a distribution and is useful for comparing two distributions.
• Box plot: plt.boxplot(data). Shows the five-number summary: minimum, first quartile, median, third quartile and maximum, and can show outliers as separate points.
• Scatter plot: plt.scatter(x, y). Shows the relationship (correlation) between two numeric variables, such as height and weight.

Exam tips

• Purpose decides the chart: trend → line; compare categories → bar; distribution → histogram; parts of a whole → pie; relationship → scatter; spread and outliers → box plot.
• In a histogram the x axis is a numeric range; in a bar chart it shows categories.

Common mistakes

• Using a pie chart for data that does not add up to a meaningful whole.
• Confusing a histogram with a bar chart because both use bars.