The type of data you have determines which chart or table to use. Picking the wrong one is a common mistake in data questions.
Bar charts — comparing frequencies across categories (e.g. favourite subjects). Bars should be equal width with gaps between them.
Pie charts — showing proportions of a whole. Calculate each angle using: $$\text{angle} = \frac{\text{frequency}}{\text{total}} \times 360°$$
Frequency tables — the starting point for almost all data questions. Use grouped frequency tables (with class intervals) when the data range is large.
Two-way tables — show two categories at once. Rows and columns should each add up to their totals — use this to fill in any missing values.
Scatter graphs — plot two numerical variables to look for a relationship. See Correlation for how to interpret them.
Histograms — for grouped continuous data where the class widths may vary. The y-axis shows frequency density, not frequency: $$\text{frequency density} = \frac{\text{frequency}}{\text{class width}}$$
Cumulative frequency graphs — show a running total of frequencies. Used to find the median, quartiles, and interquartile range.
The golden rule: match the chart to the data type. Categorical → bar chart or pie chart. Continuous → histogram. Two variables → scatter graph. Running totals → cumulative frequency.