###### Comparing GDP, National Income, Person ...

GDP GDP stands for Gross Domestic Product. It refers to the market value... **Read More**

- For numerical data, use a histogram, frequency polygon, or cumulative distribution chart.
- For category-based data, use a bar chart, tree-map, or heat map.
- For unstructured data, use a word cloud.
- For displaying relationships between two variables, use a scatter plot.
- For displaying relationships among more than two variables, use a scatter plot matrix or heat map.
- For displaying numerical data distributions, use a histogram, frequency polygon, or cumulative distribution chart.
- For displaying categorical data, use a bar chart, tree-map, or heat map.
- To draw a comparison over time, use a line chart for two variables and a bubble line chart for three variables.
- For drawing a comparison among categories, use a bar chart, tree-map, or heat map.

The following are the four common pitfalls that we should avoid while using charts and graphs:

- Selecting the wrong chart type will lead to inaccurate interpretation of data.
- Selectively plotting data could prove an inaccurate relationship.
- Data plotted in a truncated graph, e.g., having a y-axis that does not start at zero, could inaccurately show significant differences when there is really just a small difference.
- Scaling the axes incorrectly, such as setting a maximum that is higher than necessary on the y-axis, can result in a graph that appears less steep and less volatile.

QuestionWhich of the following graphs is the

most appropriateto use for plotting stock price change over a five-year period?

- Line chart.
- Bubble line chart.
- Frequency polygon.

SolutionThe correct answer is

A.A line chart is used to display the change of data series over time.

B is incorrect.In a bubble line chart, data points are represented by bubbles of varying sizes. The bubbles represent a third dimension of the data.

C is incorrect.The frequency polygon and the cumulative frequency distribution chart are also line charts, but they represent data frequency distributions.