## Thursday, 1 September 2011

### Bars & Bubbles!!

Bar Charts

Bar Charts, like pie charts, are useful for comparing classes or groups of data. In bar charts, a class or group can have a single category of data, or they can be broken down further into multiple categories for greater depth of analysis. Bar charts are familiar to most people, and interpreting them depends largely on what information you are looking for. You might look for the tallest bar, the shortest bar, growth or shrinking of the bars over time, one bar relative to another, change in bars representing the same category in different classes.

For each bar in the bar chart, the following statistics are useful:

 Mean the average height of all the bars. Maximum the maximum value (tallest bar) in the series. Minimum the minimum value (shortest bar) in the series. Sample Size the number of values (bars) in the series. Range the maximum value minus the minimum value. Standard Deviation Indicates how widely data is spread around the mean.

Whenever a bar chart is used to represent data the following points should be noted

· Watch out for inconsistent scales. If you're comparing two or more charts, be sure they use the same scale. If they don't have the same scale, be aware of the differences and how they might trick your eye.

· Be sure that all your classes are equal. For example, don't mix weeks and months, years and half-years, or newly-invented categories with ones that have trails of data behind them.

· Be sure that the interval between classes is consistent. For example, if you want to compare current data that goes month by month to older data that is only available for every six months, either use current data for every six months or show the older data with blanks for the missing months.

Bubble Charts

A bubble chart displays a set of numeric values as circles. It is especially useful for data sets with dozens to hundreds of values, or with values that differ by several orders of magnitude.

How bubble charts work?

The circles in a bubble chart represent different data values, with the area of a circle corresponding to the value. The positions of the bubbles don't mean anything, but are designed to pack the circles together with relatively little wasted space.

Because a bubble chart uses area to represent numbers, it is best for positive values. If your data set includes negative values, they will be shown in a different color: a circle for 100 and a circle for -100 will both be the same size, but 100 might be blue and -100 might be red. If your data set has many negative numbers, consider using a block histogram instead.

To see the exact value of a circle on the chart, move your mouse over it. If you are charting more than one dimension, use the menu to choose which dimension to show. If your data set has multiple numeric columns, you can choose which column to base the circle sizes on by using the menu at the bottom of the chart.

To highlight a circle, click it. The highlighted circle will turn orange. To remove the highlight, just click again. To highlight more than one circle, control-click the circles. The bubble chart will display the sum of the values of all the highlighted circles. Your highlights are saved with any comments you make, so you can easily refer to particular circles in the chart. Highlights are also useful for following a particular set of items as you switch between numeric columns.

Some people don’t like using bubble charts because It can be difficult for readers to compare the sizes of bubbles more accurately than bigger, smaller, a little bit bigger, much smaller, etc. As a result, it would be a mistake to rely on bubbles to convey more important differences that those rough, relative judgements. But if the bubbles are used to display the primary dimensions or data of the analysis, that is exactly what they are being used to do.

Written by: Saumya Rauniyar
Group: Marketing1