Use error charts to plot means and either confidence intervals or error bars for one or more variables. An error chart illustrates both a measure of central tendency and variability of the data. |

•Groups by Category:

Choose when the category names are in one column and numerical data are in another column.

oMeasurement: Select a column of numeric data.

oCategorical Variable: Enter a column of grouping (categorical). VisualStat displays a separate bar for each category, on the same graph.

•Groups by Columns:

Choose when each row in a column represents a single observation.

oVariables: Enter one or more columns of numerical data. VisualStat displays a separate bar for each column.

Data may be arranged in one of two forms:

•Groups by Category:

You need one column of categorical data and one columns of corresponding numerical data. The resulting chart contains a bar for each category.

•Groups by Columns

You need one or more columns of numeric values to be graphed. Each row of your data should represent a single observation.

•Data Labels:

Use to label bounds or means. After creating a chart, you can change the data labels or their attributes, or add data labels (see Editing Labels).

oBounds: Choose to label interval endpoints.

oMean: Choose to label each mean.

•Data Options

oExclude missing values: Check to excludes rows that have missing values.

oShow ToolTip: Check to display messages in a ToolTip window

oInverted Bar: Specify whether to draw the error bars vertically or horizontally.

•Bars Represent:

Specify the method to use for computing the error bars: Confidence Interval for Mean, Standard Error of Mean or Standard Deviation.

oConfidence Interval for Mean: Choose to display confidence intervals. The confidence intervals assume a normal distribution for the data and use t-distribution critical values.

oStandard Error of Mean: Choose to display standard error bars, where the error bars extend one or more standard errors away from the mean.

oStandard Deviation: Choose to display standard deviation error bars.

Level: Confidence level used when bars display confidence interval for the mean. Typical confidence levels are 90%, 95%, and 99%.

Multiplier: Number of standard deviations or standard errors above and below the mean that each bar represents.

The data represents cell phone towers' signal strength in various directions by more than one carrier. This data is hypothetical and is simply an example of how the Polar chart is used. The first column represents a cellular service provider company name, followed by angular orientation and signal strength.

The data is being grouped by the chart when plotted. AT&T entries are grouped together as are the Verizon entries.

1.Open the DataBook summary.vstz

2.Select the sheet CellPhone

3.Choose the tab Graph, and then the group Chart, which displays a collection of gallery of items

4.Select Stock and then use tool-tip to select Errors Bars.

All available command is displayed in the gallery of graphics

5.Select group Data and subgroup Groups by Category

6.In Measurement, select Strength

7.In Categorical Variable, select Company.

8.Click Finish

Chart window output