# Statistics Wizard for Excel - ChartSmartXL

## What is the ChartSmartXL Statistic Wizard?

I've noticed that people often have a hard time deciding what statistics to choose and use. So since the ChartSmartXL already know what data you've selected, they can choose the most likely statistical analysis tools and run them for you.

Decision Trees are Cumbersome. Rather than having to think your way through the forest of decision trees like the ones in Minitab®, wouldn't it be great if the decision tree was already coded into the software? In the ChartSmartXL, it is!

You'll find the Stat Wizard on the ChartSmartXL ribbon menu (XL2007-2016):

### How It Works

The Stat Wizard will look at your data to determine if it has 1, 2 or more columns of data. Then, depending on the number of columns and whether the data has decimals or integers, it will run the following statistics:

1 Column
Descriptive Statistics
1 Sample T Test for Means
2 Columns
F Test for Variances
t Test for Means
Chi-Square Table for independence (if data is integers)
Fisher's test for 2X2 tables
Regression
3+ Columns
ANOVA for Means
Levene's Test for Variances
Chi-Square Table for independence (if data is integers)

### 4-Column Example

By selecting the hardwood concentration data below, the Stat Wizard will run a one-way ANOVA to evaluate the means, Levene's test to evaluate the variances and Chi-Squared table to evaluate independence of the variables.

Select the data with your mouse and click on the ChartSmartXL Menu to choose:
Stat Wizard. The Stat Wizard assumes a significance of 0.05, but you can change this after it has completed.

The Stat Wizard will run the one-way ANOVA:

Then run Chi-Squared Table to evaluate independence of the variables:

Then run Levene's Test to evaluate Variances:

Then report the results:

### 2-Column Example

By selecting two columns, Stat Wizard will use F test, t Test, Chi-square, Fisher's Exact test and regression.

Stat Wizard will run the F Test to evaluate variances:

Then run t test to evaluate the means depending on whether the variances are equal or unequal. (Note: the Stat wizard will not run a paired two-sample t test.)

Then run Chi-Squared table for independence to determine if they are related:

Then run regression for correlation and fit:

And then report results:

### 1-Column Example

By selecting one column of data, Stat Wizard will use Descriptive Statistics and a 1-sample t Test.

Stat Wizard will give you descriptive statistics (note p value is > 0.05 so data is normal):

Using Test Mean of 2500 hours you will get
:

You can also input a Test Stdev to get 1-sample Stdev test.