# How to perform statistical analyses

After you have entered data in the spreadsheet, it is advised to save the data on disk. This is done by selecting the command *Save as* in the *File* menu. You will have to enter a name for your data file in the *File selector box*.

To load a data file from disk, select the command *Open* in the *File* menu. Next, select the name of the file in the *File selector box*. For example, you can select the data file *Patients - sample data.mc1*. This file contains data on length and weight for a number of persons.

When you want to obtain summary statistics for the variable *Weight*, select the command *Summary statistics* in the *Statistics* menu. The following dialog box appears on the screen.

In this dialog box you can identify the variable to be analyzed.

You can click the button to obtain a list of variables.

From this list you can select a variable by clicking the variable's name.

If the variable requires a logarithmic or square root, or any other mathematical transformation, then you can enter a formula in the *variable* field:

SQRT(WEIGHT)

LOG(LENGTH)

or you can enter a formula combining different variables, e.g.

WEIGHT/LENGTH

By doing so, new variables will be added to the variables list: SQRT(WEIGHT), WEIGHT/LENGTH, etc..

Optionally, you may also enter a filter in the *Filter* field of the dialog box, in order to include only a selected subgroup of cases in the statistical analysis.

The filter may contain a combination of different criteria, using the AND and OR functions (see logical functions):

These expressions may make use of all mathematical and other spreadsheet functions available.

For an overview of mathematical operators and functions available in MedCalc, refer to the following pages: operators, mathematical functions, string functions, logical functions, and date functions.

After you have entered a filter in a dialog box (in order to include only a selected subgroup of cases in the statistical analysis), this filter will be "remembered" by the program and will be selectable in the *Filter* list.

After you have clicked OK, the following statistics are calculated and displayed in the *results* window:

For more details about the displayed statistics, see Summary statistics in the Statistics menu.

Note that when you change the data in the spreadsheet, this will not automatically cause a recalculation of the displayed statistics, but you have to repeat the statistical analysis for the new data.

To facilitate this, use the *Saved tests and graphs* feature: by clicking the button in the toolbar you can save the analysis so you can recall the analysis later, possibly with new data. After you have clicked this button the graph is added to the *Saved tests and graphs* list in the Contents bar, with a default name, which you can edit.