# Summary Statistics

## Contents

## Description

Summary Statistics produces various statistics such as: Mean, Median, Standard Deviation, Min, Max, Range, Standard Error, Counts, Skew, kurtosis, etc to summarise a set of observations in a dataset.

## Inputs

*: a dataset containing the variable(s) to be tested.*

**Dataset Input*** Variable Name*: the variable(s) to be tested – more than one can be selected.

*Analyse your data > Tools > Statistical Analysis > Summary Statistics >*

**enter parameters**

*> Add Tools > Show/Hide > Execute*]

## Outputs

**A summary of all of the variables:**

* Num rows*: Number of rows (observations) in the data

* Num cols*: Number of columns (variables/attributes) in the data

* Means*: Mean value(s) of the variables/attributes

* Medians:* Median value(s) of the variable/attributes

* Quantiles*: 1st (“25%”) and 3rd (75%) quartile values for each of the columns (variables/attributes)

* Storage mode*: The type of value that each variable/attribute is recorded as:

*logical*, i.e. “TRUE” or “FALSE”;

*integer*, i.e. 1, 2, -1, 7;

*double*, i.e. 1.03472, 2.49227; and

*character*i.e. “green”, “red”.

**A summary for each of the variables/attributes:**

* Sd*: Standard deviation for a variable/attribute

^{1}

* Trimmed*: Trimmed mean value for a variable/attribute calculated by dropping a proportion of the observations from both ends of the sample, and can allow you to determine whether long tails/outliers have an impact on the mean value

^{2}

* Mad:* Median absolute deviation

^{3}, is the median value of the residuals of each observation from the sample median for each variable/attribute

* Min:* The minimum value for a variable/attribute

* Max:* The maximum value for a variable/attribute

* Range:* The range of values (Max – Min) for a variable/attribute

* Skew*: The measure of how much a range of values is skewed to the left or right of the mean (asymmetry) for a variable/attribute

^{4}

* Kurtosis:* The measure of how flat or “chopped off” a range of values is around the mean for a variable/attribute

^{5}

* Se:* The standard error of the mean for a variable/attribute

^{6}

* Missing:* number of missing values for a variable/attribute

* Unique:* number of distinct values for a variable/attribute

* Breaks:* Break classes defined by min and max of a variable/attribute with an interval of 5%

* Counts:* Number of observations/rows in each break class for a variable/attribute

* Density:* The percentage of observations/rows within each break class for a variable/attribute

* Mids:* The median value within each break class for a variable/attribute

* Equidistant*: Indicating if the distances between breaks are all the same for a variable/attribute (true/false)