For continuous variables, the Summary Statistics report in the Distribution platform shows the mean, standard deviation, and other summary statistics. You can control which statistics appear in this report by selecting Customize Summary Statistics from the red triangle menu next to Summary Statistics.
Tip: You can specify which summary statistics show in the report each time you run a Distribution analysis for a continuous variable. Select File > Preferences > Platforms > Distribution Summary Statistics, and select the ones that you want to appear.
• Description of the Summary Statistics Report describes the statistics that appear by default.
• Additional Summary Statistics describes additional statistics that you can add to the report using the Customize Summary Statistics window.
Mean
Estimates the expected value of the underlying distribution for the response variable, which is the arithmetic average of the column’s values. It is the sum of the nonmissing values divided by the number of nonmissing values.
Std Dev
The normal distribution is mainly defined by the mean and standard deviation. These parameters provide an easy way to summarize data as the sample becomes large:
– 68% of the values are within one standard deviation of the mean
– 95% of the values are within two standard deviations of the mean
– 99.7% of the values are within three standard deviations of the mean
Std Err Mean
The standard error of the mean, which estimates the standard deviation of the distribution of the mean.
Upper and Lower Mean Confidence Limits
The 95% confidence limits about the mean. They define an interval that is very likely to contain the true population mean.
N
The total number of nonmissing values.
N Missing
The number of missing observations.
Sum Weight
The sum of a column assigned to the role of Weight (in the launch window). Sum Wgt is used in the denominator for computations of the mean instead of N.
Sum
The sum of the response values.
Variance
The sample variance, and the square of the sample standard deviation.
Skewness
Measures sidedness or symmetry.
Kurtosis
Measures peakedness or heaviness of tails. See Kurtosis for formula details.
CV
The percent coefficient of variation. It is computed as the standard deviation divided by the mean and multiplied by 100. The coefficient of variation can be used to assess relative variation. For example, it can be used when comparing the variation in data measured in different units or with different magnitudes.
N Zero
The number of zero values.
N Unique
The number of unique values.
Uncorrected SS
The uncorrected sum of squares or sum of values squared.
Corrected SS
The corrected sum of squares or sum of squares of deviations from the mean.
Autocorrelation
(Appears only if you have not specified a Frequency variable.) First autocorrelation that tests if the residuals are correlated across the rows. This test helps detect non-randomness in the data.
Minimum
Represents the 0 percentile of the data.
Maximum
Represents the 100 percentile of the data.
Median
Represents the 50th percentile of the data.
Mode
The value that occurs most often in the data. If there are multiple modes, the smallest mode appears.
Trimmed Mean
The mean calculated after removing the smallest p% and the largest p% of the data. The value of p is entered in the Enter trimmed mean percent text box at the bottom of the window. The Trimmed Mean option is not available if you have specified a Weight variable.
Geometric Mean
The nth root of the product of the data. For example, geometric means are often used to calculate interest rates. The statistic is also helpful when the data contains a large value in a skewed distribution.
Note: Negative values result in missing numbers, and zero values (with no negative values) result in zero.
Range
The difference between the maximum and minimum of the data.
Interquartile Range
The difference between the 3rd and 1st quartiles.
Median Absolute Deviation
(Does not appear if you have specified a Weight variable.) The median of the absolute deviations from the median.
Proportion Zero
The proportion of nonmissing values that are equal to zero.
Proportion Nonzero
The proportion of nonmissing values that are not equal to zero.
3*StdDev
The value of 3 times the standard deviation. You can set the value of the multiplier using the K Multiplier platform preference under Preferences > Platforms > Distribution Summary Statistics.
3*StdDev Above Mean
The value of the mean plus 3 times the standard deviation. You can set the value of the multiplier using the K Multiplier platform preference under Preferences > Platforms > Distribution Summary Statistics.
3*StdDev Below Mean
The value of the mean minus 3 times the standard deviation. You can set the value of the multiplier using the K Multiplier platform preference under Preferences > Platforms > Distribution Summary Statistics.
Robust Mean
The robust mean, calculated in a way that is resistant to outliers, using Huber's M-estimation. See Huber and Ronchetti (2009).
Robust Std Dev
The robust standard deviation, calculated in a way that is resistant to outliers, using Huber's M-estimation. See Huber and Ronchetti (2009).
Enter (1-alpha) for mean confidence interval
Specify the alpha level for the mean confidence interval.
Enter trimmed mean percent
Specify the trimmed mean percentage. The percentage is trimmed off each side of the data.
The red triangle menu next to Summary Statistics contains these options:
Customize Summary Statistics
Select which statistics you want to appear from the list. You can select or deselect all summary statistics.
Show All Modes
Shows all of the modes if there are multiple modes.
For statistical details, see Statistical Details for Summary Statistics.