|
Descriptive Statistics includes
the tools shown on the left. These are typical tools for exploring the
descriptive summaries, frequencies, and cross-tabulation tables. These
exploring tools along with graphical tools are not only useful for data
exploration, but also are useful for data cleaning. Various data errors can
be identified through these descriptive tools and can be corrected before any
further analysis. P-P plots and Q-Q plots are useful
for checking the distribution assumption required by statistical
techniques. |
This procedure gives frequency tables. It is usually used for categorical
data. There are three submenus, which are:
|
Statistics: This submenu allows users to choose various descriptive
measures such as average, median, standard deviation, percentiles, and so on |
|
|
Charts: This submenu provides Bar charts and Pie charts for categorical
data and histograms for quantitative data. |
|
|
Format: This submenu allows users to define the format of the output. |
This is usually for continuous (as known as interval or scale) data. This procedure gives options of computing descriptive measurements such as mean, variance, standard deviation, and so on.
This procedure gives several graphical presentation and descriptive
measurements. This is a quick tool for checking some trivial data errors. It
has three submenus:
|
Statistics: This submenu gives descriptive measurements that are also
available in the Descriptive Procedure. One can request for outliers. |
|
|
Plots: This submenu gives several graphical plots for any selected
variable. These plots include stem-leaf plot, histogram and normality plot.
They are useful for checking if a variable satisfies normality assumption for
further analysis. |
|
|
Options: This submenu allows users to decide how the missing data will be
treated. |
This procedure is for exploring the relationship between two or more
categorical variables in cross-tabulation form.
This procedure has three or four submenus. If you have license for Exact Tests,
this will be the fourth submenu. The submenus are
|
Statistics: This submenu allows you to do exact tests. |
|
|
Statistics: This submenu provides many statistical tests that are used for
categorical data analysis, including chi-squared test, Spearman correlation
and many others. |
|
|
Cells: This submenu allows users to choose the information to display in
the output pivot tables. For example, expected counts, row percentage and
column percentage can be requested. |
|
|
Format: This submenu allows users to choose the order of the category in
the output. |
Ratio:
This procedure provides summary statistics (e.g. median, mean, weighted
mean, standard deviation, concentration index) for describing the ratio between
two scale variables. It has one submenu.
Probability Plots:
P-P and Q-Q probability plots are used to determine whether the distribution
of a variable in the data matches a given distribution. If the variable matches
a test distribution, the probability plots cluster around a straight line. The
test distributions include uniform, normal, Student’s t, gamma and beta. The
P-P procedure plots a variable’s cumulative proportions (or empirical
cumulative distribution function) against the cumulative proportions of a test
distribution. The Q-Q procedure plots the quantiles of a variable’s
distribution against the quantiles of a test distribution.