* Encoding: UTF-8.
* The FREQUENCIES command.

* Examples will use the data from.
GET FILE='C:\Program Files\IBM\SPSS\Statistics\23\Samples\English\Employee data.sav'.

* The FREQUENCIES command is a simple procedure that you will use over and over and over 
  again.  Counts of values, and their related percents and cumulative percents, tell us how
  sample data are distributed, and give us an empirical distribution function and an empirical
  cumulative distribution function (where appropriate), from which we calculate probabilities
  of events or collections of events.

*## Basic Syntax

* Basic use of the dialog box for FREQUENCIES looks like this (see [Basics: Simple Statistics]
(http://www.ssc.wisc.edu/~hemken/spssKB/classintro/spss_students2.html#frequencies-counts-and-percents) .

* If you *Paste* the syntax from the dialog box, SPSS generates this.
FREQUENCIES VARIABLES=gender
  /ORDER=ANALYSIS.

* However, if you look at the **Command Syntax Reference** (a handy shortcut to do this:  type "frequencies" in the syntax editor, then
  press the F1 key, the relevant Help page opens in a web browser), you will see by looking at the syntax diagram (link to
  [SPSS Syntax]) that you could more simply type.
frequencies variables=gender.

* Pasted syntax is verbose - it typically includes a number of default options made explicit.  This makes
  your code harder to read, especially when you come back to it days or weeks later.  If you use the
  Paste button as an aid to writing syntax (it can be faster than trying to look everything up!), you should
  also make an effort to trim out the default specifications.  A good workflow while you are learning to
  write syntax is to Paste - edit - run.

* You can extend this command to give you frequency tables for multiple variables, just include their
  names in the list of variables.  This is more efficient than typing multiple FREQUENCIES commands,
  in a couple of ways.  First, it requires less typing on your part.  And second, all the tables are produced
  with one pass through the data set, rather than multiple passes.
frequencies variables=gender minority jobcat.

*## Default Output.
* Statistical procedures have collections of statistical output that they give you by default.  There are usually
  options that allow you to ask for more (or sometimes, less) output.  (There are numerous ways to customize
  your output through various external scripting languages, as well.)

* The default output for FREQUENCIES is pretty simple, a "Statistics" table that tells you how many missing
  data values are in every frequency table, and the the frequency tables themselves, with counts, percents
  (with the full sample N as the denominator), "valid" percents (just the non-missing N in the denominator),
  and the cumulative percents (again based on the non-missing total).  For any given analysis, you will
  seldom be interested in all four columns, but each may answer important questions that are of interest to you.

*## Useful Options.