@c PSPP - a program for statistical analysis.
-@c Copyright (C) 2017 Free Software Foundation, Inc.
+@c Copyright (C) 2017, 2020 Free Software Foundation, Inc.
@c Permission is granted to copy, distribute and/or modify this document
@c under the terms of the GNU Free Documentation License, Version 1.3
@c or any later version published by the Free Software Foundation;
@cmd{AGGREGATE} both ignores and cancels the current @cmd{SPLIT FILE}
settings (@pxref{SPLIT FILE}).
+@subsection Aggregate Example
+
+The @file{personnel.sav} dataset provides the occupations and salaries of
+many individuals. For many purposes however such detailed information is
+not interesting, but often the aggregated statistics of each occupation are
+of interest. In @ref{aggregate:ex} the @cmd{AGGREGATE} command is used
+to calculate the mean, the median and the standard deviation of each
+occupation.
+
+@float Example, aggregate:ex
+@psppsyntax {aggregate.sps}
+@caption {Calculating aggregated statistics from the @file{personnel.sav} file.}
+@end float
+
+Since we chose the @samp{MODE = REPLACE} option, in @ref{aggregate:res} cases
+for the individual persons are no longer present. They have each been replaced
+by a single case per aggregated value.
+
+@float Results, aggregate:res
+@psppoutput {aggregate}
+@caption {Aggregated mean, median and standard deviation per @exvar{occupation}.}
+@end float
+
+Note that some values for the standard deviation are blank.
+This is because there is only one case with the respective
+occupation.
+
+
@node AUTORECODE
@section AUTORECODE
@vindex AUTORECODE
@end float
+@float Screenshot, autorecode:scr
+@psppimage {autorecode}
+@caption {Autorecode dialog box set to recode @exvar{occupation} to @exvar{occ}}
+@end float
+
Notice in @ref{autorecode:res}, how the new variable has been automatically
allocated value labels which correspond to the strings of the old variable.
This means that in future analyses the descriptive strings are reported instead
For some purposes, neither height nor weight alone is of interest.
Epidemiologists are often more interested in the @dfn{body mass index} which
can sometimes be used as a predictor for clinical conditions.
-The body mass index is defined as the weight of the person in kg divided
+The body mass index is defined as the weight of the person in kilograms divided
by the square of the person's height in metres.
@footnote{Since BMI is a quantity with a ratio scale and has units, the term ``index''
is a misnomer, but that is what it is called.}
so that a more descriptive label appears in subsequent analyses, and this can be seen
in the ouput from the @cmd{DESCRIPTIVES} command in @ref{bmi:res}.
+@float Screenshot, bmi:scr
+@psppimage {compute}
+@caption {Using the dialog box to generate a new variable and compute its values}
+@end float
+
The expression which follows the @samp{=} sign can be as complicated as necessary.
@xref{Expressions} for a precise description of the language accepted.
+Normally it is easiest to enter the code directly, however there is a dialog box
+available if desired. This is illustrated in @ref{bmi:scr}.
+One advantage is that it offers a list of mathematical
+functions which can be selected and pasted into the expression.
@float Results, bmi:res
@psppoutput {compute}
In @ref{count:ex} the @cmd{COUNT} transformation creates a new variable, @exvar{low_counts} and
its values are shown using the @cmd{LIST} command.
+If using the graphic user interface, a two step process must be used to set
+up the @cmd{COUNT} transformation. The first dialog box (@ref{count:scr}) provides for the
+variables to be chosen.
+Then, one must click on the button marked ``Define Values...'' to reveal
+the dialog box for selecting the values to count.
+
+@float Screenshot, count:scr
+@psppimage {count}
+@caption {The variables @exvar{v1}, @exvar{v2} and @exvar{v3} selected, ready
+to define values to count}
+@end float
+
+In this dialog box, you must select the values you wish to count
+--- in this case all values up to and including 2 --- as shown in @ref{count-define:scr}
+and click ``Add''. As many ranges or may be added as you desire.
+When all desired ranges have been added click ``Continue''.
+
+@float Screenshot, count-define:scr
+@psppimage {count-define}
+@caption {Count ``Define Values'' dialog with @samp{lowest thru 2} selected}
+@end float
+
In @ref{count:res} we can see the values of @exvar{low_counts} after the @cmd{COUNT}
transformation has completed. The first value is 1, because there is only one
variable amoung @exvar{v1}, @exvar{v2} and @exvar{3} which has a value of 2 or less.
@cmd{SORT CASES} is a procedure. It causes the data to be read.
@cmd{SORT CASES} attempts to sort the entire active dataset in main memory.
-If workspace is exhausted, it falls back to a merge sort algorithm that
-involves creates numerous temporary files.
+If workspace is exhausted, it falls back to a merge sort algorithm which
+creates numerous temporary files.
@cmd{SORT CASES} may not be specified following @cmd{TEMPORARY}.
@subsection Sorting Example
+In @ref{sort-cases:ex} the data from the file @file {physiology.sav} is sorted
+by two variables, @i{viz@:} @exvar{sex} in descending order and @exvar{temperature} in
+ascending order.
+
+@float Example, sort-cases:ex
+@psppsyntax {sort-cases.sps}
+@caption {Sorting cases by two variables.}
+@end float
+
+In @ref{sort-cases:res} you can see that all the cases with a @exvar{sex} of
+@samp{1} (female) appear before those with a sex of @samp{0} (male).
+This is because they have been sorted in descending order.
+Within each sex, the data is sorted on the @exvar{temperature} variable,
+this time in ascending order.
+
+@float Results, sort-cases:res
+@psppoutput {sort-cases}
+@caption {The @file{physiology.sav} file after sorting.}
+@end float
+
+Note that @cmd{SORT CASES}, like all other transformations, affects only the active file.
+It does not have any effect upon the @file{physiology.sav} file itself. For that, you
+would have to rewrite the file using the @cmd{SAVE} command (@pxref{SAVE}).
+
+When using the graphic user interface, it is often simpler to perform a sort
+directly from the data view.
+To do this, switch to the data view. Select the column corresponding to the
+variable by which you want to sort and click button 1 and then click button 3.
+A popup menu will appear like that shown in @ref{sort-simple:scr}. Select
+either ``Sort Ascending'' or ``Sort Descending'' from this menu.
+
+@float Screenshot, sort-simple:scr
+@psppimage {sort-simple}
+@caption {Sorting the data on a single variable @exvar{height}}
+@end float
+
+However, sometimes you will want to sort on two or more variables, and that is
+not possible using this method. In this case, you must either use some code or
+the ``Sort Cases'' dialog from the Data menu. @ref{sort:scr} shows the dialog
+box set up to perform a sort on both @exvar{sex} and @exvar{height}.
+Note that the order in which you enter the variables is important. In this case,
+the data will be first sorted on @exvar{sex}, and then all cases for which @exvar{sex}
+is the same will then be sorted by @exvar{height}.
+
+@float Screenshot, sort:scr
+@psppimage {sort}
+@caption {Sorting the data on two variables @exvar{sex} and @exvar{height}}
+@end float
projects / pspp / blobdiff