Home Page Overview
New Features Downloads & Prices
P-STAT documentation Technical Support
Contact P-STAT Unistat for Windows

Report writing Commands

LISTING A FILE

The P-STAT LIST command lists data concisely and attractively, automatically formatting the listing for the current output device - the terminal, a disk file or a printer. Conciseness is important when the output is directed to the terminal because typically only 80 columns and 22 lines are available for listings. Attractiveness contributes to the readability and clarity of a listing. Thus, the two prime goals of LIST are:

      a meaningful and attractive layout, and
      a maximum information in the least amount of space.

These goals are implemented as completely as possible. In situations where they are incompatible, economic use of space takes priority. However, there are a number of LIST command options that may be used to change or improve the appearance of the output when that is more important than conserving space. LIST requires only the command itself and the name of the file to be listed:

      LIST File1 $

GENERATING REPORTS

The TEXTWRITER command produces text or reports that summarize the data in a P-STAT system file. The text is formatted much the same way as text produced by a word processing software package, with justification, paragraphs and pagination. In addition, the reports can include character strings, values from the file, and evaluations of complex expressions containing functions and operators.

TEXTWRITER uses the P-STAT Programming Language (PPL) instructions PUT and PUTL to specify the strings, values and expressions. Additional PPL may be included to test values and output appropriate strings. Thus, if Gender equals "M", the string "Mr." is written, but if Gender equals "F", "Ms." is output. Control words format the text and position it in specific columns and lines.

Special features include the use of labels files to supply character string replacement for integer values and PostScript controls to supply font information. Information such as the current date and time can be included by using P-STAT system variables in the PUT statements.

PostScript Plots

Scatter Plots

The PLOT command graphs numeric variables on scaled axes and produces scatter plots. The coordinates of the plotted points are the paired values (X,Y) for two variables. It is possible to produce plots of a single variable, plots of three or more variables, and multiple plots on one set of axes. In addition, separate plots for subgroups may be requested by using the identifier BY in the PLOT command.

Desired plots and options are specified at the subcommand level. Reasonable settings for output parameters are automatically selected by the PLOT command but may be reset by the user.

PostScript support provides camera ready plots whenever there is a PostScript printer available. It is independent of the types of computer or operating system. Plots can be approximated first in character format with the PostScript information added when the test plot is satisfactory. If there is a PostScript previewer (for example, ghostscript) available, the PostScript output can be viewed on the screen before it is printed. Most of the PLOT subcommands are available in both modes; character and PostScript.

One feature that is only available when PLOT is used with PostScript controls is the ability to have multiple plots on a page. In addition these plots can be combined on a page with TEXTWRITER BOXPLOT, and SURVEY command output.

Box Plots

The BOXPLOT command is designed to produce boxplots in PostScript format. It can be used by itself or as part of a PostScript block. The language for BOXPLOT is derived from both the existing BOXPLOT subcommand in EDA and the PLOT command. Most of the EDA BOXPLOT subcommands were retained with the new PostScript features using the language of the PLOT command.

The BOXPLOT subcommand in EDA constructs all of the plots using Tukey's formulas for hinges, outliers, and whiskers. The BOXPLOT command permits a choice of method. The assumed method is TUKEY (also called the STANDARD method). The alternate method is based on percentiles. The choice of method determines the defaults for the printing of outliers and whiskers.