350 likes | 462 Views
This guide explains how to produce high-resolution bar and pie charts using the GCHART procedure in SAS. Learn to control the displayed statistics, summarize distributions, and visualize averages for different categories. The document covers vertical and horizontal bar charts, pie charts, and options for customizing charts such as defining chart variables, specifying analysis variables, and controlling output aesthetics. With practical examples, you can effectively graphically display the total and average values of various variables, enhancing your data presentations.
E N D
Chapter 15 Introduction to Graphics
Section 15.1 Producing Bar and Pie Charts
Objectives • Produce high-resolution bar and pie charts. • Control the statistics displayed in the chart.
Graphically Summarizing Data • You can use bar or pie charts to graphically display the following: • distribution of a variable’s values • average value of a variable for different categories • total value of a variable for different categories
Specifying a Chart • When you use the GCHART procedure, perform the following tasks: • specify the physical form of the chart • identify a chart variable that determines the number of bars or pie slices to create • optionally identify an analysis variable to use for calculating statistics that determine the height (or length) of the bar or the size of the slice • By default, the height, length, or size represents a frequency count (N).
The GCHART Procedure • General form of the PROC GCHART statement: • Use one of these statements to specify the desired type of chart: PROC GCHART DATA=SAS-data-set; HBAR chart-variable . . . </options>; VBAR chart-variable . . . </options>; PIE chart-variable . . . </options>;
Chart Variable • The chart variable • determines the number of bars or slices produced within a graph • can be character or numeric.
Vertical Bar Chart • Produce a vertical bar chart that displays the number of employees in each job code. proc gchart data=ia.crew; vbar JobCode;run; JobCodeis the chart variable.
Horizontal Bar Chart Produce a horizontal bar chart that displays the number of employees in each job code. proc gchart data=ia.crew; hbar JobCode;run; JobCodeis the chart variable.
Pie Chart Produce a pie chart that displays the number of employees in each job code. proc gchart data=ia.crew; pie JobCode;run; JobCodeis the chart variable.
Character Chart Variable • If the chart variable is character, a bar or slice is created for eachuniquevariable value. • The chart variable is JobCode.
Numeric Chart Variable • For numeric chart variables, the variables are assumed to be continuousunless otherwise specified. • Intervals are automatically calculated and identified by midpoints. • One bar or slice is constructed for each midpoint.
Numeric Chart Variable • Produce a vertical bar chart on the numeric variable Salary. proc gchart data=ia.crew; vbar Salary;run; Salaryis the chart variable.
The DISCRETE Option • To override the default behavior for numeric chart variables, use the DISCRETEoption in the HBAR, VBAR, or PIE statement. • The DISCRETEoption produces a bar or slice for each unique numeric variable value; the values are no longer treated as intervals.
Numeric Chart Variable Produce a vertical bar chart that displays a separate bar for each distinct value of the numeric variable Salary. • Salary is the chart variable,but theDISCRETEoption modifies how the values are displayed. proc gchart data=ia.crew; vbar Salary / discrete;run;
Summary Statistic • By default, the statistic that determines the length or height of each bar or size of pie slice is a frequency count (N).
Analysis Variable • To override the default frequency count, you can use the following HBAR, VBAR, or PIE statement options: SUMVAR=analysis-variable TYPE=MEAN | SUM
SUMVAR= and TYPE= Options • If an analysis variable is specified, the default value of TYPE is SUM. • If an analysis variable is not specified, the default value of TYPE is FREQ.
Using an Analysis Variable • Produce a vertical bar chart that displays the average salary of employees in each job code. proc gchart data=ia.crew; vbar JobCode / sumvar=Salary type=mean;run;
RUN-Group Processing • PROC GCHART supports RUN-group processing, which means that • the procedure executes the group of statements following the PROC statement when a RUN statement is encountered • additional statements followed by another RUN statement can be submitted without resubmitting the PROC statement • the procedure stays active until a PROC, DATA, or QUIT statement is encountered.
Pie Chart Produce a pie chart that displays the total salary of employees in each job code. proc gchart data=ia.crew; pie JobCode / sumvar=Salary type=sum; format Salary dollar8.; run;
Pie Chart • You can use the FILL= option to specify whether to fill the pie slices in a solid (FILL=S) or crosshatched (FILL=X) pattern. pie JobCode / sumvar=Salary type=sum fill=x; format Salary dollar8.; run;
Exploding a Pie Slice • You can highlight individual slices of a pie chart by moving them away from the rest of the pie with the EXPLODE= option. pie JobCode / sumvar=Salary type=sum fill=x explode='PILOT3'; format Salary dollar8.; run;quit;