The following article was originally published on my blog, The Chemical Statistician.   I previously explained the statistics behind box plots and how to produce them in R in a very detailed tutorial.  I also illustrated how to produce side-by-side box plots with contrasting patterns in R. Here is an example of how to make box plots in SAS using the VBOX statement in PROC SGPLOT.  I modified the built-in data set SASHELP.CLASS to generate one that suits my needs. The PROC TEMPLATE statement specifies the contrasting colours to be used for different classes.  I also include code for exportingthe result into a PDF file using ODS PDF.  (I used varying shades of grey to allow the contrast to be shown when printed in black and white.)   There are some very useful SAS documents on the different colours and marker symbols.  There is a separate documentation page colour schemes in SAS/GRAPH programs, and it has a specific section on grey-scale colour codes. ods path(prepend)                work.template(update);   proc template;                define style mystyle;                parent = styles.printerl;      class graphdata1           /    color = grayaa                                    contrastcolor = gray00                                    markersymbol = 'circle';      class graphdata2           /    color = graycc                                    contrastcolor = grayaa                                    markersymbol = 'square';                end; run;     data heights;               set sashelp.class;                if age < 13                                then teenager = 'Pre-Teenager';      else teenager = 'Teenager';           if sex = 'F'                                    then gender = 'Female';                 else gender = 'Male';                  label            gender = 'Gender'                                                          teenager = 'Age Group';      keep height gender teenager; run;        ods pdf file = 'INSERT YOUR DIRECTORY PATH HERE\heights.pdf'                notoc style = mystyle;     proc sgplot                data = heights;                vbox height /                          category = gender                          group = teenager; run;     ods pdf close; ... View more

The following blog post was originally published on my blog, The Chemical Statistician.   Introduction A common task in data manipulation is to obtain all observations that appear multiple times in a data set - in other words, to obtain the duplicates.  It turns out that there is no procedure or function that will directly provide the duplicates of a data set in SAS*. *Update: As Fareeza Khurshed kindly commented, the NOUNIQUEKEY option in PROC SORT is available in SAS 9.3+ to directly obtain duplicates.   The Wrong Way to Obtain Duplicates in SAS You may think that PROC SORT can accomplish this task with the nodupkey and the dupout options.  However, the output data set from such a procedure does not have the first of each set of duplicates.  Here is an example. data test;           input var1 $;           cards;                111                222                333                444                444                444                555                555                666                666                666                777                888                888                999                999                999; run;   The data set test has 5 sets of duplicates.  A simple PROC FREQ shows the frequency counts of each possible value of var1. proc freq           data = test;           tables var1 / nocum nocol norow nopercent; run;   111 1 222 1 333 1 444 3 555 2 666 3 777 1 888 2 999 3   Here is the WRONG way to get the duplicates from test. proc sort           data = test                nodupkey                     dupout = dups;           by var1; run; proc print           data = dups; run;   1 444 2 444 3 555 4 666 5 666 6 888 7 999 8 999 The PROC FREQ of the data set dups will show a different frequency count than that of test as shown previously. 444 2 555 1 666 2 888 1 999 2 When you use nodupkey and dupout in PROC SORT, you will remove the first of each set of duplicates in the output data set.  That's why each duplicated value has a frequency in dups that is one less than its frequency in test.     The Right Way to Obtain Duplicates in SAS   To obtain ALL duplicates of a data set, you can take advantage of first.variable and last.variable.  Here is the code to do it with the above example data set of test; you will get both the single observations and the duplicate observations.   (Update: As Robert Matthews commented, you can alter this code to just obtain the duplicates - this would be more efficient if you have many single observations and few duplicate observations, and you don't care about the singles.)   data single dup;           set test;           by var1;           if first.var1 and last.var1                     then output single;           else output dup; run;     Here is the logic behind the code.  We are producing 2 data sets: single for the observations that appear only once, and dup for the duplicate observations.   If first.var1 and last.var1 both equal to "1" for the same observation, then there is no duplicate, and this observation belongs to single. Otherwise, it's a duplicate, and this observation belongs to dup.   Here is what you get as the output data set. 1 444 2 444 3 444 4 555 5 555 6 666 7 666 8 666 9 888 10 888 11 999 12 999 13 999   Notice that it has the correct number of duplicates for each value. ... View more