SAS PROC REPORT PROC TABULATE Computing for Research

SAS PROC REPORT PROC TABULATE Computing for Research I January 27, 2014

PROC REPORT

Proc Report Intro l l Similar to Proc Print, with more control over output appearance and format Produces detail reports (data listings) and summary reports

Proc Report Syntax Basics PROC REPORT data=dataset <options>; COLUMN var 1 var 2 var 3; DEFINE var 1 / <options>; DEFINE var 2 / <options>; RUN;

Proc Report Statement l l l DATA= input data set NOWD/NOWINDOWS Option - non-interactive mode Additional Options - control layout of the report – – – COLWIDTH= MISSING SPLIT=‘character’ SPANROWS (for ODS only) STYLE<(location(s))>=<style-element-name><[styleattribute-specification(s)]> (for ODS only)

Other Statements Column Statement - describes the arrangement of columns and column headers that span >1 column Define Statement – Variable Appearance (format, label, justify) – Variable Usage DISPLAY – default for character variables ANALYSIS – default for numeric variables ORDER – determines row order ORDER=FORMATTED | INTERNAL | DATA | FREQ GROUP – consolidates unique values of GROUP variables into one row Rbreak Statement - produces a default summary at the beginning or end of a report or at the beginning or end of each BY group.

Vital Signs Dataset l l Subject. ID (47 subjects) Visit. ID (1 -16), Visit Nm (Baseline, Day 1 – Day 15) SBP, DBP, Heart Rate, Central Venous Pressure 1 record per subject per visit

Proc Report Code: Example 1 title 'Baseline and Day 1 Blood Pressure' ; proc report data=vitals nowd split='*'; where visitid in (1, 2) and 1020 le subjectid le 1029; columns subjectid visitnm ('Blood*Pressure' SBP DBP); define subjectid / 'Subject' order; define visitnm / 'Visit' ; define SBP / 'Systolic' format=8. center; define DBP / 'Diastolic' format=8. center; run;

Proc Report Output: Example 1 Baseline and Day 1 Blood Pressure Subject 1020 1021 1022 1023 1024 1025 1026 1027 1028 Visit Baseline Day 1 Baseline Day 1 Baseline Blood Pressure Systolic Diastolic 160 80 145 100 151 70 195 105 145 85 135 75 158 94 186 80 146 82 155 85 165 85 134 80 134 82 178 83

Proc Print Comparison: Example 1 proc sort data=vitals; by subjectid; run; proc print data=vitals label noobs; where visitid in (1, 2) and 1020 le subjectid le 1029; var subjectid visitnm SBP DBP; run; Subject. ID 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 Visit. Nm Baseline Day 1 Baseline Day 1 Baseline Day 1 Baseline Systolic Blood Pressure 160 145 151 195 145 135 158 186 146 155 165 134 178 160 180 Diastolic Blood Pressure 80 80 100 70 105 85 75 94 80 80 82 82 85 85 80 82 83 86 75

Proc Report Code: Example 2 ods rtf; title 'Average Blood Pressure by Visit'; proc report data=vitals nowd split='*' style(header)=[fontsize=14 pt textalign=center foreground=white background=black]; columns visitnm ('Average*Blood Pressure' SBP DBP); define visitnm / 'Visit' group order=data; define SBP / 'Systolic' format=8. center mean; define DBP / 'Diastolic' format=8. center mean; rbreak after /summarize; run; ods rtf close;

Proc Report Output: Example 2 Average Blood Pressure by Visit Average Blood Pressure Systolic Diastolic Baseline 159 81 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day 10 Day 11 Day 12 Day 13 Day 14 Day 15 163 169 170 174 173 174 170 166 170 168 161 159 169 83 87 89 91 91 89 90 90 90 88 91 91 90 89 86 88

Proc Means Comparison: Example 2 ods rtf; title 'Average Blood Pressure by Visit'; proc means data=vitals mean; class visitnm; var SBP DBP; run; ods rtf close; Visit. Nm N Obs Variable Label Baseline Day 10 Day 11 Day 12 Day 13 Day 14 Day 15 Day 2 … 47 SBP DBP 46 SBP DBP 33 SBP DBP 26 SBP DBP 24 SBP DBP 19 SBP DBP 47 SBP DBP Systolic Blood Pressure Diastolic Blood Pressure Systolic Blood Pressure Diastolic Blood Pressure Systolic Blood Pressure Diastolic Blood Pressure Mean 158. 9361702 80. 5744681 162. 7826087 83. 3043478 169. 8888889 88. 3888889 166. 3030303 91. 0000000 170. 2692308 91. 0384615 167. 8461538 89. 5769231 161. 4166667 89. 0833333 158. 7368421 85. 6315789 168. 5744681 87. 4680851

Proc Report Resources • SAS Documentation • Learning Proc Report http: //www 2. sas. com/proceedings/forum 2008/170 -2008. pdf • Proc Report Advanced Tutorial http: //www 2. sas. com/proceedings/sugi 28/015 -28. pdf

PROC TABULATE

Proc Tabulate Intro l l Similar to Proc Means and Proc Freq, in tabular format Creates summary reports with descriptive statistics

Proc Tabulate Syntax Basics PROC TABULATE data=dataset; CLASS classification variables; VAR analysis variables; TABLE page dimension, row dimension, column dimension / <options>; RUN;

Class and Var Statements Class Statement l identifies the class variables (character or numeric) l determines the categories that Proc Tabulate uses to calculate statistics l Missing Option - if not included, observations with any missing class variable values are excluded from the table Var Statement l Identifies numeric variables to use as analysis variables.

Table Statement l l l Required for procedure Describes the table to create Format: page variables, row variables, column variables Can include variable names, keyword statistics, and operators Options – – – PRINTMISSTEXT= BOX=

Table Statement Operators l l Comma – starts a new dimension Blank – concatenates elements Asterisk – creates categories from the combination of values of the class variables and calculates appropriate statistics if class and analysis variables are crossed Parentheses – groups elements, associates operator with each element in the group

Table Statement Statistic Keywords N NMISS PCTN COLPCTN ROWPCTN MEAN STD MIN MAX MEDIAN Q 1 Q 3 many more…

Other Statements l l l KEYLABEL – labels a statistic keyword KEYWORD – specifies a style element for keyword headings CLASSLEV – specifies a style element for class variable level value headings

Proc Tabulate Code: Example 1 proc tabulate data=vitals; where visitid in (1, 2) and 1020 le subjectid le 1029; class subjectid visitnm; table all subjectid, (visitnm='Visits')*(n pctn='%') / box=[label='Subjects x Visits'] misstext='MISSING' printmiss; run;

Proc Tabulate Output: Example 1 Subjects x Visits All Subject. ID 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 Visits Baseline N % Day 1 N % 10 52. 63 9 47. 37 1 5. 26 1 5. 26 MISSING 0. 00

Example 1: Proc Freq Comparison proc freq data=vitals; where visitid in (1, 2) and 1020 le subjectid le 1029; table subjectid*visitnm /norow nocol; run; Table of Subject. ID by Visit. Nm Subject. ID(Subject. ID) Visit. Nm(Visit. Nm) Frequency Percent Baseline Day 1 Total 1 1 2 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 Total 5. 26 1 5. 26 10 52. 63 5. 26 1 5. 26 0 0. 00 9 47. 37 10. 53 2 10. 53 1 5. 26 19 100. 00

Proc Tabulate Code: Example 2 title 'Average Blood Pressure Across Baseline and Day 1'; proc tabulate data=vitals; where visitid in (1, 2) and 1020 le subjectid le 1029; class subjectid ; var SBP DBP; table all subjectid, (SBP DBP)*(n mean*f=8. 1 std*f=8. 1); keylabel std='SD'; run;

Proc Tabulate Output: Example 2 Average Blood Pressure Across Baseline and Day 1 All Subject. ID 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 Systolic Blood Pressure Diastolic Blood Pressure N Mean SD 19 158. 9 18. 6 19 83. 6 8. 3 2 160. 0 2 80. 0 2 148. 0 4. 2 2 85. 0 21. 2 2 170. 0 35. 4 2 95. 0 14. 1 2 146. 5 16. 3 2 84. 5 13. 4 2 186. 0 0. 0 2 80. 0 2 146. 0 0. 0 2 82. 0 0. 0 2 160. 0 7. 1 2 85. 0 0. 0 2 134. 0 0. 0 2 81. 0 1. 4 2 169. 0 12. 7 2 84. 5 2. 1 1 180. 0 . 1 75. 0 .

Example 2: Proc Means Comparison proc means data=vitals mean std; where visitid in (1, 2) and 1020 le subjectid le 1029; class subjectid; var SBP DBP; run; N Subject. ID Obs Variable Label 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 2 SBP DBP 2 SBP DBP 2 SBP DBP 1 SBP DBP Systolic Blood Pressure Diastolic Blood Pressure Systolic Blood Pressure Diastolic Blood Pressure Systolic Blood Pressure Diastolic Blood Pressure Mean Std Dev 160. 0000000 80. 0000000 148. 0000000 85. 0000000 170. 0000000 95. 0000000 146. 5000000 84. 5000000 186. 0000000 80. 0000000 146. 0000000 82. 0000000 160. 0000000 85. 0000000 134. 0000000 81. 0000000 169. 0000000 84. 5000000 180. 0000000 75. 0000000 0 0 4. 2426407 21. 2132034 35. 3553391 14. 1421356 16. 2634560 13. 4350288 0 0 7. 0710678 0 0 1. 4142136 12. 7279221 2. 1213203. .

Proc Tabulate Resources l l SAS Documentation Proc Tabulate Introduction http: //www. sascommunity. org/mwiki/images/4/49/Proc_Tabulate_how_to_-_version_2. 0. pdf l Using Style Elements in the REPORT and TABULATE Procedures http: //support. sas. com/resources/papers/stylesinprocs. pdf

Reporting Data l Plan the layout – – l Summary statistics, analysis output or data listings? Content of columns/rows? Order of columns/rows? Font Use a report writing tool

QUESTIONS?