1          OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;
 61         
 62         /********************* 2101f17senic1.sas ***************************/
 63         title 'Open SENIC Data';
 64         /* I reluctantly changed NA to . with a text editor. */
 65         
 66         data senic;
 67              infile '/folders/myfolders/2101f17/openSENIC.data.txt' firstobs=2;
 68                                           /* Skip the first line that R uses */
 69              length region $ 12; /* Otherwise names are truncated to 8 */
 70              input   id region $ mdschl $ census nbeds nurses lngstay age
 71                      xratio culratio infpercent;
 72              label region     = 'Region of U.S.A.'
 73                    mdschl     = 'Medical school affiliation'
 74                    census     = 'Aver # patients in hospital per day'
 75                    nbeds      = 'Average # beds during study period'
 76                    nurses     = 'Aver # nurses during study period'
 77                    lngstay    = 'Av length of hospital stay, in days'
 78                    age        = 'Average patient age'
 79                    xratio     = '# x-rays / # no signs of pneumonia'
 80                    culratio   = '# cultures  / # no hosp acq infect'
 81                    infpercent = 'Percent acquiring infection in hospital';
 82              /* Make dummy variables. Proc reg needs them, proc glm does not. */
 83              if mdschl = 'No' then mschool = 0;
 84                 else if mdschl = 'Yes' then mschool = 1;
 85              /* Region  */
 86              if region = 'NorthCentral' then nc = 1;
 87                 else if region = ' ' then nc = .;
 88                 else nc = 0;
 89              if region = 'Northeast' then ne = 1;
 90                 else if region = ' ' then ne = .;
 91                 else ne = 0;
 92              if region = 'South' then s = 1;
 93                 else if region = ' ' then s = .;
 94                 else s = 0;
 95              if region = 'West' then w = 1;
 96                 else if region = ' ' then w = .;
 97                 else w = 0;
 98         
 99         /* First priority is to check the dummy varables. */
 100        
 
 NOTE: The infile '/folders/myfolders/2101f17/openSENIC.data.txt' is:
       Filename=/folders/myfolders/2101f17/openSENIC.data.txt,
       Owner Name=root,Group Name=vboxsf,
       Access Permission=-rwxrwx---,
       Last Modified=07Sep2017:19:15:05,
       File Size (bytes)=8585
 
 NOTE: 100 records were read from the infile '/folders/myfolders/2101f17/openSENIC.data.txt'.
       The minimum record length was 83.
       The maximum record length was 83.
 NOTE: The data set WORK.SENIC has 100 observations and 16 variables.
 NOTE: DATA statement used (Total process time):
       real time           0.00 seconds
       cpu time            0.01 seconds
       
 101        proc freq;
 
 102             tables mschool*mdschl / norow nocol nopercent missing;
 
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: PROCEDURE FREQ used (Total process time):
       real time           0.05 seconds
       cpu time            0.04 seconds
       
 103        proc freq;
 
 104             tables (nc ne s w) * region / norow nocol nopercent missing;
 105        
 
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: PROCEDURE FREQ used (Total process time):
       real time           0.07 seconds
       cpu time            0.08 seconds
       
 106        proc freq;
 
 107             title2 'Frequency distributions of categorical variables';
 108             tables region mdschl;
 109        
 
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: PROCEDURE FREQ used (Total process time):
       real time           0.03 seconds
       cpu time            0.03 seconds
       
 110        proc means;
 
 111             title2 'Table of means and standard deviations';
 112             var census nbeds nurses lngstay age xratio culratio infpercent;
 113        
 
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: PROCEDURE MEANS used (Total process time):
       real time           0.05 seconds
       cpu time            0.05 seconds
       
 114        proc univariate plot;
 
 115             title2 'Detailed descriptive statistics with boxplots and histograms';
 116             var census -- infpercent;
 117             histogram;
 118        
 
 NOTE: PROCEDURE UNIVARIATE used (Total process time):
       real time           3.15 seconds
       cpu time            1.68 seconds
       
 119        proc sgplot; /* Bar charts for categorical variables */
 
 120             vbar region;
 
 121        proc sgplot;
 NOTE: PROCEDURE SGPLOT used (Total process time):
       real time           0.19 seconds
       cpu time            0.08 seconds
       
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 
 122             vbar mdschl;
 123        
 124        /* Boxplots for different values of a grouping variable */
 125        /* Need to sort by the grouping variable first */
 
 126        proc sort;
 NOTE: PROCEDURE SGPLOT used (Total process time):
       real time           0.16 seconds
       cpu time            0.08 seconds
       
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 
 127             by mdschl;
 
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: The data set WORK.SENIC has 100 observations and 16 variables.
 NOTE: PROCEDURE SORT used (Total process time):
       real time           0.00 seconds
       cpu time            0.00 seconds
       
 128        proc boxplot;
 
 129             plot infpercent*mdschl;
 
 NOTE: Since an input data set was not specified, data set SENIC is assumed as a DATA= data set.
 NOTE: Processing beginning for PLOT statement number 1.
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: PROCEDURE BOXPLOT used (Total process time):
       real time           0.20 seconds
       cpu time            0.10 seconds
       
 130        proc freq;
 
 131             title2 'Relationship between region and medical school affiliation';
 132             tables mdschl*region / norow nopercent chisq;
 133        
 
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: PROCEDURE FREQ used (Total process time):
       real time           0.05 seconds
       cpu time            0.05 seconds
       
 134        proc ttest;
 
 135             title2 'Less risk at Hospitals with Med School Affiliation?';
 136             title3 'Compare t = -2.542';
 137             class mdschl;
 138             var infpercent;
 139        
 
 NOTE: PROCEDURE TTEST used (Total process time):
       real time           0.67 seconds
       cpu time            0.31 seconds
       
 140        proc glm;
 
 141             title2 'Regional differences in average infection risk?';
 142             title3 'Compare F = 2.674, p = 0.0519';
 143             class region;
 144             model infpercent = region;
 145             means region;
 146        
 147        /* Could get the means from proc means, with no extra boxplots */
 148        
 
 NOTE: PROCEDURE GLM used (Total process time):
       real time           0.46 seconds
       cpu time            0.24 seconds
       
 149        proc means;
 
 150             class region;
 151             var infpercent;
 152        
 
 NOTE: There were 100 observations read from the data set WORK.SENIC.
 NOTE: PROCEDURE MEANS used (Total process time):
       real time           0.03 seconds
       cpu time            0.02 seconds
       
 153        proc corr;
 
 154             title2 'Correlation matrix of quantitative variables';
 155             var census nbeds nurses lngstay age xratio culratio infpercent;
 156             /* The nomiss option gives casewise deletion of missing values. */
 157        
 
 NOTE: PROCEDURE CORR used (Total process time):
       real time           0.12 seconds
       cpu time            0.11 seconds
       
 158        proc reg plots=none; /* Suppress diagnostic plots for now*/
 
 159             title2 'Simple regression (One explanatory variable)';
 160             model infpercent = nurses;
 161        
 
 NOTE: PROCEDURE REG used (Total process time):
       real time           0.10 seconds
       cpu time            0.08 seconds
       
 162        proc reg plots=none;
 
 163             title2 'Just the hospital size variables';
 164             model infpercent = census nbeds nurses;
 165        
 166        /* Fit the model with all quantitative variables, and test the size
 167           variables (census nbeds nurses) simultaneously */
 168        
 
 NOTE: PROCEDURE REG used (Total process time):
       real time           0.06 seconds
       cpu time            0.05 seconds
       
 169        proc reg plots=none;
 
 170             title2 'All the quantitative explanatory variables';
 171             title3 'Compare F-test for size: F = 2.7662';
 172             model infpercent = census nbeds nurses lngstay age xratio culratio;
 173             size: test census = nbeds = nurses = 0;
 174        
 
 NOTE: PROCEDURE REG used (Total process time):
       real time           0.08 seconds
       cpu time            0.08 seconds
       
 175        proc reg plots=none;
 
 176             title2 'Full model including categorical variables';
 177             title3 'Reference category for region is South';
 178             model infpercent = census nbeds nurses lngstay age xratio culratio
 179                   mschool nc ne w;
 180             region: test nc = ne = w = 0;
 181             size: test census=nbeds=nurses = 0;
 182        
 183        /* Proc glm will make dummy variables for you. */
 184        
 
 NOTE: PROCEDURE REG used (Total process time):
       real time           0.11 seconds
       cpu time            0.09 seconds
       
 185        proc glm;
 
 186             title2 'Full model with proc glm';
 187             title3 'Compare tests of region (F=4.0769) and size (F=3.5856)';
 188             class region mdschl;
 189             model infpercent = census nbeds nurses lngstay age xratio culratio
 190                                mdschl region / solution;
 191             /* The solution option requests the regression coefficients;
 192             /* Test region and size*/
 193             contrast 'Test of Region'
 194                region 1 -1  0  0,
 195                region 0  1 -1  0,
 196                region 0  0  1 -1;
 197             contrast 'Test of size'
 198                census 1,
 199                nbeds  1,
 200                nurses 1;
 201             lsmeans region;
 202        
 
 NOTE: PROCEDURE GLM used (Total process time):
       real time           0.28 seconds
       cpu time            0.20 seconds
       
 203         proc glm;
 
 204             title2 'Try lsmeans another way';
 205             class region;
 206             model infpercent = census nbeds nurses lngstay age xratio culratio
 207                                mschool region;
 208            lsmeans region / at mschool=0;
 209        
 
 NOTE: PROCEDURE GLM used (Total process time):
       real time           0.25 seconds
       cpu time            0.15 seconds
       
 210         proc reg plots=none;
 
 211             title2 'Just hospitals with no medical school affiliation';
 212             where mdschl = 'No';
 213             model infpercent = census nbeds nurses lngstay age xratio culratio
 214                   nc ne w;
 215        
 216        
 217        
 218        OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK;
 231