topic Re: Multiple types of ANALYSIS in PROC REPORT in SAS Studio

Multiple types of ANALYSIS in PROC REPORT

marianhabesland — Tue, 21 Nov 2017 20:56:45 GMT

I am trying to create this output in my PROC REPORT:

Blood Pressure Summaries by Race

	Average Pressures
	Systolic	Diastolic
Race Code	N	MEAN	STD	MEAN	STD
Asian	11	117.1	11.1	69.1	9.0
Black	83	123.3	15.3	74.0	7.0
Other Race	30	115.2	15.5	69.8	11.0
White	449	120.8	15.5	70.0	7.6
	573	120.8	15.4	70.6	7.8

*The format of this shows up kinda off. The systolic and diastolic labels are supposed to be over the Mean and STD variables.

So far I have this code:

options fmtsearch = (HypTabs.HypFormats WORK LIBRARY);
proc format;
value $RaceCd
' ' = 'Unknown'
'A' = 'Asian'
'B' = 'Black'
'O' = 'Other Race'
'W' = 'White'
other = 'Unexpected Value';
run;

title 1 "Blood Pressure Summaries By Race";
proc report
data = HypAnl.HypPrimAnl
nowindows
split = '/';
column RaceCd N ("Average Pressures" SBP DBP);

define RaceCd / groip 'Race Code';
define SBP / analysis mean format = 5.1 'MEAN' analysis std format =4.1 'STD';
define DBP / analysis mean format = 4.1 'MEAN' analysis std format =4.1 'STD';

rebreak after / summarize;

format RaceCd $RaceCd.;
run;

The issue is getting the MEAN and STD variables split up in the table, because now only one is showing up. Any ideas?

Re: Multiple types of ANALYSIS in PROC REPORT

Cynthia_sas — Wed, 22 Nov 2017 01:43:25 GMT

Hi:

Your DEFINE statements are incorrect. Since you did not post data, I used SASHELP.HEART which has diastolic and systolic measurements, but there is no ethnicity variable in SASHELP.HEART, so I used CHOL_STATUS for categories/grouping. I imagine that you want something like this:

which I produced with this code (I added an overall summary line just in case you wanted something like that too):

title 'Using SASHELP.HEART';
proc report data=sashelp.heart;
where agechddiag gt 65;
column chol_status N
       ('Average Pressures' Diastolic,(mean std) Systolic,(mean std));
define chol_status / group;
define N / 'Count';
define Diastolic / analysis 'Diastolic';
define Systolic / analysis 'Systolic';
define Mean / 'Mean';
define std / 'Std';
rbreak after / summarize;
run;

PROC REPORT allows you to "cross" multiple statistics "under" an analysis variable as I show above. So when the COLUMN statement has this:

Diastolic,(mean std) or

Systolic,(mean std)

then I am asking for the MEAN and STD statistics to be calculated using the variable named DIASTOLIC and then again, calculated using the variable SYSTOLIC.

Hope this gives you a better idea of how to get what you want with PROC REPORT.

cynthia

Re: Multiple types of ANALYSIS in PROC REPORT

marianhabesland — Fri, 24 Nov 2017 22:44:25 GMT

I tried this code based off of what you gave me, but it is not working. I get the error message listed below.

OPTIONS FMTSEARCH = (HypTabs.HypFormats WORK LIBRARY);

PROC FORMAT;

VALUE $RaceCd

' ' = 'Unknown'

'A' = 'Asian'

'B' = 'Black'

'O' = 'Other Race'

'W' = 'White'

OTHER = 'Unexpected Value';

RUN;

TITLE1 "Blood Pressure Summaries By Race";

PROC REPORT

DATA = HypAnl.HypPrimAnl;

COLUMN RaceCd N

('Average Pressures' Diastolic,(MEAN STD) Systolic,(MEAN STD));

DEFINE RaceCd / GROUP "Race Code";

DEFINE DBP / ANALYSIS "Diastolic";

DEFINE SBP / ANALYSIS "Systolic";

DEFINE MEAN /"Mean";

DEFINE STD / "STD";

RBREAK AFTER / SUMMARIZE;

I get this error, but I am not sure how to solve it:

WARNING: SBP is not in the report definition.

WARNING: DBP is not in the report definition.

ERROR: A statistic appears above or below a computed variable.

Name Usage

-------------------------------- --------

Diastolic COMPUTED

MEAN STATISTIC

ERROR: A statistic appears above or below a computed variable.

Name Usage

ERROR: A statistic appears above or below a computed variable.

Name Usage

-------------------------------- --------

Systolic COMPUTED

MEAN STATISTIC

Here is part of my data just for reference:

Re: Multiple types of ANALYSIS in PROC REPORT

marianhabesland — Fri, 24 Nov 2017 22:55:24 GMT

Figured out my problem. Was just referencing the variables wrong.