I tried to merge both datasets based on the variables ClassVal0 and param. For some reason though the coding I used is not working. First I tried to update based using this code below before merging but it said SQL Update: ERROR: Subquery evaluated to more than one row. proc sql; update bout.allm as a set ClassVal0 = (select (ClassVal0) from nhns.chdsimplogtab as b where a.Param = b.Param group by b.Param); quit; Then I just tried to combine them together based on this code below and it just didn't work out with values mismatched. proc sql; create table mamlogistictable as select a.*, b.M_OR_CI, b.M_p_value, b.M_g_p_value from btab.lnelmam_simplogtab a left join bout.allm b on a.ClassVal0 = b.ClassVal0 and a.Param = b.Param order by Param; quit; What am I doing wrong? I'd appreciate the explanation. Thank you ... View more

Hi all, I am having a problem looping through a list for class statement in the proc surveylogistic step.  The code is below %let catclass = agegrp(ref="40-49") raceth(ref="NH White") maritals(ref="Married") educat(ref="Some College/College Grad") wrkstat(ref="Employed") income(ref=">= $50,000") insure(ref="No") hlthcost(ref="No") doctor(ref="No") hltstat(ref="Excellent/Very Good"); %macro ccls(); %do i = 1 %to %sysfunc(countw(&catclass)); %let logref = %scan(&catclass,&i); %end; %mend; %ccls; When I loop through it, I get this error (see below). MLOGIC(REF): Beginning execution. SYMBOLGEN: Macro variable CATCLASS resolves to agegrp(ref="40-49") raceth(ref="NH White") maritals(ref="Married") educat(ref="Some College/College Grad") wrkstat(ref="Employed") income(ref=">= $50,000") insure(ref="No") hlthcost(ref="No") doctor(ref="No") hltstat(ref="Excellent/Very Good") MLOGIC(REF): %DO loop beginning; index variable I; start value is 1; stop value is 29; by value is 1. MLOGIC(REF): %LET (variable name is LOGREF) SYMBOLGEN: Macro variable CATCLASS resolves to agegrp(ref="40-49") raceth(ref="NH White") maritals(ref="Married") educat(ref="Some College/College Grad") wrkstat(ref="Employed") income(ref=">= $50,000") insure(ref="No") hlthcost(ref="No") doctor(ref="No") hltstat(ref="Excellent/Very Good") SYMBOLGEN: Macro variable I resolves to 1 MLOGIC(REF): %DO loop index variable I is now 2; loop will iterate again. MLOGIC(REF): %LET (variable name is LOGREF) SYMBOLGEN: Macro variable CATCLASS resolves to agegrp(ref="40-49") raceth(ref="NH White") maritals(ref="Married") educat(ref="Some College/College Grad") wrkstat(ref="Employed") income(ref=">= $50,000") insure(ref="No") hlthcost(ref="No") doctor(ref="No") hltstat(ref="Excellent/Very Good") SYMBOLGEN: Macro variable I resolves to 2 ERROR: Literal contains unmatched quote. ERROR: The macro REF will stop executing. MLOGIC(REF): Ending execution.   Additionally, when I try it another way, I get another error. Code is below and error is after that %let ref = "40-49" "NH White" "Married" "Some College/College Grad" "Employed" ">= $50,000" "No" "No" "No" "Excellent/Very Good"; %macro ref(); %do i = 1 %to %sysfunc(countw(&ref)); %let logref = %scan(&ref,&i); %end; %mend; %ref; MLOGIC(REF): Beginning execution. SYMBOLGEN: Macro variable REF resolves to "40-49" "NH White" "Married" "Some College/College Grad" "Employed" ">= $50,000" "No" "No" "No" "Excellent/Very Good" MLOGIC(REF): %DO loop beginning; index variable I; start value is 1; stop value is 19; by value is 1. MLOGIC(REF): %LET (variable name is LOGREF) SYMBOLGEN: Macro variable REF resolves to "40-49" "NH White" "Married" "Some College/College Grad" "Employed" ">= $50,000" "No" "No" "No" "Excellent/Very Good" SYMBOLGEN: Macro variable I resolves to 1 ERROR: Literal contains unmatched quote. ERROR: The macro REF will stop executing. MLOGIC(REF): Ending execution.   How do I figure this out?   Thanks for the help. ... View more

Hi all, I'm trying to extract values from a surveyfreq procedure using two types of code structure and they both give out 0 observations. Below are the two statement types.  First DATA &YES &NO &ALLTOTAL; SET COMBINEALLtesTaBC; SELECT (F_&DISEASE); WHEN ('&YES') OUTPUT &YES; WHEN ('&NO') OUTPUT &NO; WHEN ('TOTAL') OUTPUT &ALLTOTAL; OTHERWISE; END; RUN; Second DATA &YES &NO &ALLTOTAL; SET COMBINEALLtesTABC; if F_&disease = "YES" THEN OUTPUT &YES; if F_&disease = "NO" THEN OUTPUT &NO; if F_&disease = "TOTAL" THEN OUTPUT &ALLTOTAL; RUN; What am I doing wrong?  I included the dataset and the log when i run the code. Thank you all. ... View more

Hello, I am trying to left join two datasets but it is failing miserably! Can someone help rectify this? I attached the pics of before and after in a pdf format. In the pics, you can see the simple log reg output and the other is for multiple log reg. The issue comes when I try to combine the dataset to generate one output and my rows don't merge properly resulting in missing values for adjusted odd ratios and p-values. Thank you for the help rendered. proc sql; create table logit_table as select a.*, b.M_OddsRatios_CI , b.M_p_value, b.M_global_P_value from simplelog1 a left join multilog1 b on a.Category=b.Category and a.Demographic_Characteristics=b.Demographic_Characteristics order by f_order; quit;   ... View more