Programming the statistical procedures from SAS

Follow-up to previous question

Accepted Solution Solved
Reply
Occasional Contributor
Posts: 5
Accepted Solution

Follow-up to previous question

I have been implementing several techniques to address this issue, but I need some help. I am trying to predict students success based on the number of years in high school.  Please see below clarification of my initial post. Number of years is dichotomized 1 = 3years and 2= 4 years. The response variable is ordered based on the 1 to 5, with 5 being the highest level of success. I also have type of school attended which is dichotomized top tier = 1 and lower tier = 2. I also have gender, male and female. The sample size is 56000 representing 24 schools, which randomly selected from 400 schools. So this is what I have:

 

Predictor = number of years (coded 1 and 2)

Outcome = Admissibility to university (coded 1 – 5)

Gender = (female and Male, coded 1 and 2)

School Type (A and B, coded 1 and 2)

 

Run Ordinal Logistics see SAS code and output below:

 

 

ODS LISTING CLOSE;

  ods graphics on;

  ODS RTF FILE = '\\Client\C$\SHS_DATA_CURRENT\LOGIT_YEARS.RTF';

   proc logistic data=SHS plots(only)=(effect(polybar)oddsratio(range=clip))DESCENDING;

      class YEARS(param=ref ref= "4YEARS");

           WHERE YEARS NE ('3N4YEARS');

      model ACCEPT=YEARS / SCALE=NONE AGGREGATE covb;

      oddsratio YEARS ;

           ODDSRATIO ACCEPT; 

           OUTPUT OUT=PREDICTED2 PRED=PRED;

      title PREDICTING STUDENT ADMISSIBILITY TO UNIV. BASED ON YEARS IN HIGH SCHOOL;

   run;

   ods rtf close;

   ods graphics off;

   ods listing;

 

The Problem I am having is the PO assumption is not held ---please see my print out.

 

I have also tried other suggested techniques such as empirical test of parallelism of my variables – see sample code below and out on document

 

proc freq data=SHS;

     table ACCEPT*YEARS / out=os;

WHERE NOT MISSING(ACCEPT);

     run;

 

 PROC SORT DATA = OS; 

BY YEARS;

RUN;

   proc transpose data=os(WHERE=(YEARS NE '3N4YEARS')) out=tran;

     by YEARS; var count;

     run;

 

   data a; set tran;

     const=0;

     c1=log((sum(of col1-col1)+const)/(sum(of col2-col5)+const));

     c2=log((sum(of col1-col2)+const)/(sum(of col3-col5)+const));

     c3=log((sum(of col1-col3)+const)/(sum(of col4-col5)+const));

     c4=log((sum(of col1-col4)+const)/(sum(of col5-col5)+const));

   run;

 

     ODS RTF FILE = '\\Client\C$\SHS_DATA_CURRENT\LOGIT_YEARS.RTF';

   TITLE 'EMPIRACAL PLOTS OF ACCEPT ON YEARS';

   proc sgplot;

     series y=c1 x=YEARS; 

     series y=c2 x=YEARS;

     series y=c3 x=YEARS;

     series y=c4 x=YEARS;

     yaxis values=(-6 to 6);

     xaxis integer;

     run;

 

My question what do I do next. Is this the end of my analysis. Should I continue interpreting my results based on the fact that the lines are parallel?

 

Any help will  be appreciated.

 

 

 


 

 


Accepted Solutions
Solution
‎03-12-2017 08:11 PM
Valued Guide
Posts: 832

Re: Follow-up to previous question

Instead of "follow-up to previous question", how about making a subject line that actually describes the question?

View solution in original post


All Replies
Solution
‎03-12-2017 08:11 PM
Valued Guide
Posts: 832

Re: Follow-up to previous question

Instead of "follow-up to previous question", how about making a subject line that actually describes the question?

☑ This topic is solved.

Need further help from the community? Please ask a new question.

Discussion stats
  • 1 reply
  • 130 views
  • 0 likes
  • 2 in conversation