A questionnaire has been given to men in South Africa who are HIV positive and have an active genital ulcer. The baseline and lab data collected includes age, whether they are on antiretroviral medication (ARVs), their ethnic group, the area/size of their ulcer in millimeters, their plasma HIV viral load , herpes 2 status, and whether their ulcer is shedding HIV.

The variable names in the SAS dataset are as follows:

AGE ARV ETHNIC AREA HIVLoad HERPES HIVShed

Coding Guide

Ulcer shedding HIV:1=Yes 0=No

Taking antiretrovirals:1=Yes 0=No

What is your Ethnicity? 1=Xhosa2=Zulu

3=Sotho 4=Tsonga 5=Other

Herpes 2 Status: 1=Positive 0=Negative

The data is stored in SAS dataset called baseline.sas7bdat in a folder on your computer (just make up a name of the folder).

Which SAS program would accomplish the following below? (2 points each)

1.Create user-defined formats for the variables listed in the coding guide above.

2.Use a Data step to read the permanent SAS data set into a temporary dataset called HIV. Use a libname statement consistent with the version of SAS you have been using this semester.

3.Assign the user defined formats to the appropriate variables in the data step.

4.Have SAS draw a histogram of plasma HIV viral load.

5.Have SAS examine whether there is a relationship between HIV shedding and Herpes result. Request the odds ratio and the p-value for the chi-square test.