SAS Programs
Standard Evaluation Processing and Sample Data
To use the Standard Evaluation Processing program on Instruct or Research accounts, you need to upload the program and your data file to your account before proceeding. You can copy and paste the contents of the above SAS program into a text file on your desktop machine - name it stdeval.sas. Then, using FTP, upload both it and your data file (also a text file) to your account.
1. Before you run this program, you must edit the variables listed in the first few lines. NOTE: you can either edit the SAS program before you upload it to your Instruct or Research account, or while it is in your account, by using the PICO editor (type pico stdeval.sas to edit it). The relevant lines read:
%let head = Put Header Text Here ; /* HEADER*/
%let dir = /usr3/netid; /* DIRECTORY OF WORK AND DATA*/
%let data = data.txt; /* NAME OF RAW DATA SET*/
Everything in bold above needs to be edited.
-
%let head = Put Header Text Here
This is the text printed at the top of every page of output and can include any text you like. -
%let dir = /usr3/netid
This should be the directory where the input and output files are located. For a Research account, this will be /usr3/netid - where netid is your Cornell NetID. -
%let data = data.txt
This input file should be text only and should be located in the directory named in the line above.
2. Run the program by typing sas stdeval (or whatever you named the program).
3.The resulting data will be saved in a file called stdeval.lst (or whatever you named the program with .LST as an extension).
Standard Exam Processing and Sample Data
To use the Standard Exam Processing program on Instruct or Research accounts, you need to upload the program and your data file to your account before proceeding. You can copy and paste the contents of the above SAS program into a text file on your desktop machine - name it stdeval.sas. Then, using FTP, upload both it and your data file (also a text file) to your account.
1. Before you use this program, you must edit the variables listed in the first few lines. NOTE: you can either edit the SAS program before you upload it to your Instruct or Research account, or while it is in your account, by using the PICO editor (type pico examdata.sas to edit it). The relevant lines read:
%let outfile= program2.sas; /* NAME OF SAS PROGRAM TO BE CREATED*/
%let dir = /usr3/netid; /* DIR OF RAW DATA*/
%let data = data.txt; /* NAME OF RAW DATA*/
%let weight1 = 1; /* WEIGHT OF EACH QUESTION in subtest 1 */
%let weight2 = 1; /* WEIGHT OF EACH QUESTION in subtest 2 */
%let weight3 = 1; /* WEIGHT OF EACH QUESTION in subtest 3 */
%let weight4 = 1; /* WEIGHT OF EACH QUESTION in subtest 4 */
%let weight5 = 1; /* WEIGHT OF EACH QUESTION in subtest 5 */
%let numkeys = 1; /* NUMBER OF ANSWER KEYS (OR SUBTESTS) */
%let numqu = %eval(200*1); /* NUMBER OF QUESTIONS*/
%let head = Put Header Text Here ; /* HEADER*/
Everything in bold above needs to be edited.
-
%outfile= program2.sas
This output file (program) can be named anything you like, but should have .sas as the extension. -
%dir = /usr3/netid
This should be the directory where the data file is located. For a Research account, this will be /usr3/netid - where netid is your Cornell NetID. -
%let data = data.txt
This input file must be text only and should be located in the directory named in the line above. -
%let weightx = 1
These should be the numeric weight assigned to each problem in each subtest. -
%let numkeys = 1
This should be the number of answer keys (or number of subtests) in the data file - multiple answer keys are used either for different subtests (with different weights), or when you have multiple answers for various questions. -
%let numqu = %eval(200*1)
This should be the total number of exam questions -
%let head = Put Header Text Here
This is the text printed at the top of every page of output and can include any text you like.
2. Run the program by typing sas exam (or whatever you named the program). This will NOT process the data, but rather, create a second program that will process the data.
THEN, you need to run the second program by typing sas program2 to process the data - the program name is designated in the %outfile field - in the example above it is named.
3. The resulting data will be saved in a file called program2.txt (or whatever you named the %outfile with a .txt as an extension).
NOTES:
1. Deleting Questions - If questions are to be removed from
the exam (i.e., after giving the exam you want to remove one), this should
be done BEFORE you submit the bubble sheets for scanning - and it should be noted at that time.
(The office that does the scanning will simply ERASE the answer for each question to be removed in the answer key only.)
2. Multiple Answers - If one or more questions are to have multiple answers, then additional answer keys must be scanned. For example, suppose questions 5 and 10 have answers a and b respectively, but you would like them to have alternate answers of c and d (respectively) as well. You should submit the main answer key with answers a and b, then submit a second answer key with just answers c and d for questions 5 and 10 (all other answers are left blank). You can even submit a third answer key if you another alternate answer is allowed.
Note that this is complicated if you are using other "subtests" because of different weighting of questions (for instance, if the first 10 questions are worth 5 points each, and the next 10 questions are worth 7 points each). In this case, you would have to submit all alternate answers for the first subtest on an answer key separately from the alternate answers for the next subtest. In other words, you would submit 4 answer keys -one main answer key for each subtest, then alternate answers for the first subtest, then alternate answers for the second subtest.
(The scanning office will then scan all main answer keys first, followed by all "alternate" answer keys.)
On This Page
Related Links
Contact Information
Office of the University Registrar
B07 Day Hall
Cornell University
Ithaca, NY 14853
(607) 255-4232
Hours
Mon-Fri 8:00-4:30
Questions?
Email us at:
univreg@cornell.edu.
