How do I implement a grading rubric in concert with Cody Coursework?

With the introduction of Cody Coursework, I'm often asked how faculty implement a grading rubric, especially mindful of privacy concerns. For this reason, Cody Coursework was designed such that MathWorks systems need never know which programming problems you assign for practice vs. for credit. This way, as a third-party to the school, we can never correlate students with grades, thus protecting privacy.

Answer 1

[MathWorks Community Team]

Cody Coursework provides for exporting all student solution attempts, as input into any further assessment you may wish to conduct. Each row of the CSV export from Cody Coursework represents a student solution attempt.

With this raw data, your grading rubric can be implemented a number of ways, for example

1. MATLAB script
2. Spreadsheet pivot table

This way, you decide which problems were assigned for credit vs. practice, which exceptions to grant, and what weights to place on each as input into final grades.

[C Lira]

It would be very helpful to have a script to extract the student email an submission that: 1) passes the test; 2) was the latest submission for the student.

The csv file is good for importing into a gradebook, but I need to look at successful code. Some students are passing the test suite with code that uses a different route than specified or is poorly written, and I don't think I can write an assert test for that evaluation. It is awkward to view submissions in the spreadsheet. I'll see if I can write something, but if someone already has something I would really appreciate it!

[C Lira]

Here is what I came up with:

% sort spreadsheet by 
% problem ID (asc), then e-mail (asc), 
% then correct (dec), then submission time (dec).
% save as xls or xlsx.
% After running this script, open the .txt file in MSWORD,
% and each solution will start on a new page.
[num,txt,raw] = xlsread('../../../../Desktop/cody.xlsx')
fid = fopen('../../../../Desktop/codyresults.txt','w+');
% number of data rows
nrows = size(raw,1);
prevprid = -1; % previous problem id
prevuid = ''; % previous user id
for i = 2:nrows
    prid = raw{i,10}; %problem id
    if (prid ~= prevprid)
        % output problem title
        fprintf(fid,'############## %s #########\n',raw{i,11})
        prevprid = prid;
    end
    uid = raw{i,1};
    uid = strtok(uid,'@');
    % test if username is different than last row and test passed.
    if (~strcmp(uid,prevuid) && (raw{2,8}==1))
        fprintf(fid,'*********** %s ************\r\n',uid)
        fprintf(fid,'%s\r\n\f',raw{i,3})
    end
    prevuid = uid;
end 
fclose(fid)