[ARCHIVED] Quiz Student Analysis CSV file - what do the columns mean?

[bkompe]

The Canvas Guide page https://community.canvaslms.com/t5/Instructor-Guide/Once-I-publish-a-quiz-what-kinds-of-quiz-statist... discusses the "Student Analysis" report that can be downloaded (both for quizzes and surveys).  The resulting CSV file does not have informative headers to enable you to work out the use of each column. (For example, some columns have the header "1" - very informative).

The Canvas guide has a link to a PDF https://s3.amazonaws.com/tr-learncanvas/docs/CanvasQuizItemAnalysis.pdf, The section titled "CSV Information" refers to a number of columns but I find that the information provided is almost unusable as it is impossible to relate the listed calculations and counts to the headers in the CSV file. I wish to provide a summary for staff so that they will be able to interpret their own download instead of having to spend a number of hours trying to work out what the CSV file means so does anyone have a summary of what each column in the CSV file represents?

[ctitmus]

Hi @bkompe,

I downloaded a Student Analysis Report CSV and it the numbered column headers ("1" in the example you shared) refer to the total points possible for the question and each cell below displays the points earned by the student.

As far as your question is concerned with the PDF, those column headers relate to the Item Analysis CSV and not the Student Analysis CSV. Do you have questions about any of the other column headers form the Student Analysis Report CSV?

Cody

[bkompe]

Hi Cody,

Thanks very much for your reply. You are correct that the PDF refers to the Item Analysis CSV file.

I would be grateful if you are able to point me to any resource that explains the Student Analysis CSV. I would like to be able to provide an explanation for academics who wish to retain a copy of student responses to a quiz/survey outside of Canvas. I find that the CSV file is almost impossible to interpret (including the poorly-labelled columns following the question columns) unless you are prepared to spend a lot of time comparing the file with the actual student responses and then guessing what each column represents.

Thanks.

Regards, Barbara.

[MegWaraczynski]

I don't have an answer for you but want to add that it would also be very, very useful if the student analysis could be sorted alphabetically by student last name!  Example: sometimes students will type in a slightly misspelled answer in a fill in the blank question.  I want to give them credit.  I get a rather randomly organized list of who did that from the quiz itself, then have to search another rather randomly organized student analysis to see whether each student's answer is accurate enough to give credit.  Alphabetizing both lists would save a lot of time.

[bkompe]

Hi Cody,

Thanks for that. I suppose my gripe is that when the Student Analysis report is downloaded, the columns are titled very badly and give no information about what each column represents. It looks as if there has been no work done on formatting of the report since the data download was originally designed.

It would be very valuable if someone has produced a key to what each column means rather than having to sit down each time and try to nut it out.

Barbara.

[MTChan]

My guess is as follows:

1. The first few column headings are self explanatory.
2. After the "attempt" column, the heading is an id followed by the question, e.g. 487101:What is ...; the id appears to be related to the question sequence in the question bank (not confirmed). If this column is a row of a student then it is the answer selected by the student.
3. The next column is the no. of points of the question. If this is a student row then the point is either 0 or the score.
4. These two columns would repeat until it comes to the score columns showing no. of correct, no. of incorrect and the score

This works for me as I try to break down the total score into different categorised scores according to the question type.