10

We're building a teaching platform for projects, and one of the goals is to provide rich feedback to students concerning their performance.

I'm fairly data-driven as a student and always appreciate seeing a lot of data about each project submission (min/mean/median/max/histogram/percentile/rank...)

My advisor argues that this drives competitiveness unnecessarily, and while it may help the best students in the class, it would discourage those who are struggling. He would prefer to optimize the learning experience for each student in the class.

Is there published data or science of learning research regarding the tradeoffs between driving competition by revealing grades and ranks after every homework/project submission?

Improve this question
3
  • To describe comparisons with other students as "data-driven" I think puts a too innocent and misleading face on it. One might argue abstractly that "having information" is good, but what the question refers to as "data" is only information about (the taken-for-granted) competition. It is not about the subject matter, for example. Unless the subject matter is an otherwise meaningless filter (as, for example, many low-level math classes often are), acquisition of the subject matter should be primary, and could be excellent even if/while everyone else "does better" in the "competition".
    – paul garrett
    Jun 28, 2016 at 22:30
  • @paulgarrett, We're trying to tie each graded component to a well-defined learning objective (LO). The assessments are online and adaptive to ensure that the grade matches the acquisition of an LO. The absolute score on a submission is still the primary identifier of whether a student has learned the subject matter (and we have this down). (1/2)
    – Jedi
    Jun 29, 2016 at 0:05
  • My proposal is that by providing relative score indicators (ranks/percentiles/distributions), students will be driven to perform "better". This could be measured by A/B testing with two groups of students in the same cohort and comparing: i) the mean/median scores of the two groups, and ii) the number of extremely low/high scores per assignment in this cohort. (2/2)
    – Jedi
    Jun 29, 2016 at 0:05

1 Answer 1

Reset to default
2

If you Google something like "pros and cons of ranking students" or something similar you will find several websites and other resources that deal with this topic.

Ranking students is common in education. The best examples are standardized tests that always include a percentile ranking (SAT, ACT, GRE, etc.). This is one form of norm-referencing, which is a common tool used in assessing. However, since standardized testing is nationwide, it is hard to be too competitive with the person sitting next to you during the exam, since you probably do not even know their name.

Within the classroom context, ranking could be detrimental if there is a higher consequence of the ranking. For example, only the top 20% will pass and the rest will fail. This could lead to some cutthroat behavior and really lead to a negative learning environment. If the ranking lacks deeper implications, students may not even pay attention to it.

In the end, whether or not to rank or not depends primarily on a person's philosophy of education. Some believe this is beneficial while other do not and there is empirical support for both positions. It is impossible to say that this should always be used or never be used.

Improve this answer
3
  • 1
    Thanks @DarrinThomas. Even with your search and other similar searches, I did not find anything that was very relevant from any "good" publication (SIGCSE, L@S, ICLS, JEE, RER). I should add that the ranking is not meant to have any deeper implications. You mentioned: Some believe this is beneficial while others do not -- I'm hunting down published research on either side of this divide.
    – Jedi
    Jun 29, 2016 at 0:41
  • There is not much on class ranking. However, there is a lot about ranking universities. It might be useful to look at ranking universities and see if you can extrapolate principles for class ranking
    – Darrin Thomas
    Jun 29, 2016 at 0:51
  • I don't think it maps well. The absence of existing publications is an opportunity for me though :-)
    – Jedi
    Jun 29, 2016 at 1:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .