Suggestions for an alternative grading system

Question

I'm designing a leadership development program for a graduate college at a research university. As part of the project, I've designed a summary sheet which each participant will receive. The summary is designed to give them an impression of their overall leadership development. It includes an "overall leadership score," which is currently shown as a number on a 1-10 scale.

My intent is for all students to begin the program at a baseline level of 5. Students who demonstrate multiple positive leadership traits will raise their score above a 5. Students who show a lack of these traits will lower that score.

However, I'd like to replace the 1-10 scale since it will be difficult to keep the baseline or average score of 5 from being equated with a 50, an F, etc. For the same reason, I've avoided the use of a letter scale.

Does anyone have suggestions on an alternative grading system that is well suited to this type of grading task? I'm also trying to find something that won't be too discouraging for students whose leadership scores are below average (seeing a 0 or an F would not get the right message across).

Answer 1

I would suggest a rubric that includes the desired qualities of each category. If you're really worried about the scale, don't include the numbers or letters at all, or change to something like Roman Numerals. I would not include more than four or five categories, unless you can really break down differences. E.g.,

I. Not making progress. The student needs to improve his/her leadership skills, to include participating in mentor/mentee relationships with junior students, or actively seeking out lead roles in the lab.

II. Pre-leadership. The student is new to the lab, and has not yet developed leadership roles. The student should start looking for opportunities for leadership, and should begin considering the type of leader he or she wants to be. The student should thoughtfully observe other leaders and their leadership styles.

III. Making progress. The student is actively engaged in a leadership role in the lab and/or has made mentor/mentee relationships with junior students. The student should continue to develop his/her leadership skills in these roles.

IV. Excellent leadership development. The student has actively developed his/her leadership opportunities, and is capable of independent leadership in the lab. Other students view him/her as a quality leader and are comfortable going to the student for guidance.

V. Phenomenal leader. The student is qualified to independently lead large groups of other students, and is able to mentor others in leadership roles.

In this example, I would start everyone off at (II). Everyone will know they should look for leadership opportunities, and it should be relatively easy to get into the (III) category. Obviously, this is only an example, and can be tailored to your situation.

I'm also trying to find something that won't be too discouraging for students whose leadership scores are below average (seeing a 0 or an F would not get the right message across).

I do think having some sort of forcing function is necessary; otherwise, what is the reason to improve? In my example, you could tone down "Not making progress" (or take it out), but if you do want to show students who aren't improving that they need to, you should have at least one negative category.

Answer 2

It's a challenging question. I believe your goal is to measure progress and not overall leadership-levels. That is, Student A could enter with strong leadership skills and would start at a 5. Student B could enter with weak leadership skills and would also start at a 5. If Student B shows more progress in improving their leadership skills, then Student B should end up with a 7 (for example) while Student A who was a bit lax and did not improve might end up with a 4 or even 3. The goal, as I understand, is to measure improvement only.

If my understanding is correct you have a challenge in that if students know this, they will try to corrupt (by under-performing) the baseline. Then they will naturally appear to have improved more than they really have giving them a better grade not by improving but by gaming your grading system.

You could structure it by learning outcomes whereby points of improvement are each an outcome. For example:

• Outcome 1: Improve ability to communicate to those in subordinate positions
• Outcome 2: Improve......

By this, you do not fix the levels but show what areas in which each student should actually improve. This also makes it extremely clear what the goals are. You could assign a certain number of points per outcome (10 for outcome 1, 5 for outcome 2, etc.) to sum up to a total score as well.

I know this is not a discussion board but I'd love to hear what you ended up doing.

Answer 3

It is not clear what you are trying to accomplish with your grading system. If the point is to evaluate/rank the students then 1-10 is probably as good as anything else. If it is to be used as a formative tool then using an absolute scale is probably less helpful and what the students need to know are their relative strength and weaknesses and areas that they can improve on. What this means is that a student who gets a 9 out of 10 still can improve his/her leadership, similarly a student with a 2 out of 10 likely has leadership areas/traits that can be improved more than others. It is not clear to me that the competition aspect of ranking the students will be beneficial in this case. If you go with a relative scale you could just make the total of the numbers sum to 10 (or 1 or 100). Better would probably be a stacked bar or pie chart so it is clearer that comparisons across individuals are meaningless.

I think the best "grading" system for leadership is probably one that doesn't assign presence/absence of leadership traits, but rather one that characterizes in positive terms the type of leadership each individual is most suited for (e.g. the managerial grid model).