[eDebate] translating point scales

jregnier at gmail.com jregnier
Sun Nov 1 11:24:11 CST 2009


I'll admit that I didn't go back in the archives to read everything that  
has been written on this topic, so maybe my concern has already been  
addressed.

It seems to me that as long as we keep trying to "translate" point scales,  
the problem of point differentiation can't really be dealt with. The  
Harvard Tournament's practice of establishing a strict ratio between the  
two scales is the worst form of this. All it does is lop off the first  
digit of the 30pt scale and then move the decimal over one point to the  
right (27.5 => 7.5 => 75). This may make us feel better (because we can say  
that the difference between speakers is "5 points" instead of ".5 points"),  
but the relative difference is the exact same. There is no difference  
between the two ratios, which means that there is no functional difference  
at all. We could just as well be using a 10pt scale (with tenths).

Speaker points have become too self-referential. I think there is a  
tendency when deciding speaker points to jump too quickly to the number:  
that was a 28 quality speech, this is a 29.5 quality speech, etc. As long  
as the numbers themselves are the reference point, new point scales will  
only replicate the problem. In other words, we should never ask  
ourselves "What does a 27 look like in the new scale?" The problem is not  
the scale. The problem is the set of reference points. If we want to  
differentiate more between speakers, then what we need is a new rubric for  
interpreting those differences. And it probably needs to be more detailed  
than "Perfect: 30/100, Excellent: 29.5/95, Very Good: 29/90, Good: 28.5/85,  
Fair: 28/80, Below Average: 27.5/75, Poor: 27/75, Bad: 26.5/26."

Many of us also teach oral communication as our day job. There we evaluate  
speeches based upon a number of categories (eg delivery, composure,  
organization, sound argumentation, use of evidence, etc) each of which can  
be further subdivided into specific evaluable factors. One way of  
establishing a new rubric could be to establish reference points for  
speeches that accomplish each of the criteria to varying degrees.

Another possible rubric would be to base it on projected competitive  
success. For instance, 95s are speeches that we would expect to hear from  
debaters who win major tournaments. 90s are those we would expect to be  
competing in late elims. 85s are for those we think should consistently  
break and sometimes see in middle elims. 80s for people who we think would  
probably break. And so on down the line.

I don't know if either of these is the right solution, but I am pretty  
confident that just switching the numbers up on people without giving them  
a new rubric for evaluation will only mean that we continue to chase our  
tails.

Jason
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.ndtceda.com/pipermail/edebate/attachments/20091101/df136576/attachment.htm 



More information about the Mailman mailing list