Why I am worried about PARCC and other Common Core assessments, Part 1

One reason I am worried about common core style assessments is that I don’t really have much confidence in how such assessments will be scored.  With multidimensional questions (and this is a good thing), students will inevitably have some  multiple interpretations.  I recently used this questions from EngageNY (Grade 6 Math Module 1, Topic C, Lesson 19, or thereabouts)

Screen Shot 2015-03-08 at 4.34.41 PM

When I did the problem I came up with two possible answers, and when grading my student’s quizzes, I noticed that they came up with two more.  Now this isn’t a particularly intricate problem with lots of different interpretations, but the question isn’t especially precise.

The answers I expected included 3 mph faster or 1.6 times faster.  Most of my students wrote 9 minutes faster because they just referred to the actual walk to school which is of course the relevant distance.  Other students wrote 4.5 minute per mile faster, not the way we typically think of speed, but sound proportional reasoning nonetheless.

My worry is that the person whose charge is to write the code that is meant to interpret student responses to such questions simply isn’t going to think much about whether students are demonstrating sound proportional reasoning.  The machine will look for the expected answer, and be done.  Our students, teachers and schools will be told their failing, and it simply won’t be the case.