The Validity of a Test

While browsing the Facebook page for the Badass Teachers Association, I came across the following post:

A large group of Facebook followers commented on Chace's regarding students who simply draw imagines in the "bubbles".  Urban Dictionary refers to this phenomenon as "Christmas Treeing".  It was then stated that these score sheets are removed from the testing sample and simply shredded.  Other readers responded that they are familiar with students who attempt to create patterns on their answer sheets.  I believe that there are also a large number of students who simply make a mistake.  Students fill in answers on the wrong section of an answer sheet.  Students accidentally skip a problem and then all of their answers are out of place for the remainder of a section.  I am certain that in addition to the Christmas Treeers, a large students fall into the "oops" category.  Lastly there are the students who are simply having an "off day".  This group includes the students who woke up or went to bed too late, or perhaps students who got in trouble that morning and have simply shut down.

How can we take the results of a test seriously if we know that student responses are flawed?  Throwing out score sheets of students who simply do not care is not a solution.  These students, who I am assuming fall into the lower academic range, are a testimony to what is being taught and learned in classrooms.  It is possible that the removal of these students scores are raising the overall average.  Is it right to have so much merit leaning on one test?  Perhaps smaller, more frequent assessments gives a larger data pool for each student, allowing one to average their test scores.

In my classroom, I am under the approach that more assessment is better.  No, I don't test or quiz my students daily.  However, I do check in with them, near daily, to assess their understanding.  This allows students to have a large range of grades when I am calculating quarter or semester grades.  It's possible that a student who ends up with an A had an F or two along the way.

I am in favor of some testing. It is necessary to track student progress over time, and also to evaluate the efficiency of a school as a teaching institution.  However, I don't think that testing once a year is the way to accurately obtain this data.