Student Achievement and Test Rewrites

I was first introduced to the idea of allowing students to rewrite tests about 6 years ago.  The rational behind allowing students to rewrite tests or quizzes is simple.  As educators we are looking for evidence that students have mastered a concept or topic.  However there is nothing that says they need to master it the first time through.  In fact, it is more likely that different students will require different time periods to acquire the same knowledge.  By allowing them to re-write a test we provide them with the opportunity to demonstrate their ultimate mastery of a particular concept.  Simple right?

Unfortunately no.  As with most things in education the issue is not quite that black and white.  While that may be the intended outcome, along with this exist numerous unintended consequences like student entitlement and assessment inaccuracies due to what I'll call the lottery effect (I'll explain more about this shortly). It was because of these unintentional outcomes that for many years I questioned the educational value of allowing students more than one opportunity to write a test.  However over the past few months I've refined my own rational and established some guidelines that make it very clear to my students when and under what circumstances they should be allowed the rewrite an assessment.

The first thing educators need to consider is the purpose of the assessment.  I believe that if an assessment is primarily formative (quiz, class assignment, etc) students should have the opportunity to redo it as many times as they feel they need to (with certain conditions of course).  If these assessments are weighted appropriately (being formative they shouldn't count for a substantial percentage of the overall mark) the overall effect on a students grade will be minimal; however the intended outcome of demonstrating student learning will be met.  In contrast, rewriting summative assessments should be approached much more carefully.

As a result of students being allowed to rewrite exams without consequence I have found that many students view them as a sort of "Get out of Fail free card".  If they do poorly on the first assessment they will simply rewrite it again and again until they get a mark they are happy with. I would also argue that the reliability and validity of summative assessments are more or less destroyed when students have more than one or two attempts. Its what I call the lottery effect and its simply a matter of statistics.  Take the lottery for example.  I seem to have terrible luck when it comes to lottery tickets.  On a typical 6/49 ticket I rarely match more than one number.  Lets assume that my 'test' is how many numbers I correctly match.  On my first attempt I get 1 out of the 6.  So next week I 're-test' and still only get 1.  The following week I get 2.  Not satisfied I continue for another month, each week buying 1 ticket and never matching more than 2 numbers.  Finally after 6 attempts I match 4 numbers.  Now I'm satisfied with my 'grade' and decide to stop.  Does this  indicate I have become more proficient at buying lottery tickets? Have I learned anything about buying lottery tickets other than the obvious fact that I should stop wasting my money? (The odds of matching numbers never change. But given enough tickets you will eventually win).  So how can we improve our confidence that learning is happening if students are allowed to retest?

Since September I have implemented what I call a 'two step retest program'.  To qualify for a retest students must first come for extra help on the areas they did poorly in.  However in order to be allowed extra tutorials students first must show me that all their homework is complete.  If students don't complete some sort of review and remediation there is no justifiable reason for them to redo an assessment.  And if students haven't at least attempted all the homework there is no justifiable reason for me to spend extra time helping them.

The 'two step retest program' gives the students a very clear cut set of steps they must complete if they want to retest.  It also has the added benefit of providing additional motivation for students to complete their homework on a daily basis.  They already know the homework doesn't count for much of their grade, but using this system it still has extrinsic value for them.

Of course there are many other circumstances when it may be justified to allow students to rewrite something but at least this way the majority of students know how, when and why they will be allowed a second chance.

Using Problem Based Learning for Differentiated Instruction

Experience has taught me that differentiated instruction is one of the most difficult things to successfully accomplish,  particularly in a high school science class with 30 students.  How does a teacher divide his or her time between 30 students in a meaningful way? How can we recognize and adapt lessons to the various personalities and abilities of all those students?  For a long time I seriously questioned whether or not it could even be done. Using PBL has given me a glimmer of hope that DI is possible.

To apply DI in PBL requires careful planning and preparation. One of the most difficult things about designing a successful PBL lesson is to scaffold the material enough so that students are equipped to begin the problem but do so in a way that does not give them a direct route to the answer.  What I've found is that once this is done students can approach the problem from numerous different angles and with a variety of different resources.  This is the first link in the DI chain; students can approach a problem in ways that they are comfortable.  The second link involves the teacher;s role in PBL. Instead of being a passive observer, teachers need to constantly evaluate where students are and how they are progressing. 

Using cell phones, laptop computers or tablets students in my class are expected to find definitions, formulas, constants and the 'information'. Once they have done this they are tasked with figuring out how to use it.  Many of the students find this a daunting challenge and I don't expect them to find all the answers and applications themselves.  If they could wouldn't that make me quite redundant?

What do I expect is that students make a solid attempt at applying their new found facts.  This is where PBL really shines when it comes to DI.  Some students, or groups of students, are able to accomplish this task without much guidance.  These students can either go on to the next problem or look at more complex aspects of the problem while I am free to work with individuals to try and develop the concept in a more complete, comprehensive fashion.

I don't think I know the best way to organize my PBL lessons yet.  In some ways I feel like I'm groping in the dark for some I can't only vaguely make out.  But what I've seen has convinced me that there is merit in techniques like PBL and it just requires a bit of work and practice to make it successful.