For the Teachers Feeling Low Because of VAM
By: Mercedes Schneider
I would like to take this opportunity to write a post for my teaching colleagues who are feeling bad about themselves because of value added modeling (VAM). I too am a public school teacher, and I too have a standardized testing “passing” quota that I have been told I must reach in order to be considered “effective.” One advantage I have is that I hold a Ph.D. in statistics, so I know the lie that VAM is, and I do not allow any VAM score to determine my worth as a teacher.
I want to encourage my colleagues to take heart. I want to show you what a mess VAM is, in hopes that you, too, will hold your heads high and say, “I know what I do for my students, day in and day out, and I know there is no way VAM can capture my professional value no matter what the numbers might say.”
When I looked at the data for the accuracy of VAM reclassification from one year to the next given that the teachers do not alter their teaching, I realized VAM was useless. (See my piece on the pilot study for VAM for more detailed information: (click on this link to Schneider's blog). If VAM were able to classify teachers accurately, then for all categories of teacher “effectiveness,” reclassification rates would have been highly consistent from year to year. Instead, reclassification rates for teachers, given no change in how the teachers teach, were erratic, with some reclassifications as low as 15% and only as high as 64%. Let’s turn these numbers around: If we assume that VAM only correctly classified 15% – 64% of teachers to begin with, then it misclassified from 36% -85% to begin with (100% - 15% = 85%; 100% - 64% = 36%). Therefore, VAM presents a huge problem from the very beginning of the classification process. It cannot get classification anywhere near accurate from the start.
If a teacher were classified as “ineffective,” it is highly likely not because of anything he/she did or did not do in his/her classroom. It is a VAM problem. Conversely, if a teacher were classified as “highly effective” one year and did not receive the same classification the next year, again, it is highly likely that this “loss” of teacher “effectiveness” had absolutely nothing to do with the teacher’s performance in the classroom.
This is why the teachers in Houston call the receiving a “highly effective” rating “the lottery.” (click on this link to an earlier Louisiana Educator post) They know from experience that receiving a high VAM rating is hit and miss.
VAM accuracy is like a bathroom scale with broken springs. If I stand on a scale, and the scale reads, 156, then I step off, step back on, and the scale now reads, 142, which is correct? I have no reason to believe that either number is correct, actually. What is “correct” is to throw away the scale.
And what of the arbitrary predetermination for the state to classify 10% of teachers as “ineffective”? The court already declared such quotas as unconstitutional (http://theadvocate.com/home/4706171-125/judge-upholds-most-of-teacher). My colleague, Herb Bassett, in his five-minute video on VAM problems, notes the arbitrariness of such quotas. https://docs.google.com/file/d/0B0Pq4gPXaWVySjRZTXN3a200Vkk/edit?usp=sharing).
Finally, let me encourage my teaching colleagues by noting that Act 1 (Teacher Evaluation) will be back in the courts this week (http://deutsch29.wordpress.com/2013/01/25/the-constitutionality-of-la-teacher-eval-legislation-rests-in-the-title/). The reason: The judge based his initial decision on what he thought was the title of the legislation—but was actually a subtitle.
Hold your heads up, my colleagues. You are valuable no matter what the VAM nonsense purports.