Running throughout the L.A. Times articles is a curious slippage between two concepts.  The main point is clear:  some teachers are more effective than others at helping their students to score well on standardized tests in Math and English.  At the same time, the reporters acknowledge that there is more to effective teaching than producing high test scores in English and Math.  But again and again this distinction is forgotten, and the reporters continuously equate “effective teaching” with improved student test scores.   

There are a number of problems with this equation, and with the method that the reporters used—the so-called “value added” approach—to arrive at their conclusions.  First, the statistical validity of using this methodology has been seriously questioned by a variety of statisticians who have examined it. They have found that it simply is not ready for prime time. For example, a U.S. Department of Education institute warned against “likely system errors” if “value added” was applied to high stakes decision-making. Results from value-added studies have been shown to vary significantly from year to year for the same teacher. The Times articles, in fact, fail to point out the lack of reliability of the results or the statistical significance of different scores.  Teachers who were evaluated as causing less growth were at other times found to be quite adequate. Likewise for highly-rated teachers. Further, "value added” says nothing about why actual student growth versus predicted growth varies from class to class, only that it does.  

Statistical analysis of this type can be useful in providing preliminary, quantitative indicators of pedagogical problems and successes.  However, statistical analysis alone cannot reveal the specific changes to be made, which requires both direct observation and expertise in pedagogy and professional development.  

Thus value-added is a poor foundation on which to build what proposes to be an assessment of “effective teaching.”  Teachers and teacher unions generally have no problem with the idea of factoring student test score progress into a comprehensive tool to help improve instruction.  It is another matter when this single flawed measure is elevated to be the most important measure for the public to judge teacher competence - and it is being so elevated by the LA Times articles.  Why else devote so much space in a shrinking newspaper to the story?  And to publishing the data?  

Attaching teachers’ names publicly to the scores in the context of this interpretation of data—and without what is arguably more important contextual information—is an invasion of the teachers’ privacy while being unfairly destructive of their reputations.  What Song and Felch’s articles did not provide is the bigger picture into which we might fit this data and thus give it a sense of relative importance in answering the real question, which is, “How do we improve student learning and teacher effectiveness?”  If this is what the L.A. Times was after, rather than continuing its erosion of journalistic firewalls between editorial and news departments seemingly in pursuit of a teacher demonization agenda, it would have noted a few other things:  

• California is $2,500 per year behind the U.S. average of per pupil funding as a result of enormous budget cuts.     
• Teacher evaluation is deeply affected by the chronic underfunding of our system, including lack of effective training of administrators to do evaluations, too few administrators to perform that task, and the gutting of programs such as Peer Assistance and Review (PAR), which was designed to help beginning and struggling teachers improve.   
• Nearly fifty percent of teachers leave the occupation before their fifth year, due to the enormous difficulties of serving in a severely underfunded system.  We have a far bigger problem in retaining teachers than getting rid of them.  Why not focus attention on addressing the most important factors that cause good teachers to leave the classroom?   

The Times is doing a disservice to individual teachers, to public education in hard times, and to the public, by presenting a misleading picture of what is important to consider in improving the education of our students.  Student standardized test scores are one part of the puzzle. But standardized tests do not measure student excitement in learning, ability to complete tasks, and the ability to think critically - or even look at the citizen that is being produced.  In short, there are many other parts to the puzzle of how we assess teacher effectiveness.  These articles use clumsy and inadequate statistical methods and while asking the wrong questions, predictably come up with the wrong answers.