Student Test Scores as Evaluation of Teacher Performance

There was an interesting item posted in the New York Times a few weeks ago regarding the use of student mandatory test scores as the source for evaluation of teacher performance.  The original article can be found here:

http://www.nytimes.com/2011/11/28/education/principals-protest-increased-use-of-test-scores-to-evaluate-educators.html

My first surprise when reading this article is that not all principals were happy with linking test scores to teacher effectiveness.  That was actually wonderful to hear.  Here in Southern California, there are so many schools within solid districts that pull the district's state standings down.  The problem is that these schools are generally in the poorest (and predominantly Latin) neighborhoods, where parents, although holding education important, hold family survival higher.  In such districts, many students come to school exhausted or hungry, mainly because they are in charge of younger siblings and cousins while parents are working several jobs each to keep even scraps of food on the table.  A tired and hungry student--no matter how old or how young--will not be as responsive in class as a well-fed and well-rested student from a household with better financial resources and only 1--maybe 2--jobs per adult.  No matter how good a teacher is and no matter how much he/she cares about the students' overall welfare as well as academic growth, the growth will be slow.

When I wrote my dissertation looking for specific characteristics of Hispanic dropouts from across the country (I used the NELS:88 database from the National Center for Education Statistics--now part of the government's Institute of Education Sciences), I missed an important piece of information that I want to explore more fully in the future.  First, among Hispanic students who had ever dropped out, whether they return to school before the age of 21 is inexorably linked to level of poverty before beginning high school.  Since I was looking for predictive characteristics in 8th grade (the first available in this database) as a regression on final dropout/completion status by traditional college graduation (8 years later), I found that the deeper the family poverty, the less likely the dropouts would even attempt to return to school.  In fact, it was interesting to note that NCES never considered that a lot of their targeted students would have already withdrawn from the school establishment before the spring of 1988 when the data was originally collected.  A more critical finding was the incredibly high correlation between reading and math skills (.97) AND the reading/math correlation to poverty indexes (.95).  My first reaction was that the co-variance of math and reading was due to the fact that much of math includes written instructions and word problems, so that reading would be the driving factor, and that reading materials are very hard to come by when staying alive is a family's major goal. 

Over the years, in speaking to educators who work with students in such circumstances, I have come to believe that there was much truth to all the ado during the late 1960s and 1970s about free lunch programs.  But even back then, a student's family responsibilities seemed to be subsumed by just how early in life many of these kids have to contribute to whatever stability a family unit can attain. Students who are actively involved in keeping things together at home while Mom and Dad are out working at incredibly poor-paying jobs for 16 hours a day are not going to do well in school, no matter the quality of academic resources.  We've all heard reports on the number of middle and high school students who come to school for free breakfast and lunch, and disappear before the afternoon classes start. Many of these students' basic skills are so low that there is no longer any understanding of class materials.  According to NELS:88, most Hispanic dropouts during that era had reading levels at or below 2nd grade, with no indication of learning disabilities (at the time, students with any identified exceptionality were excluded from the surveys), or ESL/ELL needs. 

There are depths upon depths upon depths of information that educators--terachers, administrators,researchers--are unable to fathom. Yet these have little or nothing to do with teacher quality or how effective the teacher would be in a class of less impoverished students.  It doesn't even seem to to be related to the language spoken in the home.  My suspicion is that, as a society, we are not providing for the basic human needs of our poorest citizens--food, shelter, clothing, medical care.  I cannot believe that all the nation's poor want to be poor.  I cannot belive that most very poor parents are happy about keeping their eldest children home from school to care for the youngest.  I cannot believe that most poor parents like the idea of being home only long enough to eat a piece of bread and catch a few hours of sleep before returning to one of their several below-subsistence paying jobs, while leaving the oldest child (even if only 7 or 8 years old) in charge of younger ones, in essence guarateening that the child will get no homework assignments done, and probably will not get enough sleep. 

To blame a teacher when a class of students with similar problems performs poorly on tests is unconscionable.  Most of these teachers are among the most dedicated educators I have ever met.  They work so hard to help each individual in their classes learn as much as possible.  Yet, they are threatened with being fired if her/his classes continue to score below some predetermined level.  

I guess what I am trying to say is that this is not an issue that should be blamed solely on educators and educational institutions.  This is an issue that needs to be more thoroughly explored at its social roots. 

Your thoughts?

##