Some of you may have seen the very distressing news yesterday on the investigation into cheating at several Denver Public Schools including the highly acclaimed Beach Court Elementary (EdNews article here).
Over the last five years I, like many, have been a huge supporter and cheerleader for Beach Court and given what I had understood was happening from my visits and an analysis of publically available data over the years they seemed to be doing an amazing job. Beach Court saw a dramatic change in CSAP proficiency scores from the low 40's to the upper 60's in 2007 with numbers growing into the upper 80's, amongst the highest in Denver, and certainly the highest for any elementary school serving so many low-income students (growth was also in the 80 and 90 percentage points for many of those years).
Yesterday's news concerning an investigation by the Colorado Attorney General, Colorado Department of Education and Denver Public Schools raises all sorts of worries though it is important to remind ourselves that this is an on-going investigation and that people (and schools) are innocent until proven guilty.
It's also important to recall that when stakes are high, a small number will try to game or cheat the system. This is true in professional sports, on Wall Street or in almost any sector including education. Most people will do the right thing but some for greed, fame or other unique circumstances will cut corners. This is why strong oversight systems are needed whether on Wall Street or in public education. A good oversight system for educational testing is as much about protecting and validating great work as it is about identifying cheating.
Regardless of the outcome of this investigation of Beach Court and Hallet, the good news is that we have other schools in Denver and around the country that continue to demonstrate that through effective practices and proper support low-income kids are excelling. This investigation on testing improprieties seems to have only picked up a few schools among the growing number of schools in Denver that are showing positive change. Recent EdNews, Atlanta Constitution and USA Today (another one here) stories seem to indicate that Colorado had fewer statistically anomalous scores than some other states and districts.
While I want to praise the DPS administration for taking this so seriously, it's unfortunate that this was not discovered earlier given the number of years this may have been occuring. I thought I would outline a few policies and practices that could help ensure that this does not happen again and/or is identified earlier by the state and district.
- Ensure that the Colorado Department of Education (CDE) has the resources and capacity to investigate any school or district that shows some anomalous statistical improvement or is given a tip that something is suspicious at a school or district. Currently the state is very limited in it's capacity to investigate these situations. The same should be true for large districts like DPS.
- DPS should do more spot monitoring of schools and increase oversight whenever a school shows a dramatic increase in test scores. Schools and the district should be able to ask for monitoring once a school has made a dramatic change to guarantee to the public that students and the public have not been bamboozled.
- Teachers administering assessments should not be allowed to alter anything on student test booklets without other teachers and administrators that are designated assessment monitors present. Allowing teachers to "clean up" student responses without supervision could lead to all sorts of problems given the pressure for teachers to have their students perform well.
While it can take time and some resources, having a system of checks and balances in school districts and at the state level for state testing are critical to having a system of assessment and accountability that will build public trust and provide policy makers with guidance for what works.